Ac Manual10 Vol1

AMERICAN
CINEMATOGRAPHER
MANUAL
TENTH EDITION
VOLUME I
AMERICAN
CINEMATOGRAPHER
MANUAL
TENTH
EDITION
Volume I
EDITED BY
Michael Goi, ASC
The ASC Press

Hollywood, California
American Cinematographer Manual
Tenth Edition—Volume I
Copyright© 2013 by The ASC Press
Hollywood, California, USA

All Rights Reserved
ISBN 978-1-4675-6831-9
Cover design by Martha Winterhalter

Book design by Deeann j. Hoff
Production by Mark McDougal
| iii
Foreword
Y ou hold in your hands the result of five years of thought, debate, and
inspiration. When Stephen Burum, ASC, asked me to be the editor of
this 10th edition of the venerable American Cinematographer Manual, the
industry was in the birth throes of transition; digital intermediates were the
exception and not the rule, we still used the term video rather than digital,
and 4K as a viable production and post format was far beyond our reach.
All these changes and many more came in rapid succession as we labored to
bring this book to press. No sooner had we completed an article when it had
to be updated due to sweeping advances in technology.
I am at heart a low-tech person. I like things simple. I questioned whether
I was even the right person to be taking this book on. But in a strange way, it
made sense. If I could design the manual in a manner that made sense to me,
then the information it contained would be accesible to a wide spectrum of
professional and prosumer image makers. Cinematographers today need to
be closet scientists in order to decipher the tools they have at their disposal,
but all those technologies need not be daunting; they can be fun to explore
and exciting to utilize. Now more than ever, the dreams of a whole new gen-
eration can be made into real moving images. This edition contains some of
the most comprehensive information on digital that you will find anywhere,
but it doesn’t leave behind the essential building blocks of film technology,
which is at its highest level of development. Where we are now is really having
the best of both worlds.
When you embark on a journey to a new world, it’s best to take along a
crew who know the territory. The contributors to this edition have proven to
be the most helpful and dedicated group of scientists, artists and craftspeople
one could possibly hope to assemble. Thanks go to Jim Branch, Curtis Clark,
ASC; Richard Crudo, ASC; Dan Curry; Linwood G. Dunn, ASC; Richard
Edlund, ASC; Jonathan Erland; Jon Fauer, ASC; Ray Feeney; Tom Fraser;
Taz Goldstein; Colin Green and the Previsualization Society; Frieder Hoch-
heim; Michael Hofstein; Bill Hogan; John Hora, ASC; Rob Hummel; Steve
Irwin; Kent H. Jorgensen; Frank Kay; Glenn Kennel; Jon Kranhouse; Lou
Levinson; Andy Maltz and the AMPAS Science and Technology Council;
Vincent Matta; Tak Miyagishima; David Morin; M. David Mullen, ASC;
Dennis Muren, ASC; Iain A. Neil; Marty Ollstein; Josh Pines; Steven Poster,
ASC; Sarah Priestnall; David Reisner; Pete Romano, ASC; Andy Romanoff;
Dr. Rod Ryan; Nic Sadler and Chemical Wedding; Bill Taylor, ASC; Ira Tiffen
and Evans Wetmore.
iv
|
Special thanks go to Iain Stasukevich for his assistance in research, Lowell
Peterson, ASC, Jamie Anderson, ASC and King Greenspon for their proof-
reading skills, and Deeann Hoff and Mark McDougal for handling the layout
of the book.
Extra special thanks go to Brett Grauman, general manager of the ASC,
Patty Armacost, events coordinator, Delphine Figueras, my assistant when
I was handling being ASC president while trying to finish this book, Saul
Molina and Alex Lopex for their expertise in marketing and events man-
agement, Owen Roizman, ASC, who is the heart, soul and inspiration for
the organization, George Spiro Dibie, ASC, my mentor and friend, Martha
Winterhalter, whose knowledge of what we do and how to convey it to the
world knows no bounds, and Gina Goi, my wife, for her love and support
during my many twilight editing sessions.
Enjoy the Manual. Go make movies.
Michael Goi, ASC

Editor
| v
Table of Contents
Foreword Vol. I iii
Origins of the American Society of Cinematographers Vol. I 1
Responsibilities of the Cinematographer Vol. I 3
Summary of Formats Vol. I 7
Tak Miyagishima
Basic Digtal Concepts Vol. I 9
Marty Ollstein
A Primer for Evaluating Digital Motion Picture Cameras Vol. I 31
Rob Hummel
Digtal Cinematography on a Budget Vol. I 37
M. David Mullen, ASC
Putting the Image on Film Vol. I 57
Rob Hummel
Comparisons of 35mm 1.85, Anamorphic and Super 35
Film Formats Vol. I 59
Rob Hummel
Anamorphic Cinematography Vol. I 87
John Hora, ASC
Exposure Meters Vol. I 93
Jim Branch
Lenses Vol. I 111
Iain A. Neil
Camera Filters Vol. I 143
Ira Tiffen
Camera-Stabilizing Systems Vol. I 173
Previsualization Vol. I 187
Colin Green
3-D Stereoscopic Cinematography Vol. I 193
Rob Hummel
Day-for-Night, Infrared and Ultraviolet Cinematography Vol. I 207
Dr. Rod Ryan
Aerial Cinematography Vol. I 215
Jon Kranhouse
Underwater Cinematography Vol. I 229
Pete Romano, ASC
vi
|
Table of Contents
Arctic and Tropical Cinematography Vol. I 239

Filming Television and Computer Displays Vol. I 251
Bill Hogan and Steve Irwin
Digital Postproduction for Feature Films Vol. I 257

Glenn Kennel and Sarah Priestnall
ASC Color Decision List Vol. I 273

David Reisner and Josh Pines
The Academy of Motion Picture Arts and Sciences

Academy Color Encoding System Vol. I 283
Curtis Clark, ASC and Andy Maltz
The Cinematographer and the Laboratory Vol. I 289

Rob Hummel
Emulsion Testing Vol. I 299

Steven Poster, ASC
Finding Your Own Printer Light Vol. I 305

Richard Crudo, ASC
Adjusting Printer Lights to Match Sample Clips Vol. I 311

Bill Taylor, ASC
Cinemagic of the Optical Printer Vol. I 313

Linwood G. Dunn, ASC
Motion-Control Cinematography Vol. I 319

Richard Edlund, ASC
Greenscreen and Bluescreen Photography Vol. I 325

Bill Taylor, ASC and Petro Vlahos
Photographing Miniatures Vol. I 363

Dennis Muren, ASC
In-Camera Compositing of Miniatures with Full-Scale

Live-Action Actors Vol. I 367
Dan Curry
Light Sources, Luminaires and Lighting Filters Vol. I 375

LED Lighting For Motion Picture Production Vol. I 395
Frieder Hochheim
An Introduction to Digital Terminology Vol. I 401

Marty Ollstein and Levie Isaacks, ASC
| vii
Table of Contents
Safety On The Set Vol. II 423

Kent H. Jorgensen and Vincent Matta
Preparation of Motion Picture Film Camera Equipment Vol. II 427
Marty Ollstein, Michael Hofstein & Tom Fraser
Preparation of Digital Camera Equipment Vol. II 447
Marty Ollstein
Camera-Support Systems Vol. II 467
Andy Romanoff, Frank Kay and Kent H. Jorgensen
Camera Section Vol. II 501
Jon Fauer, ASC and M. David Mullen, ASC
35mm Vol. II 502
16mm Vol. II 559
Super 8mm Vol. II 578
65mm Vol. II 581
VistaVision 35mm Vol. II 591
Imax Vol. II 595
Digital Cameras Vol. II 602
Onboard recorders Vol. II 638
Table Reference Vol. II 649
Cine Lens List Vol. II 653
Formulas Vol. II 673
Evans Wetmore
Lenses Vol. II 699
Extreme Close-Up Vol. II 810
Filters Vol. II 819
Mired Shift Values Vol. II 835
Light Source Vol. II 836
Color Balancing Vol. II 839
Handheld Apps for Production Vol. II 843
Taz Goldstein
SunPATH Vol. II 855
Film Stocks Vol. II 864
Incident Light Vol. II 866
EI Reduction Vol. II 869
T-Stop Compensation Vol. II 870
viii
|
Table of Contents
Shutter Compensation Vol. II 872

Shutter Speeds Vol. II 876
Footage Tables Vol. II 878
16mm/35mm–Frame Totalizers Vol. II 889
Panning Speeds Vol. II 891
Time/Speed Effects Vol. II 894
Projection/Process Vol. II 897
Quick Picture Monitor Set-Up Vol. II 899
Lou Levinson
Lighting Fixture Intensity Vol. II 901
Further Reference Vol. II 921
Index Vol. II 931
Creativity,
your mind our
tools!
Let us help
you choose!
Providing you with the industry’s widest selection of

film and digital cameras, lenses and accessories, backed with
superior service and legendary technical support.
www.clairmont.com
what you get
is what you see
Sony TRIMASTER™ EL monitors.

If you believe black should actually be black. If you believe that moving pictures really
don’t benefit when the monitor adds motion blur. If you believe the only good color is the
accurate color. And if you believe high dynamic range is not only for cameras, Sony
has monitors you can believe in. Our award-winning TRIMASTER EL monitors change
everything thanks to Sony Super Top Emission OLED technology. Choose the BVM-E Series
for digital cinema evaluation, BVM-F Series for broadcast evaluation or the surprisingly
affordable PVM Series in 7.4, 17 and 25 inches* for picture monitoring. Seven models with
one standard of picture quality: the truth.
Visit sony.com/TrimasterEL to arrange an on-site demo.
© 2012 Sony Electronics Inc. All rights reserved. Reproduction in whole or in part without written permission is prohibited. Features and specifications are subject to change
without notice. Weights and measurements are approximate. Sony, TRIMASTER EL and the Sony make. believe logo are trademarks of Sony. Screen image simulated.
* Viewable area, measured diagonally.
ASC_DV&EDIT_2CMYK.indd 1 12/19/12 8:23 AM
Samy’s Camera
for all your digital cinema needs
Rent It or Buy It
from grip
to lights
to cameras
Rental Depar t ment

Professional Cameras, Lenses, Studio Equipment & more...
Los Angeles 323-938-4400 I Playa Vista 310-450-7062 I Pasadena 626-796-3300
Santa Barbara 805-963-7269 I Santa Ana 714-557-9400 I San Francisco
S aleS
Los Angeles 323.938.2420 I Culver City 310.450.4551 Pasadena 626.796.3300
Santa Barbara 805.963.7269 I Santa Ana 714.557.9400 I San Francisco
www•samys•com
ASC_Samys2.indd 1 12/19/12 8:14 AM
CREATIVE VISION
ANYWHERE. ANYTIME.
EC3 provides location services through the combined creativity, experience and technological
infrastructure that is EFILM® and Company 3®; two of the industry’s most trusted post
production houses. Through custom configured hardware, software solutions and highly
skilled field technicians, EC3 provides location services for all digital capture Features.
On Location Dailies Deliverables including Complete 3D capabilities including

iPad® dailies and editorial media same day convergence On Location
Color Correction On Location Data Management and Security
Instant Quality Control of your footage Complete Engineering and Tech
Support 24/7
2k /HD / 3D projector On Location
323.308.3094 / 310.255.6643 / E C 3 O N LO C AT I O N . C O M
EC3, EFILM and Company 3 are Deluxe Companies

iPad is a registered trademark of Apple Inc.
S:5.5”
STAGE
12
STAGE
3 STAGE
26
STAGE
34
S:8.5”
I N T R O D U C I N G T H E C A N O N C I N E M A E O S S Y S T E M
GO WHEREVER THE STORY TAKES YOU.

Presenting a line of cameras designed to shoot anything at every level of production. From the C100 and
C300, with their incredible low light performance, to the high resolution 4K image quality of the C500 and 1D-C.
Cinema EOS delivers everything including a range of resolutions and recording options for optimized image
capture, a Super 35mm CMOS sensor and compatibility with our full line of EF lenses, and new PL-mount
and EF-mount Cinema lenses. The complete Canon Cinema EOS System. Now, the world truly is your stage.
GET STARTED. CONTACT US: 855.CINE.EOS - CINEMAEOS.USA.CANON.COM
© 2012 Canon U.S.A., Inc. All rights reserved. Canon and EOS are registered trademarks of Canon Inc. in the United States and may also be registered trademarks or trademarks in other countries.
BD & DVD
Color On Location Content Content
Services
Science Services Preparation Management
Animation
Front End Sound Digital
Digital Cinema
Distribution
Mastering
VFX Digital Cinema

Post Content Packaging
Distribution
& Versioning
technicolor.com
e
S
He S
il
Fo
n To gR T H
R’
k
ac
R
Bl
iO O il T
Fr
os
e
lf TO
t
ll C un P
O
Ha l C
CT
5 ul
Fu Sp A
20 04 F
2
0.
6
TB
nD
Cl grid
iT AT F
ot
2 h h
20 ug
eD m e
ne le
20
Sc 1 H
rim alf
Re CTB
F fle
21 ull C ct
6 T
Di O
or
ffu A
sio cry
n lic
Pa
Ci
ne
l
twitter.com/leefilters
facebook.com/leefiltersusa Think LEE
leefilters.tumblr.com www.leefilters.com
3780_LF_CineAd_AW_5.5x8.5_V2.indd 1 16/11/2012 10:58

For more than 70 years, the American Cinematographer Manual has
been the key technical resource for cinematographers around the world.
Chemical Wedding is proud to be partners with the American Society of
Cinematographers in bringing you the Toland ASC Digital Assistant for
the iPhone or iPod Touch.
Toland is an application that provides solutions to the technical

concerns of cinematography in a unique multi-tasking interface.
Combining a calculator and logging interface, it is designed to work with
you throughout the process of cinematography.
Toland will track your photographic choices as you make them. As you
change the camera speed, you will instantly get feedback on how this
affects running time and exposure; when you change lenses, you will
see Depth of Field and Field of View updates in real time.
“ “
GREGG TOLAND, ASC
STEPHEN LIGHTHILL MICHAEL GOI
KEY FEATURES This app, named in honor of The combined intellectual and
the legendary Citizen Kane practical knowledge of the
cinematographer, Gregg world's greatest
Comprehensive database Toland, is an indispensable, cinematographers has been
on-set, exposure management poured into this, the first
of cameras and lenses tool. And, in keeping with the app-based field tool worthy of
Exposure calculator educational role of the ASC, carrying the ASC brand. The
the app is also a teacher and research and care that went
covering camera speed, mentor in its own right: as you into this was driven by our
”
learn the app, you learn the desire to make something
shutter angle & filter factor black art of exposure. innovative in design and
function that would be a fitting
”
Running time and footage
tribute to the cinematographer
calculator whose name it bears.
‘Flicker Free’ warning

indicator
Depth of Field calculator
with clever focus marks
Angle of View indicator
with pretty picture
Full camera data logging
exportable for offline
reporting
POCKET ALCHEMY CAN BE FOUND AT:

the rt of film optics
optimo lenses transform light into
the imagery of your story. Employing the
finest optics and mechanics to capture
images beyond ever y expectation.
Available in 17-80mm, 24-290mm and
now 19.5-94mm and 28-340mm.
angenieux@tccus.com • www.angenieux.com
DIGITAL CINEMA (RE)SOLUTION
7680 x 4320
8K
ADORAMA RENTAL CO
5120 x 2700 (RED EPIC)
4096 x 2160
5K WE HAVE A SOLUTION
4K FOR YOUR RESOLUTION
3840 x 2160 Quad HD
3072 x 1728 3.8K ARC rents and supports the full
2880 x 1620 (ARRI) 3K range of professional still & motion
2048 x 1152
2.8K equipment, including digital cinema
cameras, accessories, lenses,
1920 x 1080 lighting and grip.
1080P 2K
1280 x 720
720P
NEW GEAR. KNOWLEDGEABLE STAFF. GREAT PRICES. | RENT | MOTION@ADORAMA.COM | 212 - 627- 8487 | NYC | ADORAMARENTAL.COM
•
Truly Cinematic
www.arri.com
1212_ACManual2012_F.indd 1 12/4/12 3:00 PM
Profile. Calibrate. Measure.
Sekonic.com
Designed, Developed, Engineered LED’S
& Assembled in the USA ACCEPTED BY
BROADCASTERS,
NEWS AND
SPORTS GROUPS
JAB HURRICANE
LED-IP65 RATED, “WEATHERPROOF”
Single Source & No multiple shadows,
DMX compatible, AC/DC & output compatable
to a 800 Watt HMI. Optional units Jab Daylight,
Jab Variable & Jab Tungsten.
ECO PUNCH PLUS

4,044 foot candles @ 10 ft., Draws 4.8 Amps,
Daylight 5600 K, Single Source & No multiple
shadows, DMX Compatible & more
output than a 2,500 Watt HMI.
FOR ALL UNITS ETL, UL, CE & CSA approved, RoHS Compliant, No Ultra Violet or Infra Red Rays,
Cree LEDs, 60,000 hour warranty on LEDs, Green Technology, Universal power supply and much more.
JAB HURRICANE SPACE LED LIGHT

IP65 Rated-Weatherproof Draws 6.5 Amps @ 120v AC
Total Protection against Dust Up to 90% power savings
and Low Pressure Water Jets
Tungsten – 3280° Kelvin
1,804 foot candles @10 feet
Over 52 foot candles @ 25 feet height
Daylight-Balanced at 5600°K
Even spread of light
Draws 1.77 amps at max output
High CRI – 95
AC/DC Powered
Worldwide Distribution
Germany & 20 Euro Countries Switzerland Canada
U.K. & Ireland Australia & New Zealand Mexico, Brazil & Peru Italy
EvErything you Want in
a CamEra rEntal housE
ARRI SONY RED AATON
MOVIECAM JVC PANASONIC
ANGENIEUX COOKE FUJINON
CANON NIKON ZEISS
and so much more
Oppenheimer Cine Rental has long been the #1 20
Year
Film & Digital

Ser v s of
choice for camera rentals in the Pacific Northwest. ice
As the largest camera rental house north of LA, we

work long and hard to maintain the best & most
complete inventory, keep our technical skills at
SPECIAlTY PRODUCTS
the top of their game, and provide the friendliest Oppenheimer Camera Products’ offers OppCam
and most professional service, all at nationally Panhandle System, many LCD and OLED Monitor Yoke
Mounts, 60mm & 100mm Macro Lenses, ARRI ALEXA &
competitive rates.
Sony F65 OB Power Supplies, On-Camera Viewfinder
Great prices and Systems, Lens Carry Handles for Angenieux, Alura,
Canon and Fujinon zoom lenses.
great service since 1979!
Purveyors of Fine Cine Cameras
seattle 206-467-8666 toll free 877-467-8666 marty @oppcam.com www.oppenheimercinerental.com

CINEMATOGRAPHERS PAINT THE MOOD
FOR MORE THAN 50 YEARS, PANAVISION HAS PROVIDED THE BRUSH
artistically
inspired
technically
advanced
We Give New meaning to the Phrase "Best and Brightest"
Environmentally
Sustainable
LED Lighting
Nila, Inc. 723 West Woodbury Rd., Altadena, CA 91001, 8183928370 Nila.com
Film & Digital Cinematography Equipment - S3D - Wireless
www.hd4dp.com
CineMonitorHD, Creatively Connected

Connects /i lenses and CineTape to display essential informations
To Michael Lewis, Greg Marsden, Raigo Alas
Th
The ulTimaTe choices
and Michael Vellekoop for the concept,
©A.M.P.A.S.®
design and implementation of the Pictorvision
Eclipse, an electronically stabilized aerial
camera platform. for above and beyond
fo
The Pictorvision Eclipse system allows
cinematographers to capture aerial footage
at faster flying speeds with aggressive
cinema
cin Tography
platform maneuvering.
Find these and our latest inventory of cutting edge technologies at:
W W W. P I C T O R V I S I O N . C O M 8 0 0 . 8 7 6 . 5 5 8 3 8 1 8 . 7 8 5 . 9 2 8 2
Eclipse XR Series Wescam System Cineflex

Completely updated engine Six tools to explore sun data
Save images and notes to new internal image gallery Database of tens of thousands of locations around the world
Replicates popular motion picture cameras and lenses Email sun position data from within application
Add custom cameras for specific or exotic formats Network connectivity not necessary for use
Available for iPhone, Android phones and the iPad
I am working on a film called "Balls to the I'm usually scouting for locations months
Walls" I use this app on every blocking before the DP comes on board and knowing
rehearsal. I find the master and all the precisely where the sun will be on the exact
coverage and capture the frames and lens shoot date is essential. Helios is without doubt
sizes in my photo library. Then I show the the best tool out there for DPs, production
director. I can also start setting up the shot designers, location scouts and directors.
before the camera is in place. Great app and Nothing comes even close.
now an important tool for my work! RICHARD LONCRAINE
ROBERT REED ALTMAN
A lot easier and much more accurate than

holding up your two hands to frame a shot.
Simple. Elegant. Fast, Accurate... I love this!!!!
Much lighter than a heavy Director's
RUSSELL CARPENTER, ASC
Finder dangling from your neck. Every
Director and DP should have this wonder-
ful app.
FILM AND DIGITAL TIMES
POCKET ALCHEMY CAN BE FOUND AT:

“I love Schneider One-Stop Linear
Polarizers for interiors. They allow
me to handle the occasional cross-
light shine on complexions without
having to deal with lighting.”
Francis Kenny, ASC
“When shooting a multi-camera

series you are consistently cutting
It starts
back and forth between cameras.
I never worry about matching
because I always have Schneider
filters on each lens.”
Don A. Morgan, ASC
with the
glass...
“I’m a long-time fan of the
Classic Soft™. It is the best
wrinkle remover ever—and
it is light enough to use on
digital and film.”
Nancy Schreiber, ASC ...but it’s nothing

without the
cinematographers
“The Schneider ND
Attenuator is quite the
amazing tool, particularly
who use it.
for digital sensors.”
Bill Bennett, ASC
“Schneider came through for us

with some of the first sets of their
great new Platinum IRNDs, and
all our color matching problems
disappeared.”
Lowell Peterson, ASC
www.schneideroptics.com Phone: 818-766-3715 • 800-228-1254 Classic Filmmakers Choose Schneider

The finesT
anamorphic
glass
| 1
Origins of the
American Society
of Cinematographers
F or over 93 years, the ASC has remained true to its ideals: loyalty, prog-
ress, artistry. Reverence for the past and a commitment to the future
have made a potent and lasting combination in a world of shifting values
and uncertain motives.
The American Society of Cinematographers received its charter from the
State of California in January 1919 and is the oldest continuously operat-
ing motion picture society in the world. Its declared purpose still resonates
today: “to advance the art of cinematography through artistry and techno-
logical progress, to exchange ideas and to cement a closer relationship among
cinematographers.”
The origins of the ASC lie in two clubs founded by cinematographers in
1913. The Cinema Camera Club was started in New York City by three cam-
eramen from the Thomas A. Edison Studio: Phil Rosen, Frank Kugler and
Lewis W. Physioc. They decided to form a fraternity to establish professional
standards, encourage the manufacture of better equipment and seek recogni-
tion as creative artists. Meanwhile, the similarly conceived Static Club was
formed in Los Angeles. When Rosen came to the West Coast five years later,
he and Charles Rosher combined the clubs. The ASC now has more than 340
active and associate members.
The first ASC screen credit was given to charter member Joseph August
when he photographed a William S. Hart picture in 1919.
American Society of Cinematographers’ clubhouse.

2
|
Origins of the Society
The year after its charter, ASC began publishing American Cinematogra-
pher magazine, which ever since has served as the club’s foremost means of
advancing the art.
The ASC has been very active in recent years in expressing concern about
choices for Advanced Television (ATV), ranging from the choice of aspect
ratio to pushing for the abandonment of interlace displays. At the invitation
of the House and Senate in Washington, D.C., members of the ASC have
been asked to inform and advise legislators on these issues.
Currently our technology committee has created a standard test (StEM) for
digital cinema. They are advising the industry on standards in both produc-
tion and postproduction for digital capture, manipulation and presentation.
The ASC is not a labor union or guild, but is an educational, cultural and
professional organization. Membership is possible by invitation and is extend-
ed only to directors of photography with distinguished credits in the industry.
—George E. Turner
| 3
Responsibilities
Of The Cinematographer
DESIGNING THE LOOK
F ilmmaking is a uniquely collaborative form of artistic expression with

many people playing interlocking roles. Cinematographers require ar-
tistic sensibilities, exceptional organizational skills, the ability to master
complex technologies that are constantly evolving, and a special talent for
collaborating and communicating effectively.
The cinematographer’s initial and most important responsibility is telling
the story and the design of a “look” or visual style that faithfully reflects the
intentions the director, or the producer/show runner if the project happens
to be an episodic television series. It is mandatory for the cinematographer
to accomplish that primary goal within the limitations of the budget and
schedule. Other collaborators who are generally involved in this creative pro-
cess with the cinematographer and director include the production designer,
art director, and occasionally the visual effects supervisor and/or producer.
The cinematographer must also provide guidance in all technical aspects of
production. They offer advice regarding the choice of the most appropriate
film or digital video format for creating the “look” within the restraints of
the budget. There are many options today, ranging from anamorphic 35mm
film to digital video. The cinematographer is responsible for explaining those
options and the creative and financial trade-offs with precise clarity.
If the project is a feature film, the cinematographer can discuss the pros
and cons of traditional timing at an optical lab versus a digital intermediate
(D.I.) facility. If it is a film made for television, the cinematographer can
address upfront and long-term issues linked to finishing in high-definition
or standard definition formats. There are no textbook answers because tech-
nology is constantly evolving and every situation is different. It is essential
for the cinematographer to have a thorough grasp of all technical options,
including choice of film and digital cameras, lenses, cranes, dollies and other
platforms for moving cameras, new emulsions, and such special techniques
as aerial and underwater photography, blue and green screen and other vi-
sual effects, as well as the capabilities of equipment rental houses, labs and
postproduction facilities.
PREPRODUCTION PLANNING
The cinematographer is frequently invited to watch rehearsals and offer
suggestions for “blocking” scenes to provide artful coverage in the most
4
|
Responsibilities of the Cinematographer
efficient way. All aspects of production are planned at this stage, including
how many cameras are needed, how to move them and whether older or
newer prime or zoom lenses are best suited for each task. Cinematographers
also frequently provide input to directors while they are developing story-
boards and shot lists. They go on location scouts and make recommenda-
tions to the production designer and art director for dressing and painting
sets based on the vision for the “look” of the film and requirements for light-
ing and camera movement. The cinematographer also confers with the di-
rector and assistant director about the best times and places to stage exterior
scenes to take maximum advantage of available light. They must also plan
for such variables as the weather and tides if they are going to shoot on the
beach or at sea. The cinematographer consults with the production designer
regarding plans for dressing stages and the space they need for rigging lights
and moving cameras. This includes the use of wild walls, removable sections
of ceilings, placement, sizes and angles of windows, practical lights, and the
colors and textures of props and walls.
The cinematographer also organizes camera, electrical and grip crews,
whose talents and skills are the right match for the tasks at hand. They work
with the gaffer to plan placement and rigging lighting fixtures, including
deciding whether a dimmer control console is needed. The cinematographer
also confers with the key grip on issues related to camera movement, includ-
ing what gear is needed. He or she also consults with the production manager
regarding arrangements for rental of camera, lighting and grip equipment,
and such specialized tools as insert cars, on the days it will be needed.
During preproduction, the cinematographer must establish rapport with
the make-up, hair and costume designers, which frequently includes shoot-
ing tests with the actors. Using information gained by this testing, the cine-
matographer can then present a visual interpretation of the actor’s character,
helping to define and amplify the performance. They might also test new
camera films, lenses and specialized photochemical or digital intermediate
processes (DI) to determine the most efficient way to put the final touches on
the visual design. In addition, the cinematographer establishes communica-
tions with the timer at the film lab or digital postproduction facility, which
will provide dailies. They also make arrangements and check out facilities for
viewing dailies, meet stand-ins for actors, and establish rapport and an open
line of communications with the AD.
PRINCIPAL PHOTOGRAPHY
The cinematographer is responsible for executing the vision for the “look”
of the film, while helping to keep production on budget and on schedule.
On many feature films, the day begins with the cinematographer viewing
dailies during early morning hours at the lab to verify that there are no
| 5
technical problems and nuances in the “look” are working. They frequently
watch rehearsals of the first scene with the director, and suggest whether
modifications in lighting or coverage are needed. The cinematographer
stays in touch with the production manager and AD regarding any changes
in the anticipated schedule caused by unforeseen circumstances. In ad-
dition, the cinematographer approves the set up of lighting by the gaffer.
Many directors want lighting and camera coverage to be “flexible” enough
to give the actors the freedom to perform spontaneously.
Shots are often rehearsed with stand-ins. The cinematographer confers
with the director regarding adjustments in plans for lighting and coverage
and facilitates those changes with the grip, gaffer and lighting crews. They
also work with the standby painter for any last minute touch-ups needed
on sets, assist the AD in staging background action, and give the sound de-
partment the freedom to put their booms where they are needed to record
great audio of the dialogue. If the director desires, a final walk-through or
rehearsal is done with the actors. When cameras are rolling, the cinema-
tographer assures that there are no technical glitches. They also provide an
extra pair of eyes for the director on the set. The cinematographer might
suggest retaking a shot because something didn’t quite look or feel right,
while assuring the director they will find a way to make up the time.
If a DI finish is planned, the cinematographer might be recording request
that digital still images of the scenes be taken, which he or she later manipu-
lates with a personal computer to give the dailies timer and colorist a visual
reference for the “look.”
The cinematographer, director and other key collaborators watch dailies
together to judge how effectively the “look” is working. At the end of each
day, the cinematographer discusses the first set up for the next morning
with the AD and possibly the director. He or she also informs the script
supervisor if there are special camera or lighting notes, makes sure that the
camera, lighting and grip crews have all the information needed to rig their
gear, provides any special notes and instructions required by the film lab and
dailies timer, and works with the production manager regarding the need to
return or acquire equipment.
POSTPRODUCTION
The cinematographer’s role extends deep into postproduction with the goal
of assuring that the “look” that he or she rendered is what audiences see on
cinema and television screens. If possible, the cinematographer handles any
additional photography required by changes in the script. They are also called
on to supervise the seamless blending of visual effects shots with live-action
footage. The cinematographer is responsible for timing the film for continu-
ity and for adding nuances to the “look” in either a DI or optical lab envi-
6
|
ronment. They also approve the final answer print in collaboration with the
director and producer. In addition, the cinematographer verifies that what
they approved is reflected in the release print. The final task for the cinema-
tographer involves timing, and, if necessary, reformatting films for release in
DVD, HD and other television formats. All of these final steps are meant to
assure that audiences experience films on motion picture and display screens
the way they are intended to be seen by the creators of the images.
| 7
Summary of Formats
compiled by Tak Miyagishima
ASC Associate Member
APERTURE SPECIFICATIONS
w 35mm Camera – Spherical Lens
Academy Camera Aperture .866" X .630" 22mm X 16mm
w 35mm Theatrical Release – Spherical
1.37:1 .825" X .602" 20.96mm X 15.29mm
1.66:1 .825" X .497" 20.96mm X 12.62mm
1.85:1 .825" X .446" 20.96mm X 11.33mm
w 35mm Television Aperture and Safe Areas
Camera Aperture .866" X .630" 22mm X 16mm
TV Station Projector Aperture .816" X .612" 20.73mm X 15.54mm
TV Transmitted Area .792" X .594" 20.12mm X 15.09mm
TV Safe Action Area .713" X .535" 18.11mm X 13.59mm
Corner Radii = .143"/3.63mm
TV Safe Title Area .630" X .475" 16mm X 12.06mm
Corner Radii = .125"/3.17mm
w 35mm Full Aperture – Spherical Lens (For Partial Frame Extraction) Prints (Super 35)
Camera Aperture (Film Center) .980" X .735" 24.89mm X 18.67mm
Finder Markings
35mm Anamorphic 2.4:1 AR .945" X .394" 24mm X 10mm
70mm 2.2:1 AR .945" X .430" 24mm X 10.92mm
35mm FLAT 1.85:1 AR .945" X .511" 24mm x 12.97mm
w 35mm Panavision 2-Perf
Camera Aperture (Film Center) .980" X .365" 24.89mm x 9.27mm
Ground Glass 2.4:1 AR .825" X .345" 20.96mm x 8.76mm
Camera Aperture (Film Center) .980" X .546" 24.89mm x 13.87mm
1.78:1 .910" X .511" 23.10mm x 12.98mm
1.85:1 AR Spherical (FLAT) PROJ AP .825" X .446" 20.96mm X 11.33mm
2.4:1 AR Anamorphic Squeeze PROJ AP .825" X .690" 20.96mm X 17.53mm
5 perf 70mm 2.2:1 AR PROJ AP 1.912" X .870" 48.56mm X 22.10mm
w Panavision 35 and Anamorphic Squeezed Negative
Camera Aperture .866" X .732" 22mm X 18.59mm
35mm Squeezed Print
Finder Marking (2.2:1 70mm) & Proj AP .825" X .690" 20.96mm X 17.53mm
16mm Squeezed Print .342" X .286" 8.69mm X 7.26mm
Max Proj. AP
16mm Un-Squeezed Print (1.85:1) .380" X .205" 9.65mm X 5.20mm
Proj. AP matte
70mm Unsqueezed Print Proj. AP 1.912" X .870" 48.56mm X 22.10mm
8
|
Summary of Formats
w 16mm Film Apertures 1.33:1 (4:3) Television Safe Area

Camera Aperture .404" X .295" 10.26mm X 7.49mm
TV Station Proj AP .380" X .286" 9.65mm X 7.26mm
TV Transmitted Area .368" X .276" 9.35mm X 7.01mm
TV Safe Action Area .331" X .248" 8.41mm X 6.30mm
Corner Radii R = .066"/1.68mm
Safe Title Area .293" X .221" 7.44mm X 5.61mm
w Finder Markings for Enlarging to 35mm
Projector Aperture (1.37:1) .380" X .286" 9.65mm X 7.26mm
Projector Aperture (1.85:1) .380" X .206" 9.65mm X 5.23mm
(Enlarging ratio 1:2.105)
w Super 16mm (16mm Type W) for Enlarging to 35mm
Projector Aperture (1.66) .464" X .279" 11.80mm X 7.10mm
Projector Aperture (1.85) .464" X .251" 11.80mm X 6.38mm
w 65mm 5-Perf TODD-AO/PANAVISION 65mm Spherical Imaged Negative
Camera Aperture 2.072" X .906" 52.63mm X 23.01mm
35mm ’Scope Extraction 1.912" X .800" 48.56mm X 20.31mm
35mm Projector Aperture .825" X .690" 20.96mm X 17.53mm
(with 2:1 squeeze)
70mm Projection Aperture 2.2:1 1.912" X .870" 48.56mm X 22.10mm
65mm 8-Perf
Camera Aperture 1.35:1AR 2.072" X .1.485" 52.63mm X 37.72mm
w 65mm - 15-Perf IMAX/OMNIMAX

Camera Aperture 2.772" X 2.072" 70.41mm X 52.63mm
Projector Aperture (computed from cut-off) 1.172" X 2.04" 29.77mm X 51.81mm
16mm Un-Squeezed Print (1.85:1) .380" X .206" 9.65mm X 5.20mm
70mm Unsqueezed Print Proj. AP 1.912" X .870" 48.56mm X 22.10mm
w VISTAVISION 8-Perf Horz. Pull Across
Camera Aperture 1.485" X .981" 37.72mm X 24.92mm
35mm VistaVision
w Super 8mm
Camera Aperture (1.33:1) 0.224" x 0.166" 5.69mm X 4.22mm
Projection Aperture (1.33:1) 0.209" x 0.158" 5.31mm X 4.01mm
Most coMMon screen ratios
1.37 1.85 2.40

(1.33 - TV)
| 9
Basic Digital Concepts

by Marty Ollstein
A working understanding of digital theory is the foundation for success-

ful use of all digital technology, from cameras to post-production hard-
ware and software systems. Although the principles of photography still ap-
ply, the cinematographer today should understand the relationship of digital
to analog image acquisition.
1) AnAlog AnD DigitAl

The natural world is considered to be analog.1 Human vision perceives
continuously changing gradients of light from black to white, and color that
spans the spectrum uninterrupted between infrared and ultraviolet. We
model light and sound as waves whose shape varies by their amplitude and
frequency in a continuously variable signal. An analog signal is understood
to be composed of an infinite number of points. The scale used to measure
light and color is infinitely divisible into smaller units. Film is an analog
medium—it can record a continuous spectrum of color.
A digital representation of the world is finite. It uses numbers to approxi-
mate analog physical phenomena. A digital signal is defined by a finite value
in accordance with a predetermined scale. It may be hard to imagine using
numbers to define and reproduce the subtle and random sights and sounds
of our experience. But the more numbers we use, the closer we can come to
reproducing the analog original.
In digital cinematography, light and color are represented by numbers
organized in binary code. Each digit of binary code, a bit (b), defines only
two values: 0 or 1. A single bit, then, can define a scale of only two levels. But
two bits together can define four levels—two levels in each of the original
two levels (2 x 2). Three bits define 8 levels, and four bits do 16. Eight bits
are called a ‘byte’ (B). An 8-bit code defines a scale of 256 levels. The number
of levels defined by a binary code can be expressed as 2n, where n = number
of bits. The more bits used, the more levels defined, and the more precision
available to define values.
1 There is some debate as to this simple distinction. If light were modeled as discrete
particles (photons) instead of analog electromagnetic waves, it could be considered
digital. And although film grains are randomly distributed, they are separate, quan-
tifiable (not infinite) particles, which might be described using some digital format.
But for this discussion, we will consider the natural world as analog.
10
|
Film records an image by varying the density of silver or dye of the film
emulsion in a continuous gradient from clear to opaque, black to white.
Digital imaging builds an image with numbers in a binary format. The
detail displayed is limited to a scale with a discrete number of values, deter-
mined by the number of bits being used.
2) Pixels AnD ChiPs

Short for ‘PICture ELement’, a PIXEL is the building block of the digital
image. It represents one sample of picture information. Pixels are grouped
into fixed arrays of straight rows and columns. The pixels remain in the same
position, frame to frame. All pixels in an array are the same size—usually
square, but in some cases rectangular.
Instead of film behind the lens, digital cameras have chips that contain an
array of light-sensitive sensors or photoreceptors that receive image data.
The sensors convert light into voltage, an electrical charge proportional to
the light striking it. The analog voltage is sampled, or measured at speci-
fied intervals, and converted to numbers on a scale—a digital code value.
This process is called analog-to-digital conversion or A/D conversion. The
resulting digital-code values determine the brightness of each pixel viewed
on a display. The color of the pixel is defined by one of two general methods,
determined by the chip design in the camera.
The two different sensor-array designs used in production cameras today
include the three-chip design and the single-chip design. When HD video
production first developed and became an industry standard, most profes-
sional cameras used a three-chip design. Some of the most common HD
cameras in use were the SONY F900 and the Panasonic VariCam. These
three-chip cameras contain a set of three 2⁄3"-size chips, one each for mea-
suring red, green and blue light. In most cases, a prism divides the light
captured by the lens into separate red, green and blue rays, directing each
ray to a dedicated chip. (See Figure 1.) The photoreceptors on these chips are
grayscale devices that only measure
light intensity. The color of a pixel is
later determined by integrating the
respective red, green and blue light
values from the three chips.
The three-chip prism “block” sys-
tem is precise and efficient. However,
it takes up significant space in the
camera. To keep production cam-
eras small enough to be portable, the
smaller 2⁄3" chip size was used. Since
Figure 1. Three-chip prism system the 2⁄3" target is significantly smaller
| 11
Bayer Pattern 2x2 Matrix
RGB Stripe Pattern 2x3 Matrix

Figure 2. Single-chip system configurations
than a 35mm film frame, different HD lenses were manufactured to opti-
mize the optical path. Another requirement of this system is that the three
separate chips in the block must be precisely registered to align their images
with precision.
More recently, to enable the use of a larger chip (matching or exceeding
the 35mm frame size) that would provide a reduced depth of field and allow
the use of the large inventory of motion-picture lenses (optimized for the
35mm frame), camera manufacturers have moved to a single-chip format.
Single-chip cameras, such as the ARRI Alexa and RED Epic, contain a
single large chip. No prism block is needed to separate the color rays from
the spectrum. Image color is created by placing tiny red, green and blue
filters over each sensor on the chip in a particular “mosaic” pattern, then
using software matched to that pattern to calculate color values for each
pixel. Each camera uses their own proprietary chip design, mosaic filter pat-
tern, selection of filter color, and dedicated software—all of which have a
significant effect on the characteristics of the image recorded.
The most widely used mosaic filter pattern for single-chip cameras is the
Bayer pattern. (See Figure 2.) The ARRI Alexa and RED Epic both use the
Bayer pattern for their chip. This pattern uses a series of square 2x2 matrices
of filtered photo-receptor sites—two green, one red and one blue. Also called
12
|
RGBG, it is 50% green, 25% red and 25% blue. This allocation mimics the
physiology of the human eye which is more sensitive to green light. These
groups of four filtered sensors are repeated throughout the chip. Dedicated
proprietary software interprets the signal (the de-Bayering process) from
each sensor site, taking into account the particular spectral transmission of
its’ color filter along with the values of the adjacent sites in the matrix, and
assigns a color and brightness value to each pixel of the recorded image. The
image data can be recorded before this digital conversion (de-mosaic or de-
Bayer process) is performed. This ‘pre-conversion’ format is called raw, and
yields more data and a higher quality image. (See Figure 6.)
The choice of the tiny color filters on the sensors—whether narrow spec-
trum or wide band—has a significant effect on the dynamic range and color
saturation of the image captured. A wider-band filter leaves the sensor more
sensitive to light, yielding a wider dynamic range and higher native ISO.
But the color recorded by that wide-band filtered sensor is less true and less
saturated.
The Panavision Genesis camera uses an RGB stripe mosaic pattern. This
design uses a 2x3 matrix of filtered sensors (two each of red, green and blue)
to measure the data that determines the value of each pixel. In this case, the
data is “oversampled”—six sensors contribute data to define each single pixel
in the recorded image.
The SONY F65 uses a new mosaic pattern that provides red, green, and
blue data for every pixel of a recorded 4K image. The higher resolution 8K
sensor array is rotated at a 45-degree angle so as to place the filtered sensors
in position to measure all three color values for each pixel in a recorded
image, producing a ‘true’ 4K RGB output image.
3) Resolution
As the smallest element of a digital image, the pixel represents the limit of
detail that can be displayed. An image composed of a small number of pixels
can only show a rough approximation of a scene, with little detail. The more
pixels used to display an image, the finer the detail that can be revealed.
Resolution is the measure of the finest detail visible in a displayed image,
and is defined numerically by the number of pixels recorded in the image
raster—NOT by the number of sensors in the camera chip. This distinction
has created some confusion and controversy in the resolution claims of
some camera manufacturers.
Camera resolution is commonly defined by the number of lines of pixels
(scan lines) it records. The standard HD camera records 1080 lines, although
some cameras that record 720 lines are also considered HD. An increase in
the pixel line count will produce a proportional increase in resolution and
representation of fine detail. Image resolution is expressed by two numbers:
| 13
columns (or pixels per line) x lines. The HD standard image is defined as
being 1920x1080 pixels.
A doubling of pixel lines and pixels per line (as the change from 2K to
4K), increases the total pixel count by a factor of 4, requiring four times the
memory to store the image data. However, the MTF (modulation transfer
function, an optical measure of line-pair resolution) only doubles.
Some of the most common image-resolution standards include:
• Standard definition analog NTSC = 640x480 (1.33:1)
• HD = 1920x1080 (or 1280x720) (1.78:1)
• 2K = 2048 pixels per line (line count varies with aspect ratio)
• 4K = 4096 pixels per line
Pixel count is not the only element in the imaging chain that affects picture
resolution. Lens quality, precise registration of the chips in a three-chip cam-
era, and image scaling and resizing conversions also affect image resolution.
4) exPosuRe
Film has a greater dynamic range, or range of usable f-stops, than most
digital formats. And due to the response of film dyes and silver to exposure
extremes, which causes a gradual “rounding off ” of values at either end of
the tonal scale (highlights and shadows), there is a perceived extension of the
dynamic range. Shadows merge smoothly into darkness, and highlights fade
gradually (or “bloom”) into white.
Digital images, however, “clip” at either end of the tonal scale—the shad-
ows drop off abruptly into solid black, and highlights cut off into flat, white
areas with no definition. Clipping occurs when the exposure moves beyond
the specific threshold which is determined by the sensitivity and capability
of the camera or recording device.
A cinematographer can avoid clipping a digital image by monitoring and
controlling the light levels recorded in a scene. There are several useful digi-
tal tools available for monitoring exposure in a scene. A waveform monitor
displays the exposure levels across a scene, read by the camera from left to
right on a scale of 0–100 IRE (Institute of Radio Engineers). Generally, 0
IRE defines total black and 100 IRE defines total white, indicating the maxi-
mum amount of voltage that the system can handle. If the levels flatten out
at either end of the scale (0 or 100), the image will clip, and no detail will be
recorded in those areas. NTSC defines 7.5 IRE as black (called the “pedestal”
in post, and “setup” on the set). The waveform monitor displays the pedestal
and peak white levels, and indicates when clipping occurs.
Another tool is the Histogram, which graphically displays the distribu-
tion of light values in a scene, from black to white. Basically a bar chart, a
histogram indicates the proportion of image area (y axis) occupied by each
level of brightness from 0 IRE to 100 IRE (x axis). With a clear indicator of
14
|
clipping at either end of the scale (often a red line), it is useful for determin-
ing whether the scene fits within the camera’s dynamic range.
Some cameras record light levels up to 110 IRE. These brighter values can
be used throughout the post-production process, then brought back down
to a “legal” level to avoid being hard-clipped at 100 IRE when broadcast on
television.
A handy exposure tool for digital production allows the user to quickly
bring light levels in a scene within the safe range to avoid clipping. Devel-
oped by David Stump, ASC, the device consists of a hollow sphere (diameter
about 1.5 feet) painted glossy white, with a hole in one side (diameter about
2 inches) that reveals a dark interior painted matte black. To use on the set,
place the device in the brightest area of the scene, usually the position of the
main subject. Adjust the lighting and camera exposure so that the specular
(shiny) white highlight on the white ball registers under 100 or 110 IRE, and
the black hole registers 0 IRE.
Some digital cameras have software tools that help protect the shadow
and highlight detail or, if clipping is unavoidable, soften or round off the
edge of the clipping. In the toe or shadow region, the technique involves
the manipulation of the black pedestal (the level at which the image turns
black) and black stretch. Black stretch flattens the curve of the Toe, lowering
the contrast and putting more levels (and subtlety) into the lower part of
the tone scale. The resulting effect of black stretch is to reveal more shadow
detail. Going the opposite direction, “crushing the blacks” raises the slope of
the Toe curve and compresses the tonal values in the shadows. Raising the
black Pedestal turns more of the shadow area to pure black. In the highlight
region, the “soft clip” or “knee” function compresses the values near 100
IRE, rounding off the exposure curve, allowing more highlight detail to be
recorded, and giving a more pleasing shape to the edges of a clip in the very
bright areas in the frame.
5) gAmmA AnD log/lin

Film dyes have a logarithmic response to exposure. The traditional sen-
sitometric curve, used to describe the behavior of film stocks, plots density
as the y coordinate and log exposure as the x coordinate. The resulting S
curve, with a shallow slope in both the toe and shoulder regions, provides
more shadow and highlight detail. Logarithmic code values are an impor-
tant characteristic of what is considered film color space and the “film look.”
A logarithmic representation of light values is also typically used when
scanning film material for digital intermediate or visual-effects work. And
certain current digital-production cameras, such as the Arri Alexa and Sony
F65, both provide log curves as a choice for image acquisition.
CRT (cathode ray tube) monitors have a nonlinear response to any
| 15
input—the intensity of the image displayed is not directly proportional to

the video signal input. This response is called gamma, contrast or a “power
function” and is generally considered to be 2.6 for HD monitors.
Digital video cameras capture a scene as analog voltage and then digitize
it with an A/D converter to create linear digital-code values. To view these
unprocessed linear code values would require a linear display whose response
was directly proportional to its input signal. But the CRT is not a linear dis-
play; its gamma or power function curve gives a nonlinear response to input
signals. Linear digital code values do not display properly on a CRT monitor.
The tones appear desaturated with very low contrast. To generate a video
signal that will display properly on a CRT, video cameras apply a gamma cor-
rection equal to the reciprocal of the CRT power function.
To differentiate video color space from the logarithmic film color space,
video is often referred to as “linear.” This is inaccurate, however, due to the
nonlinear gamma correction applied to the video signal. Video color space
should more accurately be called a “power function” system.
As use of the CRT has decreased, new display devices have been developed
that accept a wide range of input signals. No longer constrained by the limits
of the CRT, digital cameras have been built to record image data without
applying the gamma correction of video. Different input curves have been
used in the cameras, often with some version of the log characteristic curve
of film, such as ARRI’s Log C or SONY’s S-log. Image data recorded in these
formats retains more information and can produce a higher quality image.
Instead of applying the video gamma correction, these cameras convert
the linear code values measured by the sensor into logarithmic code values
which approximate a film gamma, resulting in a wider dynamic range and
greater shadow detail.
6) ViDeo Vs. DAtA, RAW Vs. De-BAyeReD

Video vs. Data
As described above, video cameras were designed to produce an image that
displays correctly on a CRT monitor. To properly display on a CRT, a video
signal uses a gamma correction that is the inverse (reciprocal) of the CRT
gamma. The video format was also designed to limit the dynamic range of
the light values it records in order to fit into the narrower brightness range
that can be displayed by a CRT monitor. This conversion of light values into
a video signal by the camera, which involves applying the gamma correction
and limiting the dynamic range, reduces image quality and restricts its capac-
ity to portray shadow and highlight detail. Another characteristic of video is
that it records images in fields and frames, but stores them as clips—stream-
ing clips that encode all frames of a shot (head to tail) into a single file.
The highest quality video-recording format used in production today is
16
|
HDCam-SR. Originally developed for recording on videotape cassettes, the

HDCam-SR format can now also be recorded on dedicated hard drives or
solid-state media. Other video recording formats in use include DVCPro-
HD and HDCam.
More recently, digital data recording has been developed that records im-
ages in a frame-based format in which each frame is saved as a discrete data
image file. This format allows direct access to any frame without the need to
scroll through a video clip file to locate a frame. The camera and recording
system selected determine the particular recording format used. Some file-
based formats in use include .dpx, .jpg, .tiff and OpenEXR.
RAW vs. De-Bayered

As described in section “2 Pixels and Chips.” Single-chip cameras with mo-
saic patterns use software to immediately convert the light levels measured
by their filtered sensors into RGB values. This conversion process (called de-
Bayering, referring to the Bayer mosaic pattern) allows real-time viewing of
the image. However, some quality is sacrificed by performing the de-Bayer
process in real time.
A “raw” recording file format was developed to encode all the light-
level data directly from the filtered sensors, before the de-Bayer process
occurs. A raw recording of an image can result in a significantly higher
quality image. A raw file will require de-Bayer conversion for viewing and
most manipulation, but most professional systems used today provide that
capability. Some DI (digital intermediate) color-correction systems allow
raw files to be used seamlessly in their workflow. Others require a de-Bayer
conversion to some RGB data format before proceeding. Cameras that re-
cord raw data use their own proprietary raw file format, such as Redcode
Raw and ArriRaw.
7) ViDeo sCAnning—
inteRlACe oR PRogRessiVe
The video image is composed of a certain number of lines of pixels. A light
or energy beam scans the lines to “write out” each frame. Since the begin-
ning of television broadcasting, interlaced scanning was the standard. In this
process, two vertical sweeps are made for each frame—one scans all the odd
lines (1, 3, 5, etc.), the next scans the even (2, 4, 6, etc.). Each sweep produces
a “field.” The NTSC 30 fps television standard records 60 interlaced fields per
second.
Progressive scanning is simpler—all lines are scanned sequentially in one
vertical sweep. There are no fields, only complete frames, just as in motion-
picture film photography. Computer screens use progressive scanning.
When broadcast standards were first being established, the bandwidth of
| 17
the equipment was relatively small. That is, the throughput of the pipeline
used to transmit image data, measured by the amount of data it could trans-
mit over a given period of time, was very limited. The existing equipment
was not capable of sending an entire (progressive) frame 30 times per sec-
ond. It could, however, send 60 half-frames (the fields of interlaced frames)
per second, since this requires transmitting only half the frame data at any
given moment.
One benefit of interlaced scanning is that the higher frame rate (60 fields,
instead of 30 frames) makes the portrayal of movement clearer and more ac-
curate. At 24 progressive frames per second, or even 30, the shutter speed is
slow (1⁄48 or 1⁄60 second), and generates motion blur with any action. Camera
movement can appear juddery over certain backgrounds. Interlaced fields,
exposed at twice the shutter speed and frame rate, reduce these problems
and give a sharper, more precise portrayal of movement. Subjectively, an
interlaced image appears to have greater detail—an immediacy associated
with television viewing.
Image flicker is also reduced with interlaced scanning. CRT monitors have
a short persistence—the scanned image fades quickly, leaving a black screen.
At 24 or 30 fps, there would be more black-screen time, causing noticeable
flicker on the monitor. Interlaced scanning at 60 fields per second, refreshing
the image twice as fast, greatly reduces flicker. Progressive computer moni-
tors avoid the flicker problem by using a much higher frame rate.
The downside of interlaced scanning is a loss of resolution and steadi-
ness when shooting movement. With subject or camera movement, the
image changes from one field to the next. But those two consecutive fields
perceptually merge to create each complete frame, even though the mov-
ing subject has changed position. Therefore, any movement will appear
blurred and detail will be substantially reduced in areas of the frame that
contain movement. This factor also creates difficulty for any image pro-
cessing that involves spatial manipulation of the image, such as resizing or
reframing.
A film release of interlaced material requires the creation of progressive
frames from pairs of fields, so as to convert the interlaced video fields back to
progressive film frames. There are various methods of doing this video-field-
to-film-frame conversion. Some involve the interpolation of pixel values
between the different fields. This merging of field pairs into frames reduces
resolution wherever there is movement. For this reason, the progressive
format is preferable for recording back to film.
Some cameras offer the option of recording in either a progressive or
interlaced format. A key factor in deciding which format to use should be a
consideration of the primary distribution goal of the project—be it theatrical
screen, broadcast or DVD/Blu-ray.
18
|
8) ComPRession
One of the biggest challenges to creating high quality digital images is
the handling of the large amounts of data they require. Compression is the
means used to reduce the data content of an image source. The objective of
compression is to reduce image file size with the least possible sacrifice in
quality and detail.
A common method used by compression schemes or codecs is to analyze
an image, identify redundant pixels in the picture, and remove them. For
instance, a stationary solid blue sky is “redundant” and can be easily com-
pressed. Conversely, subjects in motion change position every frame, are not
redundant and are difficult to compress.
There are two categories of compression codecs—intraframe and inter-
frame. The intraframe codec processes each frame individually, only remov-
ing redundant information from within that particular frame. The interframe
codec uses a series of frames, or group of pictures (GOP), to compress the
image data. Interframe compression compares consecutive frames within
each GOP to remove redundancy from frame to frame and arrive at “dif-
ference” information. Although it’s more efficient and can achieve a higher
compression ratio, interframe compression can create challenges in editing
and postprocessing due to the multiframe dependency.
Compression is quantified as a ratio, comparing the amount of data in the
original noncompressed signal to that in the compressed version. The lower
the compression ratio, the higher the quality of the resulting image. A format
that uses a 2:1 compression ratio preserves more image information than
one that uses a 4:1 compression ratio. Some cameras and recording systems
offer several compression ratios, providing a choice between higher quality
with larger image files and lower quality with smaller files.
Some compression codecs allow the original image to be fully reconstitut-
ed when decompressed. Such a codec uses “lossless” compression. However,
a danger exists with certain codecs that claim to be “visually lossless.” The
claim may hold true for direct display of an original image that needs no ma-
nipulation in postproduction, but if it later requires significant color grading
or visual effects, disturbing artifacts may appear that significantly reduce the
quality of the image. Other codecs discard image information that cannot be
subsequently retrieved. This is considered “lossey” compression.
9) ColoR sPACe
A color space is a framework that defines a range of colors within a color
model. Most color-space systems use a three-dimensional model to describe
the relationship among the colors. Each color space has a set of specific pri-
mary colors—usually three (a particular red, green and blue)—which shapes
its color model.
| 19
The color space can be device-dependent or device-independent. A

device-dependent system is limited to representing colors from a particular
device or process, such as film. The RGB film color space describes the range
of colors created by the color dyes used in film. A device-independent color
space defines colors universally and, through a conversion process using a
“profile” can be used to define colors in any medium on any device.
The color gamut is the range of colors a system can record or display. When
a color cannot be correctly produced on a particular device, it is considered
“out of gamut” for that device. Film has a different color gamut, for instance,
than a CRT monitor. Some colors recorded on film do not properly display
on a monitor. Likewise, some extremely saturated, high-intensity colors that
can be displayed on a CRT monitor cannot be reproduced on film.
Different media and devices can have widely diverging color characteris-
tics and gamuts. Yet the hybrid film/digital production workflow passes an
image from one medium to another, and from device to device—shooting
film, scanning it to digital files, viewing it on various monitors and projec-
tors, then recording it back to film. Accurate format conversions are re-
quired at each stage to preserve image quality and color-space information.
A device-independent color space is effective for accurately transferring and
converting color information from one system, format or device to another.
The challenge is to avoid losing data as the image is processed step-by-step
through the workflow.
The CIE (Commision Internationale de L’Eclairage) XYZ color space has
long been the industry’s standard device-independent reference. Although
this color space uses a three-dimensional model to represent colors, one can
use it to plot the visible color spectrum on its flat chromaticity diagram. This
diagram is actually a 2-D “slice” taken from the 3-D model of the color space.
It has a horseshoe shape and leans left on the x-y axis. (See Figure 3.) All colors
visible to human perception can be plotted on this graph. The colors of the
spectrum lie along the horseshoe curve, left to right, blue to red. The specific
red, green and blue primaries are specified as points on the graph. The neutral
white point is nominally located where the three primaries are equal in con-
tribution, but to accommodate different color temperatures, their respective
white points are plotted along a curve across the center area of the model.
Specific colors are identified in the 2-D space by a set of two numbers, called
chromaticity coordinates (x, y), which specify their position on the diagram.
One method to compare media and devices is to plot their respective
color gamuts as triangular shapes within the chromaticity diagram. The
full horseshoe-shaped space represents the full range of human vision. A
triangle within this colored shape approximates the gamut of digital cinema.
A smaller triangle inside the digital-cinema triangle represents the more
limited gamut of HD video color space (Rec. 709).
20
|
Color spaces have been developed to define color for many particular
purposes, including film, video and digital-cinema display. The standard
HD video color space (ITU-R BT.709 or Rec. 709) uses YCrCb space, which
separates luminance (Y) and chrominance (Cr and Cb—red and blue color
values). This separation allows for choices in color sub-sampling—for
instance, the 4:2:2 sampling ratio measures color half as often as it does
luminance, thereby saving memory space and time. (See section 12.) Film
is represented by the RGB color space, in which the red, green and blue
channels are sampled equally. Both the YCrCb and RGB color spaces are
device-dependent.
The color space designated as the standard for Digital Cinema distribu-
tion is X'Y'Z,' a gamma encoded version of the CIE XYZ space on which
the chromaticity diagram is based. A device-independent space, X'Y'Z' has
.9
.8
.7
.6
.5
2000º K
.4 rve
Cu
dy
B o 4000º K 1000º K
k
ac
Bl
6000º K
.3
10000º K
.2
.1
.0
.0 .1 .2 .3 .4 .5 .6 .7 .8
Figure 3. Chromaticity diagram.
| 21
a larger color gamut than other color spaces, going beyond the limits of
human perception. All physically realizable gamuts fit into this space.
10) ACes
The digital workflow of a production today involves the passing of an
image, with a particular Look attached, from facility to facility and system to
system, requiring conversions and transfers to different formats and media.
The risk of failing to achieve the intended result is significant—image quality
(including resolution, exposure and color data) can be lost along the way.
Look management and color management, discussed below, help reduce
these risks, but a larger framework for the entire workflow has been missing.
The Academy of Motion Picture Arts and Sciences’ Science and Technol-
ogy Council developed ACES—the Academy Color Encoding System. ACES
provides a framework for converting image data into a universal format. The
ACES format can be manipulated and output to any medium or device. The
system specifies the correct path from any camera or medium to ACES by
using a dedicated IDT (input device transform). Camera manufacturers such
as Sony and Arri have already generated an IDT for their camera systems.
Display device manufacturers of projectors and monitors provide an ODT
(output device transform) for their devices. Different ODTs are also needed
for each distribution medium, including digital cinema projection, Blu-ray,
and HD broadcast. A final color grading “trim pass” is still recommended
for each master so as to insure fidelity to the intended look.
The most striking gain for the cinematographer is the ability for ACES to
encode the full range of image information captured by any camera. Using a
16-bit floating-point OpenEXR format, ACES has the potential of encoding
the full gamut of human vision, beyond what any camera can record today.
It accommodates a dynamic range of 25 stops, far past the capability of any
device. Highlights that were clipped and shadows that have been crushed
using other formats and workflows now can re-emerge with surprising clar-
ity. ACES empowers the cinematographer to use the full capability and range
of the tools at his or her disposal.
11) Bit DePth

Bit depth determines the number of levels available to describe the bright-
ness of a color. In a 1-bit system, there is only 0 and 1, black and white. An
8-bit system has 256 steps, or numbers from 0–255 (255 shades of gray).
Until recently, 8-bit was standard for video and all monitors. Most monitors
still have only 8-bit drivers, but several HD video camera systems support
10-bit and 12-bit signals through their image-processing pipelines.
A 10-bit system has 1024 steps, allowing more steps to portray subtle
tones. A linear representation of light values, however, would assign a dis-
22
|
proportionate number of steps to the highlight values—the top half of the

full range (512–1024) would define only one f-stop, while leaving 0–512 to
define all the rest. A logarithmic representation of code numbers, however,
spreads equal representation of brightness values across the full dynamic
range of the medium. In 10-bit log space, 90 code values are allocated for
each f-stop. This allows for more precision to define shadow detail. For this
reason, 10-bit log is the standard for recording digital images back to film.
Some color publishing applications use a 16-bit system for even more con-
trol and detail. The cost of the additional bits is memory and disk space,
bandwidth and processing time.
The consequence of too few bits can be artifacts, or flaws in the image
introduced by image processing. Artifacts include banding, where a smooth
gradient is interrupted by artificial lines, and quantization, where a region of
an image is distorted. If image data were recorded or scanned in 10-bit color,
down-converted to an 8-bit format for postprocessing, then up-converted
back to 10-bit for recording back to film, image information (and usually
quality) will have been lost and cannot be retrieved. Whenever possible, it
is preferable to maintain the highest quality of image data and not discard
information through conversion to a more limited format. Loss of image
information can also result from reduction in color space and gamut, color
sampling, and resolution.
12) ColoR sAmPling

Color sampling describes the precision of the measurement of light and
color by the camera system. It is represented by three numbers, separated
by colons, and refers to the relative frequency of measurement, or sampling.
The notation 4:4:4 represents the maximum possible precision, in which all
values are measured at equal intervals.
In a video color space, where luminance and chrominance are differenti-
ated, the first number represents how often the luma (brightness) signal is
sampled (measured) on each line of sensors. The second number indicates
how often the color values (red-minus-luma and blue-minus-luma signals)
are sampled for the first line. The third tells how often the color values are
sampled for the second line. The two lines are differentiated to accommo-
date interlaced systems. (See Figure 4.)
4:4:4 captures the most information, sampling color at the same frequency
as luminance. 4:2:2 is the current standard on HD production cameras. The
color information is sampled at half the frequency of the luminance infor-
mation. The color precision is lower, but is adequate in most production
situations. (Human vision is similarly more sensitive to brightness changes
than it is to color variation.) Problems can arise, however, in postprocess-
ing, such as in the compositing of greenscreen scenes, where color precision
| 23
4:1:1 4:2:0 4:2:2 4:4:4

Figure 4. Color sampling ratios
is important. It is recommended that any visual-effects shots that require
image processing be recorded in a 4:4:4 format.
13) look mAnAgement

AnD ColoR mAnAgement
The “look” is the visual style of the project. As author of the image, the cin-
ematographer is responsible for the look. While supporting the director’s vision
of the script and integrating the contributions of the other creative members of
the team, he or she should develop, execute and maintain the look throughout
the Workflow of the project. This process is look management.
Understanding the fundamentals of digital technology and the produc-
tion tools available empower the cinematographer to maintain the look and
protect the creative intent of the filmmakers. Look-management software
provides visualization tools that allow users to manipulate images to create a
particular look. That Look can then be saved in a format that can be reliably
and consistently shared with other team members. The software used ranges
from photography apps, such as Adobe Photoshop, all the way to powerful
systems created for motion-picture finishing, such as FilmLight’s on-set app,
Truelight. Other look-management systems in use include Technicolor’s
DP Lights, Assimilate’s Scratch, Adobe’s Speedgrade, Fujifilm’s CCBox and
Blackmagic Design’s DaVinci Resolve.
integrated throughout the Workflow

Look management starts from the very beginning of a project, and is active
throughout the workflow, from previsualization of the look in prep through
to the final grading of the distribution master. To start, test and scouting im-
ages can be gathered, discussed and processed with software to create a set of
“hero” images or shots that encapsulate the look of the project. These images
are saved along with their color-adjustment settings, and are used throughout
the production workflow to communicate and maintain the look.
The hero images and look settings can be used on set during production,
using look-management software, to support the crew as the set is dressed
and lit to fulfill the look. Look management can then guide the dailies timer
24
|
or Colorist as they embed the Look in the dailies they create. Visual-effects
artists depend on look management to enable them to process shots that
fit seamlessly back into the edited project. Additionally, look management
communicates the cinematographer’s intended look to the DI colorist for
the final grading sessions.
A universal interoperable Format

To reliably communicate and reproduce the Look throughout the Work-
flow, the instructions should be encoded into a format that can be under-
stood universally and easily read by all the different applications, display
devices and facilities that are used on the production. The laboratory printer
timing-light number was the standard in existence before digital technol-
ogy. Until recently, no such universal digital format existed to encode a look,
leaving plenty of room for miscommunication and misunderstanding.
The ASC Technology Committee’s Digital Intermediate Subcommittee
took a significant step toward solving this problem with the ASC CDL
(color decision list). Designed to advance the industry toward the goal of
interoperability, the ASC CDL encodes basic color-correction instructions
that can be read by many different software systems, facilitating communi-
cation and collaboration among the many artists and facilities working on
a production. The ASC CDL uses 10 parameters to encode the Look of an
image: slope, offset, and power for each of the three color channels (red,
green, blue) plus a single value for saturation. Slope, offset and power cover
a similar range of adjustment to the familiar telecine parameters gain, lift
and gamma, but have a cleaner and purer logic and math. The telecine func-
tions also tend to vary in definition from system to system. The ASC CDL
format is now integrated into most professional look management and DI
grading systems, and has become a reliable way to share and exchange Look
information.
Working together with the Academy’s Science and Technology Commit-
tee, the ASC Technology Committee has also developed a universal format
for the look-up table (LUT), another digital tool often used to encode Looks
and communicate visual choices.
Several camera systems have dedicated “LUT boxes” that generate and
apply looks that were created either in-camera or imported from external
media. A cinematographer can save encoded looks from one project and
bring them to use on another.
The same industry committees have also been developing a standard for
encoding and sharing of all metadata. Metadata encompasses all the nonim-
age data associated with a particular frame. It can be recorded within the
frame’s digital file header or, alternatively, in an external database. Metadata
guides the processing of an image that is necessary to preserve the look as
| 25
intended. This can include frame number, resolution, frame rate, motion-
control data, as well as any look-management data, such as the ASC CDL
values. As the production image is passed through the numerous postpro-
duction steps in the digital workflow, it is important to preserve all metadata
associated with an image and pass it on intact to the next stage.
This discussion of the look and look management describes a “nonde-
structive” process of creating a look and recording its parameters in meta-
data to a format such as the ASC CDL. The original photographed image is
not altered until the final DI session, thereby preserving maximum image
quality (resolution and color data) and full potential for post manipulation.
Another working style instead applies the Look modifications to the image
on the spot during production, either using a LUT, a camera menu scene
file, or an analog device, such as a glass filter. This style “bakes” the Look
into the image, in effect marrying the look to the image. This workflow is
often used by productions seeking to avoid extensive postproduction work.
The disadvantage of baking-in a look, however, is significant. Depending
on the modifications used to create the look, image quality may be reduced,
whether in dynamic range, color record or resolution. The potential for later
image manipulation in the DI may have been compromised, making it diffi-
cult or impossible to undo or alter the look already imposed upon the image.
Color management, Calibration and Viewing environment

An important requirement for successful look management is rigorous
control and consistency of the display of the digital image. This involves
color management, monitor or projector calibration, and control of the
viewing environment.
With film, an image can be reproduced with reliable consistency—once
correct printer lights are set for a given negative, the look of the resulting
image is reasonably predictable. Digital images, however, require proper
color management to assure their proper display. Particularly for the pur-
pose of communicating a precise look, it is important that the image viewed
display correct color, contrast and brightness.
When the cinematographer creates a look with camera and software, he
or she views it on a particular monitor or projector. But when it is sent to
the other members of the creative team, such as the director and production
designer, it is likely to be viewed on different systems. How can the cinema-
tographer ensure that the others see the same look—complete with the right
colors, contrast and exposure?
First, the system (both hardware and software) that reads the image file
must be able to understand the format in which the image is encoded, and
be able to interpret all the metadata in the file. Next, the monitor or projector
used to view the image must be properly calibrated to a shared standard
26
|
(such as SMPTE bars) according to the device manufacturer’s instructions.

Finally, the image should be viewed in a controlled environment, consider-
ing both brightness level and the surrounding color. The same image viewed
on a monitor in a sunny, warm room will look very different when viewed in
a dim, blue environment.
With good color management, an image portraying a look can be success-
fully shared electronically, in an instant, around the world. Without color
management, the exercise is pointless.
14) DigitAl inteRmeDiAte

Although the term “digital intermediate” was originally defined as the
entire digital postproduction process that comes between principal photog-
raphy and the final distribution master, today the DI commonly refers to
the color-grading session in which all visual elements of the production are
brought together for the purpose of making the final adjustments, enhance-
ments, effects and manipulations. (See Figure 5.)
Given the potential for reshaping the look of the picture in the DI session,
the cinematographer needs to participate actively in the process, either in
person or by remotely communicating his or her vision through images and
software. The DP can use look-management software to generate a detailed
set of instructions for the colorist. Taking representative frames from each
shot in the EDL (edit decision list), he or she can adjust the color, contrast
and density of each frame to create a scene-by-scene guide to the intended
look for the picture. This process depends on precisely calibrated display
devices and accurate digital format transforms. Format and color space
must be coordinated so that the samples provided by the cinematographer
use the same “palette” (determined by a color space’s gamut and dynamic
range) that is used by the colorist. Using ACES, this consistency is assured.
The best quality and highest resolution image material should always be
used in the DI session. Since the editorial process is usually performed with a
lower-resolution image, the EDL is used to conform the hi-res original to the
low-res edit, generating a complete high-resolution copy of the project for
use in the DI session. The final grading can be performed using any digital
color-correction system, but several powerful systems have been developed
to expand the creative potential of the DI session, including FilmLight’s
Baselight, Autodesk’s Lustre, and Blackmagic Design’s DaVinci Resolve.
Following are principal activities that can be performed by these systems
in the DI session:
1) Primary color correction: Independently adjusting the lift, gain, and
gamma of the RGB color channels .
2) Secondary color correction: Isolating specific colors in the spectrum,
modifying their hue, saturation and brightness.
| 27
Camera Original Graded Master

Figure 5. Digital intermediate color grading.
3) Region-of-interest correction: Using “windows” or articulated mattes

to isolate specific areas of the frame (which may follow or track moving
subjects) to make further primary and secondary corrections.
4) Spatial corrections;
a) Repositioning the frame, zooming and panning.
b) Applying digital filters to blur, flare or sharpen.
c) Reducing (or amplifying) grain or noise.
5) Stylized looks:
a) Emulating lab processes, as bleach bypass or ENR.
b) Overall color-gamut manipulation.
15) mAsteRing
Once the final grade is completed in the DI session, a digital source master
(DSM) is rendered, creating new altered digital image files that encompass
all the corrections and enhancements made during the DI. This master is the
source for the versions or sub-masters that will be made for distribution in
each format. Sub-masters are made for distribution in 35mm, digital cinema,
HD and SD broadcast, Blu-ray, DVD, and for online streaming and down-
loading. Creation of the sub-master in each medium requires a format con-
version—and may then require supplemental color grading (a “trim pass”),
so as to come as close as possible to matching the source master created in
the DI session. Each medium, however, has different characteristics, includ-
ing resolution, dynamic range and color gamut, and cannot exactly match
the original DSM. To ensure that the look is preserved in each distribution
format, the cinematographer should stay in the loop as the sub-master for
each medium is created and, if possible, attend the grading sessions or trim-
passes for each. The objective is to preserve the look—the original creative
intent of the filmmakers.
Film
Creating a 35mm film master involves conversion of the source master to
a digital format shaped to take best advantage of the capability of film. Tra-
28
|
ditionally, the Cineon 10-bit log format has been used in film recorders, but
by using ACES, a higher quality output format could be used. The converted
digital files are uploaded into a film recorder—either laser or CRT-based—
which records the digital files onto 35mm film, exposing a new film master.
Digital Cinema
For the digital-cinema release, the DCI (Digital Cinema Initiatives—a
joint effort of the major studios) and the SMPTE have established a set of
universal format specifications for distribution—the DCDM and DCP. The
specifications define the elements (“assets”) used for digital-cinema display,
and a structure by which the assets should be used for successful exhibition.
The assets include the picture (Digital Source Master), soundtrack, subtitles,
metadata and security keys.
The DCDM (digital cinema distribution master) incorporates all specified
assets in an uncompressed format. The DCDM encodes the image in the
device-independent CIE X'Y'Z' color space as TIFF files, with 12 bits per
color channel. This color space accommodates the entire human visual color
spectrum and gamut. It has two resolution formats: 2K (2048x1080) and 4K
(4096x2160). Source images of other sizes and aspect ratios can be stored
within the 2K or 4K image containers. The DCDM also accepts 3-D formats.
The DCP (digital cinema package) is a compressed, encrypted version of
the DCDM. Using JPEG 2000 compression, the DCP is used for the efficient
and safe transport of the motion-picture content. Upon arrival at the theater
(or other exhibition venue), the DCP is unpackaged, decrypted and decom-
pressed for projection.
16) ARChiVing
Finally, all essential digital data should be properly archived for future
generations. Secure storage should be arranged for the digital source
master code (or ACES master data), and transferred periodically to fresh
media to avoid any degradation or corruption of data. Digital data is vul-
nerable on all currently available media, whether it be magnetic tape, disk,
hard drive or a solid-state medium. The frequency of transfer is dictated
by the conservatively estimated life span of the medium being used. The
only proven, long-term archive solution is film—black-and-white C/M/Y
separations. Properly stored, black-and-white film negative can last at least
100 years.
ConClusion
With a working understanding of digital technology, the cinematographer
can confidently choose the best methods and devices suited to any project in
production. As new challenges arise and new technology becomes available,
| 29
he or she can better know what issues to consider in making the technical
and creative decisions that shape a career.
A working cinematographer, director and software designer, Marty Ollstein

is a contributing member of the ASC Technology Committee, a Fellow of
SMPTE, and member of the WGA.
He received his MFA in Film Production from UCLA, and developed Crystal
Image, the first cinematography filtration software, used on Cineon and Flame
visual-effects systems.
30
|
| 31
A Primer for Evaluating

Digital Motion
Picture Cameras
by Rob Hummel
editor of the 8th edition of
the American Cinematographer Manual
DigitAl Motion PiCturE CAMErAs
S uffice it to say that any aspect ratio is achievable with the current digital
motion picture cameras available. Their capture format is generally that
of HDTV (1920x1080) or what is commonly called 2K.1 The aspect ratio of
all but one or two of these cameras is the HDTV aspect ratio of 1.78:1 (16 x 9
in video parlance). This 1.78:1 aspect ratio is a result of the different cam-
era manufacturers leveraging what they have built for HDTV broadcasting
cameras. It’s unique to find a digital camera that doesn’t owe part of its de-
sign to television cameras.
When composing for 1.85, digital cameras generally use almost the en-
tire 1.78 imaging area. When composing for the 2.40:1 aspect ratio, most
digital cameras will capture a letterboxed 2.40 slice out of the center of the
imaging sensor, which, in 1920 x 1080 cameras, results in the pixel height
of the image being limited to only 800 lines (Star Wars Episode 3, Sin City,
Superman Returns).
There is one digital camera that employs Anamorphic lenses to squeeze
the 2.40 aspect ratio to fit within its sensor’s 1.97 aspect ratio, utilizing the
entire imaging area.
Yet another camera does creative things with the clocking of CCD pixels
so that the entire imaging area is still utilized when shooting a 2.40 image,
with a subtle compromise in overall resolution.
In the future, it is likely that more and more cameras will have imaging
sensors with 4K pixel resolution.
1. When film resolutions are discussed, and the terms 2K or 4K are used, these refer
to the number of lines that can be resolved by the film. In the case of 2K, that would
mean 2048 lines or 1024 line pairs as photographed from a resolution chart. In the
case of 4K, that would mean 4096 lines or 2048 line pairs (4096 lines) as photo-
graphed from a resolution chart. In digital imagery the term is applied a bit more
loosely. While 2K and 4K still mean 2048 and 4096, respectively, with digital scan-
ning and photography it refers to a number of photo sites on the scanner or camera
chip. Numbers of pixels does not necessarily translate into actual image resolution.
32
|
A Primer for Evaluating Digital Motion Picture Cameras
In a 1920 x 1080 Digital 1920

Camera, only the center 800
lines are used for a “Scope” or
2.40:1 aspect ratio.
1080
2.40:1 Scope Aspect Ratio 800
rEsolution vs.
MEgAPixEls
When it comes to deter-
mining which digital camera
to choose, don’t allow claims Figure 1. In a 1920 x 1080 Digital Camera, only the
of “megapixels” to influence center 800 lines are used for a “Scope” or 2.40:1
aspect ratio.
your understanding of what a
camera is capable of. It is a term that is used loosely to indicate the resolution
of a digital camera that doesn’t follow any set guidelines. There are many fac-
tors that would have to be taken into account if you were going to evaluate the
merits of a digital imaging device based on specifications alone.
The most straightforward way to understand a digital motion picture
camera’s capabilities is to shoot your own rigorous tests and evaluate them.
In the same manner when a new film stock has been introduced, the best
way to truly understand that emulsion’s capabilities is to test it, rather than
rely on the claims of the manufacturer.
Also, it will be less confusing if you focus your evaluations on the final
image delivered by a given digital camera. Claims about a camera’s imag-
ing sensor can be influenced by marketing. If we concentrate on the final
processed image that is delivered for projection, color correction, and final
presentation, we will be evaluating a camera’s true caliber.
sCAnnEr vs. CAMErAs

To clarify, at the risk of confusing, 2K and 4K film scanners generally
capture more information than their camera counterparts; a 2K film scan-
ner will usually have a CCD array of 2048 x 1556, while a 4K scanner will
capture a 4096 x 3112 image. The lesson here is that one’s definition of the
dimensions of 2K or 4K can vary; the terms 2K and 4K are only guidelines,
your mileage may vary.
It is important to understand these variables in characteristics, and the
need to be very specific when describing the physical characteristics of
film cameras, digital cameras, scanners and telecines. In the world of film
cameras (and film scanners), 2K refers to an image which is 2048 pixels
horizontally (perf to perf) and 1556 pixels vertically. This image captures the
area of either the SMPTE 59 Style C full aperture frame (.981" x .735") or the
SMPTE 59 Style B sound aperture frame (.866" x .735").
A 4K scan captures the same areas of the film frame as a 2K scan, but
| 33
the image captured is 4096 x 3112. With both 2K and 4K scanners, each
individual pixel contains a single unique sample of Red, Green and Blue.
In the digital camera world, 2K often refers to an image that is 1920 pixels
horizontally and 1080 pixels vertically. Again, each individual pixel contains
a single unique sample of Red, Green and Blue. This sampling of a unique
Red, Green and Blue value for each pixel in the image is what is called a
“true RGB” image, or in video parlance, a 4:4:4 image. While these cameras
have an image frame size that corresponds to HDTV standard, they provide
a 4:4:4 image from the sensor, which is not to HDTV standard; which is a
good thing, as 4:4:4 will yield a superior picture.
Fill FACtor
There is one area of a digital camera’s specifications that is most helpful
in determining its sensitivity and dynamic range. This is the statistic that
conveys how much area of an imaging sensor is actually sensitive to, and
captures light vs. how much of a sensor is blind, relegated to the circuitry
for transferring image information. This is called the “fill factor.” It is also a
statistic that is not readily published by all camera manufacturers.
This is an area where not all digital cameras are created equal. The amount
of area a digital imaging sensor is actually sensitive to light (the “fill factor”)
has a direct correlation to image resolution and exposure latitude. With the
currently available professional Digital Motion Picture Cameras, you will
find a range of high profile cameras where less than 35% of the sensor’s total
area is sensitive to light, to cameras where more than 85% of the sensor is
light sensitive.
As film cinematographers, we are used to working with a medium where
it was presumed that the entire area of a 35mm film frame is sensitive to
light; in digital parlance, that would be a fill factor of 100%. When a digital
camera has a fill factor of 40%, that means it is throwing away 60% of the
image information that is focused on the chip. Your instincts are correct if
you think throwing away 60% of image information is a bad idea. With this
statistic, you can quickly compare camera capabilities, or at least understand
their potential.
The higher the fill factor of a given sensor (closer to 100%), the lower the
noise floor will be (the digital equivalent of film grain) and the better the
dynamic range will be.
DigitAl sEnsors AnD

AMount oF light thEy CAPturE
Since the imaging sites on a solid state sensor are arrayed in a regular grid,
think of the 40% sensitive area as being “holes” in a steel plate. Thus, the
image gathered is basically similar to shooting with a film camera through
34
|
a fine steel mesh. You don’t actually see the individual steel gridlines of the
mesh, but it tends to have an affect on image clarity under most conditions.
In terms of sensor types with progressively more area sensitive to light,
there are basically two categories: photodiodes (less area) and photogates
(more area). Depending on the pixel size, cameras utilizing a single-chip
photodiode interline transfer array (either CCD or CMOS) would be on the
low end with less than 40 to 35% of its total area sensitive to light, up to a
theoretical maximum of 50% for multichip (RGB) photodiode sensors. Next
would be single-chip photogate based sensors that can, again, depending on
pixel size, have anywhere from 70 to over 85% of its area sensitive to light.
In light sensitivity, as in exposure index, photodiode sensors will have a
higher sensitivity than photogate sensors, albeit with an associated trade off
in latitude and resolution.
In addition, solid state sensors tend to have various image processing cir-
cuits to make up for things such as lack of blue sensitivity, etc., so it is impor-
tant to examine individual color channels under various lighting conditions,
as well as the final RGB image. Shortcomings in automatic gain controls, etc.
may not appear until digital postproduction processes (VFX, DI, etc.) begin
to operate on individual color channels.
MorE on MEgAPixEls
Much confusion could be avoided if we would define the term “pixel”
to mean the smallest unit area which yields a full color image value (e.g.,
RGB, YUV, etc., or a full grayscale value in the case of black and white).
That is, a “pixel” is the smallest stand-alone unit of picture area that does not
require any information from another imaging unit. A “photosite” or “well”
is defined to be the smallest area that receives light and creates a measure of
light at that point. All current Digital Motion Picture cameras require infor-
mation from multiple “photosites” to create one RGB image value. In some
Photogate Sensor Photodiode Sensor

Photosites/
Wells
RED GREEN BLUE
BLUE GREEN
RED GREEN BLUE

GREEN RED
Photosites/Wells
Figure 2a. A Bayer pattern Photogate sensor, Figure 2b. A Photodiode “macrocell”
where each photo site is transformed to yield design, where it takes six photosites to
one full color pixel value. yield one full color pixel value.
| 35
cases three photosites for one RGB value, in others, six photosites to create
one RGB image value, a then Bayer pattern devices that combine numerous
photosites to create one RGB value.
These definitions inevitably lead us to define the term “resolution” as the
number of pixels that yield a single full color or grayscale image value (RGB,
YUV, etc.).
The images above are just two of many examples of photo sensors used
by digital cameras. One illustrates a Bayer pattern array of photosites using
a photogate sensor, and the other an interline transfer array of photosites
employing a photodiode lenslet design.
sounD CoMPliCAtED?
Perhaps, but all you need to understand is that product claims can, and
will, be misleading. We’ve lost our way a bit when thinking that counting
pixels alone is a way of quantifying a digital camera’s capability. A return to
photographing resolution charts, and actually examining what these cam-
eras are capable of will serve you much better in understanding how a given
camera will help you tell your story.
In short, do not be seduced by technical specification mumbo jumbo.
Look at images photographed by the camera in question, and evaluate from
a proper distance of screen heights from the image.
© 2007 Robert C. Hummel III • Special thanks to Stephen H. Burum, ASC,

Daniel Rosen, Garrett Smith, Evans Wetmore, and Anne Kemp Hummel in
helping bring greater clarity to this discussion.
36
|
| 37
Digital Cinematography
on a Budget
by M. David Mullen, ASC
B efore the year 2000, video technology had only been used sporadically
for theatrical releases, mainly documentaries. For narrative fiction,
milestones include Rob Nilsson’s independent feature Signal 7 (1985), shot
on ¾" video, and Julia and Julia (1987), shot on 1125-line analog high-def-
inition video. However, using video technology for independent features—
primarily as a lower-cost alternative to film—didn’t catch on until a number
of elements fell into place: the introduction of digital video camcorders,
desktop computer nonlinear editing systems, and the increase in compa-
nies offering video-to-film transfer work, all of which began to appear by
the mid 1990s.
The major turning point came with the worldwide box office success of
two features shot on consumer camcorders, the Dogma’95 movie Festen (The
Celebration) (1998) and The Blair Witch Project (1999), proving that cinema
audiences were willing to watch movies shot in video if the subject matter
was compelling enough and the visual quality seemed to suit the content.
However, with 35mm film being the gold standard for narrative feature
production, many independent filmmakers have pushed manufacturers to
develop affordable technology that would bring an electronic image closer
to the look of 35mm photography.
24 fps progressive-scan (24P) digital video appeared in 2000 with the
introduction of the Sony HDW-F900 HDCAM pro camcorder; then in
late 2002, Panasonic released the AG-DVX100, a Mini-DV “prosumer”
camcorder with a 24P mode that cost less than $5000. Not long after that,
lower-cost high-definition video cameras appeared, starting with the JVC
GR-HD1 HDV camcorder in late 2003. The next significant trend was the
movement away from traditional tape-based recording, allowing greater
options in frame rates and recording formats within the same camera. The
first prosumer camera with this design approach was the Panasonic AG-
HVX200, released in late 2005; it came with a standard Mini-DV VTR but
could also record footage to P2 flash memory cards.
By 2009, there were HD video cameras being sold on the market for less
than $1000, and now even phones and still cameras are capable of shooting
HD video. Today many movie theaters are being converted to digital projec-
tion; this trend—combined with emerging online distribution schemes—
38
|
Digital Cinematography on a Budget
has diminished the need for a digital feature to be transferred to 35mm film
(though there remains good archival reasons for doing this).
The term “prosumer” vaguely covers a range of lower-cost video products
with a mix of consumer and professional features, often in a small-sized
camera body. Prosumer cameras, by definition, are not only used by con-
sumers but by professionals as well, either for cost reasons, or because their
portability and low profile are better suited to a particular type of produc-
tion. In fact, some of the cameras discussed in this article are actually made
and sold by the professional division of the manufacturer and thus are not
technically prosumer products.
Therefore, rather than separate cameras into debatable categories of “pro-
fessional/prosumer/consumer,” I will make the cut-off point for discussion
any camera sold for under $15,000 and under that is capable of professional-
quality video, preferably with a 24P or 25P option.
In this article, “SD” refers to standard definition video and “HD” refers to
high definition video. “24P” refers to 24 fps progressive-scan video. “DSLR”
refers to single-lens reflex still cameras with a digital sensor. “NLE” refers to
nonlinear editing.
AspeCt RAtio issues

Theatrical films are usually projected in 4-perf 35mm prints at 24 fps,
either in the matted widescreen format (using a projector mask to crop the
image, commonly to 1.85:1) or in the anamorphic widescreen format (the
printed image is approximately 1.20:1 with a 2X horizontal squeeze, stretched
to nearly 2.40:1 with an anamorphic projector lens). Current standards for
digital cinema presentations follow the same conventions for aspect ratios.
In broadcast video, however, the standard aspect ratios are 4 x 3 (1.33:1) or
16 x 9 (1.78:1). Since most video cameras shoot in one or both formats, with
16 x 9 now becoming dominant, there are framing considerations for any-
thing intended for theatrical release where the 1.85 and 2.40 ratios prevail.
In digital terms, the NTSC picture area is 720 x 480 (or 486) pixels; PAL
is 720 x 576 pixels. An HDTV picture is either 1280 x 720 pixels or 1920 x
1080 pixels.
If you do some simple math, you’ll realize that 720 x 480 equals a 1.50:1
aspect ratio, and 720 x 576 equals 1.25:1—neither is 1.33:1 (4 x 3). To con-
fuse things further, the 16x9 option in NTSC and PAL share the same pixel
dimensions as 4 x 3. The simple explanation is that the pixels in many re-
cording formats are not perfectly square. You can see this right away when
you capture a frame of SD video and display it on a computer monitor as a
still image.
On a 16 x 9 monitor, a 16 x 9 SD recording displays correctly. But when
played on a 4 x 3 monitor, it fills the screen but looks stretched vertically
| 39
because of its nonsquare pixels, unless converted into a 4 x 3 signal with a

1.78 letterbox (a DVD player, for example, can be set-up to automatically
do this but many tape decks cannot). This is why 16 x 9 SD is often called
“anamorphic,” not to be confused with the photographic process of the same
name. HD, having a native 16 x 9 aspect ratio, does not use this anamorphic
technique unless downconverted to 16 x 9 SD.
4 x 3 has given way over time to 16 x 9 as HD has become more com-
monplace in shooting, post, and distribution. Since HD cameras have 16 x 9
sensors, and the costs of these cameras have fallen to the same levels of SD
cameras, there is little reason to deal with 4 x 3 SD camera issues anymore,
even for SD distribution.
In terms of composing 16 x 9 for eventual 35mm 1.85 print projection,
most prosumer cameras do not offer 1.85 framelines for the viewfinder.
However, 16 x 9 is so similar to the 1.85 ratio that simply composing shots
with slightly more headroom will be enough compensation for a later 1.85
crop. Many viewfinders can be set up to display a slightly smaller “safe area”
inside the full frame that could serve as a rough guide for how much to
protect the image. Also, one could point the camera at a 1.85 framing chart
and tape-off any production monitors to match those framelines.
A recording of this framing chart would be useful in post as well.
If a film transfer is planned, it would be better to leave the final color-
corrected master in 16 x 9 full-frame rather than letterbox it to 1.85. The
entire 16 x 9 image would then be transferred to film within the 35mm
sound aperture area (aka “Academy”) with a 1.78 hard matte. The black
borders of the hard matte would be just outside the projected 1.85 area. If
you had transferred a 1.85 letterboxed image to film, then the 1.85 projector
mask would have to be precisely lined up with the hard matte in the print
or else the audience would see some of the matte on screen. A 1.78 hard
matte allows some mild vertical misalignment during 1.85 projection, but
not enough to allow the image to be clearly misframed by the projectionist.
Also, since full-frame 16 x 9 and 1.85 are similar, keep the entire 16 x 9
frame clear of any film equipment (like mics and dolly tracks) when shoot-
ing. Odds are high that you will need to deliver a 16 x 9 full-frame version
for HDTV broadcast, so protecting the entire frame will eliminate the need
for expensive reframing and retouching in post.
Some filmmakers wish to shoot 16 x 9 video for transfer to 35mm 2.40
anamorphic. The most common solution is to simply compose the image
for cropping to 2.40; for transfer to film, either a 2.40 letterboxed version
or the 16 x 9 full-frame master can be used. You would instruct the transfer
facility to crop the recording to 2.40 and transfer to the 35mm anamorphic
(“scope”) format. The cropping and stretching are usually done by the film
recorder. If you submit a 16 x 9 full-frame recording, it is a good idea to add
40
|
a framing leader showing the

1.78 : 1 (16 : 9)
2.40 picture area within 16 x 9.
While it is easiest to just cen-
ter the 2.40 composition verti-
cally within 16 x 9, when the
project is invariably broadcast
in full-frame 16 x 9 on some
HDTV channel, there will be
an excess of headroom in the 1.85 : 1
unmatted version. This may
require some reframing in Figure 1. 1.85:1 format inside a 1.78:1 frame.
post when objectionable—but
then it will be necessary to zoom into the picture slightly so you can raise the
frame, at the loss of side information. A better solution would be to frame the
2.40 image higher up in the 16 x 9 frame, almost to what is called “common
top” so that the headroom is
similar between the full-frame
and letterboxed version. Un-
fortunately, it’s hard in many
cameras to reposition 2.40
viewfinder framelines. Also,
for the final 2.40 letterboxed
2.40 : 1
master, you will have to verti-
16 : 9 cally recenter the 2.40 image
within 16 x 9.
FIgure 2. 2.40:1 format centered inside a 1.78:1 frame.
A less common approach to
achieve 2.40 is to use a 1.33X anamorphic lens attachment on a 16 x 9 pro-
sumer camera. Normally these optical devices are used to squeeze a 16 x 9
image onto a 4 x 3 sensor, but they would also squeeze a 2.40 image onto a
16 x 9 sensor, more or less. Since a 16 x 9 recording with a 1.33X squeeze is
nonstandard, you’ll have to work carefully with the company doing the film
recording to make sure they
properly convert the 1.33X
squeeze to a standard 2X
squeeze. You’d also have to
make an unsqueezed 2.40
letterboxed image for video
distribution. 2.40 : 1
Another possibility is to use

a 35mm groundglass adaptor 16 : 9
that took PL-mount 35mm Figure 3. 1.85:1 format positioned near “common top”
cine lenses; Vantage Film inside a 1.78:1 frame.
| 41
makes a 1.33X version of their Hawk anamorphic lenses that would have
the correct squeeze ratio for a 16x9 sensor camera. And if you were using a
B4-mount 2⁄3" video camera, Canon makes a rear-mounted lens attachment
that adds a 1.33X squeeze.
Some cameras, particularly the large single-sensor ones, have cine lens
mounts thus 35mm anamorphic lenses can be used directly. Keep in mind
that the majority of anamorphic lenses made for cinema use have a 2X squeeze
and were designed to fit a 2.40:1 image onto a 1.20:1 area of 4-perf 35mm film.
Since most HD cameras have a 16 x 9 (1.78:1) sensor, once unsqueezed, the 2X
anamorphic lens would create a 3.56:1 image and thus require as much crop-
ping of the sides to make it 2.40:1 as would be involved in cropping a normal
16 x 9 HD image top and bottom to create the same aspect ratio.
After testing some of these options, you may end up preferring to just crop
a normal spherical image to 2.40:1. The simplest solutions are often the best,
especially in independent feature production.
FRAme RAte issues

The 59.94 Hz rate used for television transmissions in 60 Hz North Ameri-
can countries started with the introduction of the color encoding for analog
NTSC in the mid 1950s. For better or worse, it has remained today in digital
ATSC broadcast rates for North America. In PAL countries, the rate is 50 Hz
both for television and for household power.
In interlaced-scan video, each field contains every other line of video nec-
essary to make up a whole frame; the second field contains the alternate lines.
In a camera with an interlace-scan sensor, each field is captured sequentially
and therefore a moving object appears in a slightly different position on each
field. Thus when two temporally sequential fields are combined into a single
frame, as for a film-out or for display on a progressive-scan monitor, the
edges of the moving object have a “toothcomb” artifact and there is some
loss of vertical resolution.
In a progressive-scan video camera, complete frames are captured by the
sensor, whether or not this information is then stored that way or split into
fields.
In 60 Hz countries, the interlaced-scan rate can be described as both
59.94 fields per second or 29.97 frames per second. It is can be expressed as
“59.94i” (though often rounded to “60i”—which I will be doing often in this
chapter). In 50 Hz countries, the interlaced-scan rate is 50 fields per second
or 25 frames per second. It is expressed as “50i.” If your camera could capture
in progressive-scan, the frame rate would be expressed with a “P” after it, as
in “24P” for 24 fps progressive-scan.
Broadcast HDTV for 60 Hz countries is either 720/60P or 1080/60i. In 50
Hz countries, it is 720/50P or 1080/50i. This means that, to some extent, the
42
|
problems of NTSC and PAL are being carried over into HDTV, at least as
far as the filmmaker attempting to use HD technology to create projects for
transfer to 24 fps film. For example, if using an interlace-scan prosumer HD
camera, it may be better to choose a 50i model over a 60i version since it is
simpler to convert 50 fields into 25 frames than it is to convert 60 fields into
24 frames. 50i footage de-interlaced to 25P can be transferred 1:1 to frames
of 35mm film for projection at 25 or 24 fps. However, with a 24P or 25P
option becoming more prevalent in HD cameras, there is little reason to deal
with conversions from interlace-scan photography anymore.
Since 60i is still used by broadcast video and many display devices, con-
sumer cameras often convert 24P capture into a 60i recording using a “pull-
down” scheme. This is particularly common for cameras that use an internal
VTR for recording. Since some users want to simply edit and display the
24P Standard Pulldown
24P A B C D
original
60i A A B B B C C D D D
video
24P A B C D
recovered
24P Advanced Pulldown
24P A B C D
original
60i A A B B B C C C D D
video
24P A B C D
recovered
Figure 4. Two versions of pulldowns from 24 fps to 60 fps.

| 43
24P material in 60i, while others wish to edit the original progressive-scan
frames, many 24P cameras offer two 60i recording modes: one using the
standard 3:2 or 2:3 pulldown, and the other using an “advanced” pulldown.
This is actually a more visible 2:3:3:2 cadence designed to make it easier to
be removed at the editing stage.
In terms of motion artifacts during playback, the normal 3:2 pulldown
cadence is designed to be as smooth as possible when viewing 24 fps ma-
terial on 60i display devices; the scheme is more effectively “buried” but
therefore also harder to extract in post. The advanced pulldown cadence is
not as smooth but allows for a cleaner extraction of the original 24P frames.
As you can see in the charts above, the original “C” frame is split between
two different video frames when using normal pulldown; this means that in
order to recover this frame, those video frames have to be decompressed and
then recovered 24P frame recompressed back into the original codec. This
can mean that the “C” frame will have suffered some possible degradation
compared to the A, B, and D frames. As you can see in the chart for the
advanced pulldown scheme, every third video frame is simply discarded in
order to recover 24P in editing, and this discarded frame was the only one
where each field came from a different original frame.
Due to its name, a number of people mistakenly believe that the advanced
pulldown is “better”—either more film-like or better-looking. It’s not. Its
only purpose, in fact, is to be removed in post. If your edited 24P project
needs to be converted back to 60i for broadcast applications, you’d then add
a standard pulldown scheme to a separate 60i master.
While some interlaced-scan cameras offer a simulated 24 fps effect, this
is really only intended to create film-like motion for 60i projects, not for
transfer to 24 fps film. However, there are many consumer HD cameras now
with progressive-scan sensors capable of true 24P capture, even if some of
them record 24P to 60i with a pulldown. A number of these cameras are
multi-format, with different frame rate options, particularly the ones using
solid-state recording instead of videotape.
It is recommended that any pulldown be removed so that one can edit
with the original progressive frames. This allows greater flexibility and
quality in post for finishing the project to multiple deliverable formats. A
progressive-scan master is optimal for a transfer to film as well distribution
on DVD, Blu-ray, and the Internet. Otherwise, once you start editing a 24P-
to-60i recording, you will be breaking up the pulldown cadence, making it
harder to remove later.
Since many of these cameras record 24P in different ways, it would be pru-
dent to find out what current editing software handles the particular camera
and recording method you will be employing. Updates to current programs
are constantly being released to accommodate these variations of 24P.
44
|
ReCoRDing FoRmAts
This list is not all-inclusive because some formats and recorders fall out-
side the budget range that this article covers.
DV
This term describes both a recording codec (compression/decompression
algorithm) and a tape format. Most lower-end SD cameras uses the DV25
codec, a 5:1 DCT intraframe compression at a fixed bitrate of 25 Mbit/sec.
Chroma subsampling is 4:1:1 (NTSC) or 4:2:0 (PAL). The recorded video
signal is 60i (NTSC) or 50i (PAL). Mini-DV and DVCAM cassettes are com-
monly used; the picture quality is the same for both tape formats. DVCAM
uses a slightly wider track pitch and faster speed to reduce drop-outs and
increase physical robustness. DVCPRO uses 4:1:1 worldwide but otherwise
is still 25 Mbit/sec. DVCPRO50 is 50 Mbit/sec with 4:2:2 subsampling and
roughly 3.3:1 compression.
HDV
This is an MPEG-2 recording format that uses both intraframe and inter-
frame compression to reduce HD down to 25 Mbit/sec for 1080i, 19 Mbit/
sec for 720P. Having the same bitrate, more or less, as DV, it can therefore use
Mini-DV tapes for recording and the IEEE 1394 (FireWire) interface. The
aspect ratio is 16x9; pixel dimensions are 1280 x 720 (720P) or 1440 x 1080
(1080i). The 720P format supports 60P/30P or 50P/25P recording with a 24P
option; the 1080i format supports 60i or 50i recording with a 30P, 25P, and
24P option. Chroma subsampling is 4:2:0.
“Pro-HD” is JVC’s name for their professional line of MPEG-2/HDV
cameras.
DVCpRo HD
This is an HD codec and tape format developed by Panasonic. It uses DCT
intraframe compression; the bitrate is 100 Mbit/sec when recording to tape.
Chroma subsampling is 4:2:2. Aspect ratio is 16 x 9; the pixel dimensions
of the recorded image are 960 x 720 (720P), 1280 x 1080 (1080/60i), or
1440 x 1080 (1080/50i).
AVCHD
This stands for Advanced Video Coding High Definition. This is an 8-bit
4:2:0 HD format that uses an MPEG-4 codec known as AVC/H.264. In pro-
sumer HD cameras, it is recorded to nontape media such as memory cards,
hard drives, and 8cm DVDs. It supports the same frame rates as HDV and
DV. The bitrate is between 12 to 24 Mbit/sec. It uses long-GOP (group of
pictures) interframe compression. The Panasonic name for their AVCHD
| 45
pro camera line is “AVCCAM”. Sony uses the name “NXCAM” for their
line. AVCHD 2.0 allows bitrates to 28 Mbit/sec for 1080/60P and for 3-D
applications.
AVC-intra
Developed by Panasonic, this variation of MPEG-4/H.264 HD offers
intraframe compression rather than interframe, at 10-bits using at higher
bitrates, either 50 Mb/s (4:2:0) or 100 Mb/s (4:2:2).
XDCAm
Uses different codecs and compression methods (MPEG-2, MPEG-4,
DV25). See above for DV recording specifications. The HD version of XDCAM
uses MPEG-2 compression at bitrates of 18, 25, or 35 Mbit/sec. Recorded pixel
resolution is 1440 x 1080. Chroma sub-sampling is 4:2:0. The Sony XDCAM
EX “HQ” mode allows full-raster 1280 x 720 or 1920 x 1080 4:2:0 recordings at
35 Mbit/sec. “HD 422” mode allows 8-bit 4:2:2 at 50 Mbit/sec.
Canon XF
Similar to XDCAM HD, allows 8-bit 4:2:2 recording using MPEG-2 com-
pression at 50 Mbit/sec, as well as 4:2:0 at 35 or 25 Mbit/sec.
Redcode
This is the name of the wavelet compression (JPEG2000 variant) scheme
used by Red cameras to record RAW sensor data. Different bitrates are
offered.
Dte (Direct to edit)

Some cameras and recorders have the option to store data in the types
of file formats used in nonlinear editing systems so that the footage can be
edited immediately without transcoding. Examples include Apple ProRes
and Avid DNxHD.
BeyonD ViDeotApe
Over the past few years, there has been a shift away from using videotape
recorders inside camcorders. These are some alternative recording devices
and mediums:
optical Disk
SD and HD can be recorded using Sony’s professional XDCAM optical
disk system (similar to Blu-ray), but there are only some lower-end con-
sumer DV cameras with built-in optical disk recorders, using the miniDVD
format.
46
|
solid-state Flash memory Cards and Drives

Camera data can be recorded to different types of flash memory cards
such as SD/SDHC (Secure Digital/High Capacity), CF (Compact Flash),
ExpressCard, Panasonic’s P2, and Sony’s Memory Stick (though Sony has
been moving away from that technology).
Panasonic’s P2 cards are basically a RAID of SD memory cards in a PC
(PCMCIA) card body. Multiple P2 cards can be used in slots in the cameras
to increase recording times. Currently 16, 32, and 64 GB cards are avail-
able but capacity keeps increasing over time. One 16 GB card can record 40
minutes of 24P/720 material in DVCPRO HD.
The Sony XDCAM EX cameras record to SxS cards, a high-speed Express-
Card SSD (solid-state drive) using PCIe interface specification. Current SxS
cards are 16, 32 and 64 GB. A 16 GB card captures 50 minutes of 1080/24P
at 35 Mbit/sec.
The Blackmagic Cinema Camera has an integrated recorder using 2.5"
SSDs. The Red Scarlet has a side-mounted recorder using 1.8" SSDs.
external Recorders
There are portable recorders that can store larger amounts of video or
data than the camera can internally; some of these external devices are small
enough to be mounted directly to the camera. Some use an HDD (hard disk
drive) or a SSD (solid-state drive); others use memory cards for storage.
Some examples:
w Focus Enhancements makes a series of portable DTE recorders under

the product name FireStore. Some units, like the FS100, record DV and
Panasonic’s DVCPRO50 and DVCPRO-HD; others record DV and HDV.
Some are designed for specific camera models. The FS-200 records DV/
HDV to CF cards. The units store footage as QuickTime or MXF files.
w Sony makes a solid-state device (HVR-MRC1K) for attachment to their

tape-based cameras that can record HDV/DVCAM/DV to CF cards.
Their PXU-MS240 mobile HHD storage unit is designed for use with the
XDCAM EX cameras to transfer data from SxS memory cards for nearly
continuous back-up or off-load of video clips in the field.
w Panasonic has a portable P2 recorder/player (AG-HPG20) that records 10-

bit 4:2:2 AVC-Intra as well as DVPRO-HD, DVCPRO50, DVCPRO, and
DV. They also make a handheld AVCCAM recorder/player (AG-HMR10).
w AJA’s Ki Pro recorder can convert any camera’s HD-SDI/HDMI/SD

output to 10-bit Apple ProRes 422 QuickTime files; the footage is then
| 47
stored on either a removable HDD or (optional) SSD storage module, or

on ExpressCards. The smaller Ki Pro Mini records to CF cards.
w Convergent Designs’ Flash XDR and NanoFlash devices can be mounted

on a camera; they use multiple CF card slots to store footage as Quick-
time or MXF files, compressed using MPEG-2 codec. Their Gemini 444
recorder can record uncompressed 10-bit 4:2:2/4:4:4 through single/dual-
link HD-SDI to 1.8" SSDs. The Gemini Raw recorder can record Canon
Raw, ARRIRAW, and CinemaDNG as well as 1080P.
w The Atomos Ninja and Samurai recorders capture HDMI (Ninja) or HD-
SDI (Samurai) output to Apple ProRes on 2.5" HDD/SSDs.
w Fast Forward Video (FFV) offers the Sidekick HD camera-top recorder,

recording Apple ProRes to 2.5" SSDs from HD-SDI or HDMI.
w Sound Device’s PIX 220 and 240 recorders capture HDMI (220) and/or
HD-SDI (240) to Apple ProRes or DNxHD to CF cards or 2.5" SSDs.
w Blackmagic Design’s HyperDeck Shuttle records uncompressed HD-SDI

output into 2.5" SSDs.
w A laptop computer can also be used during shooting to process signals for
recording to an internal or external hard drive.
AVoiDing tHe ReCoRDing CoDeC

Some cameras can output a video signal before it gets processed and com-
pressed by the codec used in the internal recorder. For example, an HDV
camera often has an HDMI-out in order to send a picture to a consumer
HDTV monitor; this is usually an 8-bit 4:2:2 HD signal that could be recorded
using a computer (probably with a RAID array), an HDCAM-SR deck, or a
data recorder. You probably will need to use an HDMI-to-HD-SDI converter
in order to connect with these devices, though some devices accept HDMI
directly. Some prosumer cameras already have an HD-SDI output. Single-link
HD-SDI is limited to 1.5 Gbit/sec; an uncompressed 8-bit 4:2:2 1920 x 1080
59.94i signal is around 1 Gbit/sec. What comes out of these HDMI or HD-SDI
connections is quite variable in quality, depending on the camera itself.
Though recording large amounts of uncompressed data at HD resolu-
tion for a feature-length project (sometimes 20 to 60 hours of footage) is
somewhat inconvenient for many independent filmmakers, there may be
reasons to do that for limited amounts of footage—for example, when doing
chroma key vfx composites, or when mixing the footage with higher-quality
48
|
formats. Again, testing is recommended to see if the results are significantly

better, enough to warrant the difficulty or extra cost in using this approach.
Over the years, it has become easier to manage larger and larger amounts of
camera data, which means that filmmakers will be able to record hours of
uncompressed HD without much effort.
DeptH oF FielD AnD sensoR size

Small prosumer video cameras often have ⅓" sensors or even smaller
(though a recent trend has been towards larger sensors.) This means that, in
order to achieve the same field of view, the focal length used is much shorter
than in 35mm cinematography—with a resulting increase in depth of field.
A ⅓" sensor is 4.8mm wide; a Super-35 film frame is 24mm wide. That’s a
5X difference or conversion factor. Therefore you would need to use a 10mm
lens on a ⅓" sensor camera to approximate the field of view of a 50mm lens
on a 35mm movie camera. Practically-speaking, this means that ⅓" photog-
raphy has about five more stops of depth of field compared to 35mm cine
photography, once you match f-stop, distance, and field of view. The tradi-
tional solution to reduce this major increase in depth of field has been to use
the widest aperture of the lens and then try to work in close-focus or at the
telephoto end of the zoom in order to throw the background into soft focus.
However, in smaller shooting spaces where wider-angle lenses need to be
employed, this solution does not reduce depth of field appreciably.
There are some 35mm lens adaptors on the market that project the lens
image onto a 35mm-sized groundglass screen, which is then rephotographed
by the prosumer camera, thus retaining the field of view and depth of field of
these lenses when used in 35mm photography. Since the groundglass screen
has a surface pattern and texture, the more advanced adaptors either vibrate
or rotate the groundglass while shooting, or use an extremely fine-grained
material for the groundglass. The simplest devices simply leave the texture
of the groundglass screen over the image. Since the image coming off of the
35mm lens is upside down, some adaptors will also flip the image; otherwise,
either the camera itself has to be flipped upside down, or just the viewing
monitors (leaving flipping the recorded image until later in post). A few cam-
eras have an image flip feature. The first of these adaptors to hit the market
was the P+S Technik Mini35, followed by many variations like the Letus35,
MOVIEtube, Redrock M2, Cinemek Guerilla35, Agus35, Brevis35, etc.
In terms of larger sensors that come closer to 35mm in size, the Sony
EX1 and EX3 cameras use ½" sensors (6.97mm wide). Most professional
HD camcorders used for ENG/EFP work have ⅔" sensors (9.58mm wide).
There is only an effective 2.5-stop increase in depth of field with ⅔" photog-
raphy over 35mm cine photography, which is easier to compensate for by
shooting at wider lens apertures—at least compared to ⅓" photography. The
| 49
Blackmagic Cinema Camera uses a sensor that is almost 16mm wide. The
Panasonic AG-FS100 uses a Micro 4/3 sensor, which is about 80% as large as
a 35mm motion picture frame (the sensor is about 18mm wide versus 24mm
wide). Finally, there are now affordable models with 35mm-sized sensors in
them—for example, the Canon C300 and Sony FS700U, NEX-FS100, NEX-
VG10, and PMW-F3 all use a 35mm-sized sensor.
There are also digital still cameras with 35mm-sized sensors that shoot
HD video. Some even have a “full-frame” (FF) 35mm sensor, 36mm x
24mm, which is the same size as the 8-perf 35mm VistaVision frame. With
FF35 cameras, there is an effective 1.5-stop loss in depth of field over 35mm
cine photography once you match field of view, distance, and shooting stop.
Today, the shallower-focus look of 35mm photography is no longer
restricted to expensive professional cameras; however, keep in mind that
with less depth of field comes greater difficulty in follow-focusing during a
scene—and good focus is critical if the goal is projection on a large theatrical
screen, or even on large consumer HDTVs.
sHooting HD ViDeo on A
DigitAl still CAmeRA
Digital still cameras have had limited video capabilities for years now, but
recently, many have been able to shoot progressive-scan HD, either 720P or
1080P. This type of photography has been labeled “V-DSLR”, “HD-DSLR”,
“HDSLR” or “DSLR Video” in various online forums and publications
(though some of the still cameras being used are not actually DSLR’s but are
mirrorless.) Their small size, low cost, and high sensitivity have opened up
new possibilities in cinematography.
some issues to consider:

w Sensor sizes. Many cameras have an APS-C sensor, which varies from
22mm to 24mm in width, roughly the same size as the 3-perf 35mm
cine frame. Because of this, sometimes this sensor size is referred to as
“Super-35” (S35). The Canon EOS 7D and the Canon Rebels fall into this
category, as well as the Nikon D90, D5000, and D300s. The Canon EOS
5D Mark III, Nikon D4 and D800 have a FF35 sensor (full-frame 35mm)
comparable in size to the 8-perf 35mm VistaVision frame (36mm wide).
Less common is the APS-H sensor used by the Canon EOS 1D Mark IV,
which is between the APS-C and FF35 sizes, around 29mm in width.
There is also the Micro 4/3 sensor used by cameras such as the Panasonic
Lumix DMC-GH2, which is around 18mm in width, somewhere between
Super-35 and Super-16 in size. Finally, more and more compact (a.k.a.
“point-and-shoot”) digital still cameras, most with small sensors, can now
shoot HD, though usually 720/30P.
50
|
w Depth of field. The larger FF35 sensor, such as in the Canon 5D Mark III
or Nikon D800, requires longer focal-length lenses be used to match the
same field of view of shorter focal-length lenses on a 35mm cine camera,
a 1.5X conversion factor. This will lead to a shallower-focus look unless
compensated for by stopping down the lens. In practical terms, it’s roughly
a 1.5-stop difference in effective depth of field. The lower noise floor of
the larger sensors help in this regard, giving you the flexibility of rating
the camera at a higher ISO in order to shoot at a deeper stop. The APS-C
sensor (such as in the Canon 7D) allows the same depth of field character-
istics as 35mm cine cameras.
w Lens selection. Due to the flange depth of most PL-mount cine lenses,
the mirror shutter in a DSLR can make it difficult to mount these lenses
without risk of hitting the mirror. An increasingly popular aftermarket
solution has been to remove the mirror shutter and add a PL-mount; still
photography remains possible but would be limited to using the LiveView
function. There are also some cine lenses made with a flange depth that
does not interfere with the mirror function. Another option is to use a
camera without a mirror shutter, such as the Panasonic Lumix DMC-
GH2. Adaptors are made that allow a PL-mount lens to use the still cam-
era’s mount; however, these still camera mounts are not as robust as the
PL-mount. Due to small variations in flange-to-sensor distances between
still camera bodies, a cine lens shimmed to the correct depth for one still
camera may be slightly off on another body, so test your set of lenses if you
plan on sharing them between multiple bodies.
w Lens coverage. PL-mount cine lenses are normally designed to cover a
Super-35 area, though many medium-to-long focal-length lenses will
cover the larger FF35 area. Therefore the problem becomes finding the
shorter PL-mount lenses that will cover FF35 for your wide-angle shots.
However, if your frame of reference is Super-35 photography and you are
trying to decide what short focal-length lenses you need, remember that a
50mm lens on a FF35 camera has approximately the same horizontal view
as a 35mm lens on a Super-35 camera. There are also some PL-mount
lenses on the market designed to cover FF35, such as the Zeiss Compact
Primes. At the moment, however, it is more common to use still camera
lenses on a FF35 camera shooting HD.
w Focusing still camera lenses. On most DSLR’s, the Auto-Focus function
does not work continuously in HD mode, unlike with many consumer
video zooms (there are some exceptions like the Panasonic DMC-GH2
and Nikon D800.) While focusing the lens by hand is possible, many opt
to add a bracket under the camera for holding standard 15mm rods, al-
lowing a follow-focus unit to be attached, as well as a mattebox and vari-
ous motors as needed. The still camera lens itself will need a toothed focus
| 51
ring, either through conversion or attachment. The short barrel rotation

of still camera lenses from minimum focus to infinity is one of the biggest
challenges for focus-pullers, hence the interest in adapting cine optics to
these cameras.
w Recording. Many of these cameras record 8-bit 4:2:0 1080P using
AVC/H.264; others record 8-bit 4:2:0 720P using Motion-JPEG. Many of
these cameras have a per-shot recording time limit of 5 to 12 minutes.
Recently the recording time has been increased to just under 30 minutes
in some models.
w Storage. Footage is recorded internally to CF and/or SD/SDHC cards. You
will need to use an SD card with a higher class speed rating for recording
HD video.
w Frame rates for slow-motion. While most of these cameras can shoot at
24/25/30 fps (or 23.98 and 29.97), 720P is often the highest resolution
possible for 50/60 fps. However, 1080/60P is beginning to appear on the
latest models.
w Monitoring. Most of these cameras have a mini-HDMI connection that
can be fed to an external monitor; however, this usually disables the cam-
era’s LCD, requiring that the operator use an attached monitor which then
is daisy-chained to other monitors. When not connected to an external
monitor, some operators use a focusing loupe attached over the camera’s
back-panel LCD as a form of eyepiece. The Canon 7D always sends a 59.94
Hz interlaced signal through the HDMI port, whereas the monitor output
of the Canon 5D Mark II drops to 480P while the camera is recording (the
Mark III sends out a 720P signal however). The Nikon D90, on the other
hand, allows you to select different formats for the HDMI-out; however
playback is always 720P. While the D90 is recording, the resolution of the
HDMI-out is reduced.
w Bypassing the internal recorder. There has been some interest in trying
to record the HDMI-out from these cameras as a way of bypassing the
highly compressed recording format; however, due to on-screen informa-
tion (such as the red dot that appears in the top corner of the HDMI-out
image during recording, or the white rectangle indicating the area used
for Auto-Focus, or any letterboxing or pillarboxing), the usable picture
information is somewhat limited even when the signal is 1080i. Also, 24P
capture would have a pulldown added for a 1080/60i HDMI-out, and it
may not be the type of pulldown designed for clean extraction in post
back to 24P. The Nikon D800 does allow clean, uncompressed 8-bit 4:2:2
HD to be sent out by HDMI to an external recorder.
w Audio. These cameras lack XLR mic inputs and many only use AGC (Au-
tomatic Gain Control) for setting the audio levels, though some newer
models are adding manual control. Of course, one could record the sound
52
|
on a separate unit and sync it with picture in post—some people are using
small portable recorders like the Zoom H4N and Tascam DR-100 along-
side their DSLR. In the meanwhile, the biggest challenge for those wishing
to record sound on a camera without manual audio controls has been to
defeat the AGC. There are various devices on the market that do this by
sending a tone to one of the two stereo channels, keeping the noise floor
consistent during moments of quiet. For example, the Beachtek DXA-SLR
and the JuicedLink DT454 have an AGC disabler, as well as XLR mic in-
puts, pre-amps and phantom power. Over time, more of these cameras
will allow manual audio control but you may still need attached devices to
allow XLR mic inputs and provide phantom power.
w Picture quality. These cameras do not offer a full-sensor RAW recording
capability as of yet, because of limitations in processing power and data
recording capacity needed to handle such large files (these cameras are
primarily designed for still photography after all.) The Canon 5D Mark
III, for example, has a 22MP sensor; a single frame of 1080P video, on the
other hand, is only 2MP. To get around this problem, most of these cameras
employ coarse downsampling techniques to reduce the amount of sensor
data to be processed into HD video. This can reduce effective resolution and
create aliasing and chroma moiré artifacts. Also, the reset rate of the rolling
shutter creates a distortion called “skewing” during fast motion.
w Postproduction. H.264 and the other similar compression schemes, plus
the color and luminance limitations of an 8-bit 4:2:0 recording, can have
an impact on flexibility in color-correction. Also, the contrast of the DSLR
video image was really designed for immediate viewing on a monitor,
making it somewhat high for material to be color-corrected. The common
solution has been to set the contrast level as low as possible through the
camera’s menu; some people are also loading custom gamma curves.
Future improvements in camera processing should reduce many of the ar-

tifacts, and more and more post solutions are being developed to maximize
the quality of the footage. For now, however, the best solution is to shoot the
best image you can in-camera, working within the more limited dynamic
range of the recording and avoiding situations where aliasing and skewing
can become a distraction. Also keep in mind that some of these still cam-
eras do not have exposure aids common to video cameras such as zebras or
histograms while rolling, nor focusing aids such expanded view or peaking.
A final note: some of the after-market alterations we are talking about,
such as removing the mirror shutter and adding a PL-mount, plus buying a
set of PL-mount lenses, not to mention all the desired accessories (mattebox,
lens support, follow-focus unit, focusing loupe, handheld rig, remote focus-
ing device, onboard HD monitor, etc.) may drive an initial investment of a
| 53
few thousand dollars beyond the budgetary limit of $15,000 that this article
is focused on.
BeyonD DslR ViDeo

The popularity of using still cameras to shoot HD video has led manufac-
turers to start making video cameras with the larger sensors. One challenge
for manufacturers is to create ENG-style electronic zoom lenses that cover
the larger sensor while maintaining affordability for the consumer (not
to mention, avoiding making physically larger lenses). Of course, adding
standard video camera features and lenses generally means that the cost of
ownership is going to be higher than with still cameras that have an HD
mode; however, there may be a need for fewer accessories as a result of buy-
ing a product specifically made to shoot video.
A seleCtion oF CAmeRAs
The breakdown is based around average retail prices for the body only.
Please note that some of these cameras will need more accessories than
others to become production-friendly, and some do not come with lenses
at their base price. All of these factors can add dramatically to the cost of
ownership. Keep in mind that it is possible to find deals on many of these
cameras. Also, this is only a selection of what is available; new products are
constantly being released, and existing products are constantly being up-
dated with new features.
A few cameras allow a significantly better image, with attendant higher
data rates, to be recorded using external recorders. However, some of these
recorders are more expensive than the cameras mentioned.
High-end (in the $8,000-$15,000 range, or slightly higher)

w Canon C300. Single 35mm (S35) CMOS sensor. Records 8-bit 4:2:2
1080P using Canon XF codec at 50 Mbps to CF cards; Canon Log gamma
option.
w Canon EOS 1D C. Still camera (DSLR) with FF35 18MP CMOS sensor.
Capable of outputting 8-bit 4:2:2 Motion JPEG 4K video to a CF card at
24 fps (4K uses an APS-H crop of the sensor). Also records 8-bit 4:2:0
1080P video at either 24P or 60P using full-frame of the sensor, with
option to use a S35 crop. Uncompressed 8-bit 4:2:2 signal output via the
HDMI-out.
w Canon XF305. ⅓" 3-CMOS, 2.1MP full-res 1080P sensors. Records 8-bit
4:2:2 MPEG-2 (50 Mb/sec) to CF cards. Comes with Canon 4.1–738mm
zoom.
54
|
w Ikonoskop A-Cam dII. 16mm CCD 1080P sensor. Sends uncompressed

RAW in CinemaDNG format to integrated SSD recorder.
w Panasonic AG-HPX370. ⅓" 3-CMOS, 2.1MP full-res 1080P sensors. Re-

cords 10-bit 4:2:2 AVC-Intra (1080i, 1080P, 720P), plus 8-bit DVCPRO-
HD, DVCPRO50, DVCPRO, and DV. 12–60 fps. P2 card recording (two
slots). Comes with Fujinon 4.5-76.5mm zoom.
w Red Scarlet. 35mm (S35) CMOS sensor, 4K. Modular design; components
sold individually. Records 4K RAW using Redcode compression to SSDs.
w Sony NEX-FS700U. 35mm (S35) CMOS sensor, 4K. Records 1080P using
AVCHD to SD card/MemoryStick or via the FMU (flash memory unit)
port, or it can output 8-bit 4:2:2 (with embedded timecode) via HDMI 1.4
or 3G/HD-SDI to an external recorder. High frame rates in short bursts
(8 to 16 seconds) up to 240 fps at full resolution or 960 fps at reduced
resolution. Future option allowing 4K output to external recorder.
w Sony XDCAM PMW-EX3. ½" 3-CMOS, full-res 1080P sensors. Records

8-bit 4:2:0 MPEG-2 (XDCAM, up to 35 Mbit/sec) & DVCAM, to SxS
ExpressCards (two slots). 1–60 fps (720P); 1–30 fps (1080P). Time-lapse
capability. Uses ½" EX Mount.
w Sony XDCAM PMW-F3. 35mm (S35) CMOS sensor. Records XDCAM

HD to SxS cards. Optional 10-bit 4:4:4 S-Log out to external recorder.
mid-range (in the $3000-8,000 range)

w Blackmagic Cinema Camera. Single CMOS sensor that is 15.1mm x
8.08mm. Records uncompressed 12-bit 2.5K RAW data (CinemaDNG)
or compressed 1080P HD (ProRes or DNxHD) to 2.5" SSD’s. EF & ZE
lens mount.
w Canon EOS 5D Mark III. Still camera (DSLR) with a single FF35 22MP
CMOS sensor. Records to CF or SD cards.
w Canon XF105. Single ⅓" CMOS sensor. Records 8-bit 4:2:2 MPEG2 (50
Mbit/sec) to CF cards.
w JVC GY-HM750U. ⅓" 3-CCD sensors. Records 8-bit 4:2:0 MPEG2 (1080P,
1080i, 720P), up to 35 Mbit/sec, to SDHC cards (two slots). Optional SxS
card adaptor. 1–30 fps (1080P/i) or 1–60 fps (720P). Comes with detach-
able Canon f/1.6 4.4–61.6mm zoom.
| 55
w JVC GY-HM250U. ⅓" 3-CCD sensors. Records 8-bit 4:2:0 MPEG2 (720P,
480P/i), up to 35 Mbit/sec, to HDV tape. Optional SxS card adaptor. 1–30
fps (1080P/i) or 1–60 fps (720P). Comes with detachable Canon f/1.6
4.4–61.6mm zoom.
w JVC GY-HM100U. ¼" 3-CCD sensors. Records 8-bit 4:2:0 MPEG2

(1080P, 1080i, 720P), up to 35 Mbit/sec, to SDHC cards (two slots).
1–30 fps (1080P/i) or 1–60 fps (720P). Fixed Fujinon f/1.8 3.7–37mm
zoom.
w Nikon D800. Still camera (DSLR) with FF35 36MP CMOS sensor. Re-
cords to CF or SD cards. Has clean 8-bit 4:2:2 HDMI-out for recording to
external device.
w Panasonic AG-AC160. ⅓" 3-CMOS sensors. Records 1080P, 1080i, 720P

using SD cards to MPEG-4 AVC/H.264. Switchable between 50Hz and
60Hz at variable frame rates.
w Panasonic AG-AF100. Single Micro 4/3 CMOS sensor. Shoots 12–60 fps;
records 1080P, 1080i, and 720P using SD cards to MPEG-4 AVC/H.264.
Has HD-SDI out. Micro 4/3 lens mount.
w Panasonic AG-HMC150P. ⅓" 3-CCD sensors. Records 1080P, 1080i,

720P using SD cards to MPEG-4 AVC/H.264 (up to 24 Mbit/sec). Fixed
f/1.6 3.9–51mm zoom.
w Panasonic AG-HPX250PJ. ⅓" 3-CMOS sensors. Allows 10-bit 4:2:2 HD

recording using AVC-Intra 100 format to P2 cards. Also records in AVC
Intra 50, DVPRO HD, as well as standard def DV formats.
w Sony HDR-AX2000. ⅓" 3-CMOS 720P sensor, records 1080/24P to 60i

using Sony Memory Stick or SD cards. MPEG-4 AVC/H.264 (up to 24
Mbit/sec). Fixed f/1.6 4.1–82mm zoom.
w Sony NEX-FS100U. Single APS-C 35mm CMOS sensor. Shoots 1–60

fps at 1080P. Records using Sony Memory Stick or SD cards to MPEG-4
AVC/H.264. Uses Sony E-mount lenses.
w Sony NXCAM HXR-NX5U. ⅓" 3-CMOS. Records 8-bit 4:2:0 ACVHD to

Sony Memory Stick PRO Duo (two slots) or SDHC cards. Optional attach-
able 128GB flash memory unit (HXR-FMU128). Fixed Sony 4.1–82mm
zoom.
56
|
w Sony XDCAM PMW-EX1R. ½" 3-CMOS, full-res 1080P sensors. Records

8-bit 4:2:0 MPEG-2 (XDCAM, up to 35 Mbit/sec) & DVCAM, to SxS
ExpressCards (two slots). 1–60 fps (720P); 1–30 fps (1080P). Time-lapse
capability. Fixed Fujinon f/1.9 5.8–81.2mm zoom.
low-end (under $3000)

w Canon EOS 7D. Still camera (DSLR) with single 35mm (APS-C) 18MP
CMOS sensor. Records 1080P/720P/480P using 8-bit 4:2:0 H.264 to CF cards.
w Canon Vixia HF S21. Single ½.6" CMOS (3264 x 1840 pixels) sensor.
Records AVCHD to SD cards (two slots) or internal 64GB flash drive.
24P/30P/60i. Fixed f/1.8 6.4–64mm zoom.
w Canon XA10. Single ⅓" CMOS sensor. Records to internal 64GB flash
drive in 8-bit 4:2:0 MPEG-4-AVC H.264 (up to 17 Mbits/sec) Fixed f/1.8
4.25–42.5mm zoom.
w Panasonic AVCCAM AG-HMC40. ¼" 3-CMOS. Records 8-bit 4:2:0

AVCHD to SD/SDHC cards (two slots ).
w Panasonic Lumix DMC-GH2. Still camera (mirrorless) with single Micro

4/3 16MP CMOS sensor. Records AVCHD to SD/SDHC cards.
w Sony NEX-VG10. Single APS-C 35mm sensor. Using Sony Memory Stick
or SD cards, records 30P to 1080/60i using MPEG-4-AVC H.264 (up to 24
Mbits/sec). No 24P. Sony E-mount lenses.
The author would like to thank Adam Wilt, Randy Wedick, Charles Crawford,
and Phil Rhodes for their technical advice.
M. David Mullen, ASC is a member of the Academy of Motion Picture Arts and
Sciences. He has photographed more than thirty independent feature films and two
television series. He has received two IFP Independent Spirit Award nominations
for Best Cinematography, the first for Twin Falls Idaho (2000) and the second for
Northfork (2004).
| 57
Putting the Image on Film

by Rob Hummel,
T he section on Exposure, together with the associated tables, is intended

as a quick-reference condensation of material explained in more detail
elsewhere in the manual. The section on Special Techniques contains addi-
tional information on many other photographic situations.
ExPosurE
Most exposure meters incorporate some sort of calculator—some simple,
some sophisticated. An exposure meter measures light, either incident or
reflected. The calculator helps you decide how to use the light. There are six
specific variables entering the calculation:
Variables Expressed as:

Film exposure index EI, ASA/ISO
Camera Speed FPS (frames per second)
Shutter Opening Degrees
Lens Aperture T-stop/F-stop
Filter Filter factor
Light Meter reading: Footcandles, Footlamberts
T-sToPs
The “T” stop number is defined as being the true “f ” stop number of a
lens if it is completely free from all reflection and absorption losses. The
T (transmission) number represents the f-stop number of an open circular
hole or of a perfect lens having 100-percent axial transmission. The T-stop
can be considered as the “effective” f-stop. It is from this concept that the
means arises for standardization of T-stop calibration. T-stops are calibrated
by measuring the light intensity electronically at the focal plane, whereas
f-stops are calculated mathematically, purely on the basis of dividing the
focal length of the lens by the diameter of the effective aperture (entrance
pupil). Thus, f-stops are based on the light that enters a lens, while T-stops
are based on the intensity of the light that emerges from the rear of the lens
and forms the image.
There is no fixed ratio, however, between T-stops and f-stops that applies to
all lenses. The difference actually represents light losses within the elements
of a given lens due to reflection from the glass-air surfaces and from absorp-
58
|
Putting the Image on Film
tion within the glass itself. Consequently, this factor is variable and cannot
be incorporated into an exposure meter, since the meter must function in
connection with many different lenses calibrated in both f-stops and T-stops.
The reason why lens and exposure tables are presented in f-stops when
all professional cine lenses are calibrated in T-stops, is that the f-stops are
required for all calculations involving object-image relationships, such as
depth of field, extreme close-up work with extension tubes, etc. Such tables
are based on the size of the “hole” or diameter of the bundle of light rays that
the lens admits to form the image. The diameter of the f-stop will normally
be the same for all lenses of similar focal length set at the same aperture. The
T-stop, however, is an arbitrary number that may result in the same T-stop
setting varying in aperture diameter with different lenses.
It is recommended that all professional cine lenses be calibrated in both
T-stops and f-stops, particularly for color work. T-stop calibration is es-
pecially important with zoom lenses, the highly complex optical design of
which necessitates a far greater number of optical elements than is required
in conventional lenses. A considerable light loss is encountered due to the
large number of reflective optical surfaces and absorption losses. A zoom
lens with a geometrical rating of f/2, for example, will transmit considerably
less light than a conventional fixed-focal-length lens of similar rating with
fewer elements.
Exposure tables are generally based on “effective” f-stops (which are, in
fact, T-stops). Small variations in emulsion speed, processing, exposure
readings, etc., tend to cancel out. Cinematographers should shoot tests with
their particular lenses, meter, light and film to find the best combinations for
optimum results.
Other variables such as direction and contrast of the light are factors cal-
culated from the experience of the cinematographer, aided by such tools as
photospheres and spot readings. Finally, manipulation of all the above, plus
special negative processing to achieve a desired “look,” is determined by the
cinematographer.
The laboratory and choice of film are closely tied to exposure. It is impor-
tant to keep exposure within limits satisfactory both to the selected film and
to the printing range of the laboratory.
The tables on pages 812, 813 and 866 through 877 will aid exposure cal-
culation for meters that lack settings for some of the factors, or will aid in
calculating constant exposure control when one factor varies from another.
| 59
Comparisons of 35mm,
1.85, Anamorphic,
Super 35 Film Formats
by Rob Hummel
Editor of the 8th edition
of the American Cinematographer Manual
INTRODUCTION
T his chapter has always addressed creative aesthetic and technical differ-
ences depending on the choice of 35mm film format. The most salient
change from earlier versions of this chapter is the impact of digital interme-
diate (DI) on final image quality.
While a DI can improve upon image quality that might degrade when cre-
ated at a film lab, it does so at greater expense. On many films, that expense
is trivial when compared to the overall cost of the film; on others, it may be
a significant portion of the budget. In 2011, John Bailey, ASC saved a low-
budget film substantially and was able to achieve everything desired from
the image with a photochemical answer print. It is becoming increasingly
rare to color correct a film in this manner, but it is still available.
Also, 2K DIs have made formats such as Super 35 and anamorphic ap-
pear almost equal in quality. Filmmakers often assume this is because the DI
process makes the Super 35 image look so much better. While a 2K DI does
improve the image quality of a Super 35 image, the only reason the image
looks competitive with an anamorphic DI, is because a 2K DI sacrifices the
image quality contained in the anamorphic frame. With the further rollout
of 4K scanning and 4K DI workflows, the superior image quality of ana-
morphic imaging over Super 35 is now readily apparent again. 4K scanning
begins to approach capturing all the resolution that the anamorphic image
has to offer. While Super 35 benefits from 4K scanning, the improvements
are not as dramatic.
Lastly, the focus of this piece has always been about image quality. That
being said, I recognize, more than most, that we’re not in the medical imag-
ing business, we’re in the motion picture business. When I reference grain,
or noise in an image, I recognize that image quality may actually be what you
desire in your picture. Where this chapter may talk about one format having
superior image quality to another, you shouldn’t take offense if the lower
quality image is the preferred method for telling your story.
60
|
Comparisons of 35mm, 1.85, Anamorphic, Super 35 Film Formats
What’s New
◗ A new section on how to evaluate images effectively, and discusses the
importance of screen heights, complete with illustrations. (Page 60)
◗ For motion pictures that will enjoy an IMAX release, you will find a
page briefly illustrating how that process works. (See Figure 11)
◗ If you are considering digital cameras instead of film, after the discus-
sion of film formats and aspect ratios, you will find a new section with
guidelines and information to help you understand what to look for
when examining a digital camera. This should allow you to make an
informed decision about what will most benefit your motion picture;
whether comparing various digital cameras or comparing digital to
film(see page 81).
Screen Heights are Critical!

When evaluating digitally captured or film originated imagery, it is im-
portant to examine the images at a proper distance from the screen, and on
a screen of sufficient size. The distance one views an image is measured by
multiples of the height of the image, or screen heights.
Using screen heights as a measurement of distance from the screen is all
about quantifying the magnification of the image you are viewing. The closer
you are to the screen, the greater the magnification, and the more demands
are placed on the quality of the projected image. For example, the more you
magnify an image in a printed magazine, the more you will start to discern
the dots that make up the image; the further you hold the magazine away
from you, the less likely you can see those dots. By way of example, you
probably want to ensure you are evaluating the images you photograph from
the same distance your audience will be viewing them. If you evaluate the
image from too far away, you can be fooled into thinking the image is of
better quality than it is.
In the case of today’s state-of-the-art stadium seat theaters, that should be
somewhere from 1 to 1½ screen heights from the screen. Most Stadium seat
S
c
r
e
e
n
H
e
i
g
h
t
Two Screen Three Screen

One Screen Height Heights Heights
Figure 1: A typical stadium seat theater design, illustrating how many screen heights the
audience is from the image.
| 61
Figure 2: As an example of how an image can improve with distance, hold the above image
at arm’s length and it will probably look just fine. Upon closer examination, you will notice
the difference between the right and left sides of the picture.
theaters are no more than 3 to 3½ screen heights deep (see Figure 1).
When you are closer than seven screen heights away from your normal
standard definition television set, you can resolve the pixels and scanning
lines that make up the image. However, further than seven screen heights
away, and it is impossible to resolve any of the picture elements that make up
the image; basically, the image will appear as sharp as your eyes can resolve.
For example, if you were looking at an IMAX image and a standard defini-
tion TV image side by side, but you were evaluating the images from 7 screen
heights away, both images would appear to have equal resolution. Once you
got closer than 7 to 6 screen heights, the TV image would begin to exhibit its
poor image quality, while the IMAX image would look exemplary at closer
than ½ a screen height. If you have been in an IMAX theater, take notice
that the back row of the theater is rarely much more than one screen height
away from the screen; something IMAX can do because of the dramatically
high resolution of IMAX photography.
At the risk of stating the obvious; the higher resolution the image, the
closer you can get to the image before noticing pixels (in the case of digital)
or grain (in the case of film).
In the case of HDTV, you have to get closer than three screen heights
before you can start to see the pixels in the image. In current stadium seat
theaters, our audiences are sitting no further than three screen heights from
the images, and, in most cases, closer than two screen heights. For this rea-
son, it’s important we evaluate imagery from the same screen distance as our
audience.
In the image above, your proximity to the image affects your perception
of the image quality.
Most studios still have screening rooms that place you anywhere from six
to eight screen heights from the image. Therefore, when sitting that far from
the screen, you can fool yourself into thinking the image quality is adequate,
perhaps even superb. Yet, when viewed from a distance of 1½ to 2 screen
heights, your conclusions may be entirely different. Please make sure when
62
|
evaluating image quality, you place yourself within 2 screen heights of the
projected image. It is important that critical evaluation of image quality be
viewed from where the audience will see the picture.
Also, the display medium must be taken into consideration as well. Many
people evaluate high-definition (HD) or 2K images on HD displays that can
resolve only 1400 of the 1920 horizontal pixels contained in an HD image.
Those 520 missing pixels of resolution can hide artifacts that will show up
clearly on a digital cinema projector, or a 35mm film out.
FIlm FORmATS
History and a Definition of Terms
Currently, in the United States (and most of the world), the most prevalent
motion picture film formats, or aspect ratios, are 1.85 and 2.40 (2.40 is still
often referred to as 2.35).1 As a point of reference, these ratios are determined
by dividing the width of the picture by the height, which is why you will also
see them written as 1.85:1 or 2.40:1. Verbally, you will hear them referred to
as “one eight five” or “two four oh.” 2.40 is also referred to as anamorphic, or
“Scope,” referring to its roots in CinemaScope.
An examination of films over the past sixty years shows that format is
not automatically dictated by dramatic content. It is a creative choice, deter-
mined by the cinematographer and director. The full range of drama, com-
edy, romance, action, or science fiction can be found in both aspect ratios.
The purpose here is to advise on the pros and cons of both aspect ratios and
the photographic alternatives available to achieve them. This should help a
filmmaker make an informed decision as to which format is best for a given
project. Most importantly, you will be presented with the “conventional wis-
dom” arguments for and against the formats; this conventional wisdom will
either be endorsed or countered with reality. This knowledge, in the end, will
help you realize that, creatively, there are no technical obstacles to choosing
any format. However, you will also understand the aesthetic impact those
choices will have upon your production.
1. A historical note regarding 2.35 vs. 2.40 vs. 2.39. CinemaScope films with an ana-
log soundtrack were originally an aspect ratio of 2.35:1. In the early 1970s, the height
of the anamorphic aspect ratio was modified slightly by SMPTE to help hide splices,
effectively changing the ratio to 2.40:1. Old habits die hard and many still refer to
the format as 2.35:1. Also, in 1995, SMPTE again made a change in the size of the
CinemaScope projection area to accommodate digital soundtracks. In both cases the
math of the aspect ratio yields an aspect ratio number of 2.39 and continues on sev-
eral places to the right of the decimal point. Cinematographers felt that rounding up
to 2.40 ensured there would be less confusion with the 2.35 aspect ratio. The actual
difference between 2.39 and 2.40 is so inconsequential that, from a compositional
standpoint, they are the same.
| 63
As a clarification, the term full aperture refers to the entire image area be-
tween the 35mm perforations, including the area normally reserved for the
soundtrack. In other literature you will find full aperture used interchange-
ably with camera aperture, silent aperture and full silent aperture. All four
terms define the same area of the 35mm film frame.
In general, the term Academy aperture is used when referring to the imag-
ing area of the negative excluding the analog soundtrack area. More prop-
erly, it would be referred to as the “sound aperture,” the term used to indicate
the area that remained when analog soundtracks were first added to 35mm
film. However, throughout this chapter, we will follow convention, and use
the term Academy aperture when referring to the imaging area excluding
the soundtrack.
Academy aperture2 is an aspect ratio of 1.37:1, centered within the sound
aperture area, arrived at jointly by the American Society of Cinematogra-
phers and Academy of Motion Picture Arts and Sciences in the early days of
sound to restore compositions closer to the 1.33 aspect ratio of silent films,
and resolve the unpleasant composition produced by framing images within
the narrow sound aperture.
All 1.85 composed films are achieved with “normal,” spherical lenses.
However, the 2.40 aspect ratio can be achieved two ways. One method is
with the use of anamorphic3 lenses that squeeze the image to fit within the
Academy aperture (see Figure 6). The alternate method (see Super 35 in
Figures 7, 8, and 9) uses normal lenses without any distortion of the image,
and then is later squeezed for theatrical film release. Both methods will be
discussed here.
The 1.85 and Super 35 formats can also be captured using cameras using
a 3-perf pull-down movement. While all 35mm illustrations in this chapter
use a 4 perforation frame, were you to use a 3-perf camera for 1.85 or Super
35, the advantages and disadvantages remain the same whether 3 or 4 perf.
3-perf camera movements do not provide adequate height for anamor-
phic photography.
Also, the film formats discussed here deal with general 35mm motion
picture photography. Formats such as VistaVision and 65mm are most often
2. If you actually calculate the aspect ratio of Academy aperture from SMPTE specs,
the math has always come out to 1.37:1. More often than not, people will refer to
Academy aperture compositions as 1.33 (the original aspect ratio of silent films), and
almost never is the ratio called 1.37. The compositional difference between the two
is negligible.
3. Anamorphic comes from the word anamorphosis, meaning an image that appears
distorted, yet under the proper conditions will appear normal again. Leonardo da
Vinci is credited with first using the technique during the Renaissance.
64
|
used for visual effects and special event cinematography and would require
an chapter of their own to discuss. However, Figures 10 and 11 illustrate how
widescreen compositions are presented when converted to IMAX, and how
films photographed in 65mm are released in 70mm.
At the end of this chapter, we’ll cover current methods for achieving 1.85
and 2.40 aspect ratios with currently available digital cameras.
COmPOSITION
Before getting into specifics
about the different formats,
I want to point out the com-
position differences between
the two aspect ratios 2.40 and
1.85, regardless of how they
are achieved photographically.
Figure 3 is an image of the
Taj Mahal with a 2.40 aspect
ratio outlined in yellow. Figure 3: The yellow rectangle within the image
In Figure 4, two 1.85 aspect outlines a 2.40:1 aspect ratio.
ratios are outlined by yellow
rectangles. The larger of those
two rectangles represents a
1.85 composition equal in its
width to the 2.40 aspect ratio
of Figure 3. The smaller 1.85
rectangle is equal in height to
the 2.40 ratio of Figure 3.
The purpose here, is to il-
lustrate that, depending on
your framing, a 1.85 image Figure 4: Two 1.85 compositions.
has potential of encompass-
ing as much width as a 2.40 composition. Although 1.85 will take in the
same width with greater height in the composition, it’s important to realize
that wide sets and vistas are not restricted to the 2.40 format.
PROS AND CONS – A ClARIFICATION

While this chapter addresses the pros and cons of these formats, it can ap-
pear contradictory upon reading when a disadvantage is countered with an
opposite perceived advantage, and vice versa. This is because, in fact, many
pros and cons are merely mythology about a certain shooting format, and
are based on nothing other than conventional wisdom, or outdated produc-
tion practices, rather than current practical experience.
| 65
I believe you will be able to sort the facts from opinions. That being said,
nothing here is intended to dissuade you from choosing a given film for-
mat. Where there are possible true disadvantages to a format, it is presented
so you have your eyes wide open to any potential challenges or creative
compromises. We are working in a very artistic medium where there can
truly be no absolute rights and wrongs. What one person may find an ob-
jectionable grainy image, another filmmaker may feel enhances the story
being told.
Where possible, when listing outdated opinions, I immediately counter
with the present fact, or contrary point of view. With this presentation, I
want you to be able to draw your own conclusions and inform you with
all the arguments and points of view about these formats. If you study it
carefully, you will be able to effectively articulate your desire to photograph
a film in the aspect ratio you feel will most benefit the story.
I. THe 1.85 ASPeCT RATIO

Photographed in Normal Academy Aperture Photography
Until the early 1990s, 1.85 was far and away the most common aspect
ratio for motion pictures filmed in the United States since the late 1950s.
This trend shifted a bit in the 1990s with the advent of more and more films
composed for the 2.40 aspect ratio. With the proliferation of films photo-
graphed in Super 35, in recent years, less then one third of the top 100 mo-
tion pictures are composed in 1.85. Around the world the spherical aspect
ratio most commonly used swung between 1.85 and 1.66 depending on the
country; however, in many countries that used to photograph exclusively in
the 1.66 aspect ratio, 1.85 has become more common.
Figure 5 portrays how 1.85 films get from camera to screen. In the il-
lustration, the red box indicates the 1.85 composition area of the frame. In
actual practice, there is no indication on the exposed negative of the area
composed for. The area above and below that area contains picture informa-
tion, but is masked out when projected or eliminated entirely in the digital
intermediate.
A. ADVANTAGeS OF 1.85
1. Intimate Format: Many perceive 1.85 as more appropriate for pictures
that lend themselves to a more compact visual. Since close-ups virtually
fill the entire frame, it is often considered a more “intimate” format.
However, it can be argued that many “intimate” films have been com-
posed for the 2.40 aspect ratio.
2. Interiors: If a film is largely interiors, 1.85 is often argued as the pre-
ferred format, since they don’t involve the wide panoramic vistas associ-
ated with 2.40. Conversely, many do not weigh interiors or exteriors in
their choice of format.
66
|
Figure 5: Standard Academy 1.85:1

The image at left is a 1.85:1 aspect ratio, and is
how that image might appear when composed in
the viewfinder of a camera.
The workflow described here is a traditional film
laboratory process, that can still be implemented.
In a digital intermediate workflow, usually multiple
printing negatives are recorded out, eliminating the
IP and IN process.
Negative: Unless a hard matte is used in the

camera, the entire frame is exposed with image
information; including the soundtrack area to the
left of the image. The 1.85 aspect ratio is achieved
by masking off information above and below the
composition. The red boundary delineating the 1.85
composition here are for illustration purposes only,
and do not appear on the negative.
Interpositive (IP): To produce a release print,

the standard 1.85 negative is first contact printed
to a color interpositive The interpositive is similar
in appearance to a color negative, yet contains a
positive image.
Internegative (IN): The interpositive is then con-

tact printed to a color internegative. This produces
the dupe negative that enables mass release print
manufacture without risking the original negative.
Release Print: The internegative is then contact

printed to color positive print stock to produce the
release print. When projected in a theater, the
projector will mask out the unwanted image area
above and below the 1.85 composition. If the film
is not framed properly in the theater, unwanted
information can be shown accidentally. Because of
this, many filmmakers will use either a 1.66:1 hard
matte in the camera, or have the lab build a hard
matte into the printing internegative. In this illustra-
tion, we present a print that has been hard matted
at the lab or DI facility.The Dolby d igital tracks
are between the track side perforations; SDDS is
outside the perfs; DTS is between the analog track
© 2012 Robert C. Hummel and the picture area.
| 67
3. Depth of Field: Greater depth of field (the total area in focus at a given
distance). Since 1.85 uses shorter focal length lenses as compared with
Anamorphic, greater depth of field is available.
This advantage is often negated when lenses are shot “wide open,”
resulting in a little or no gain in depth of field.
4. Composition: Considered by many to be a “taller” composition. Lend-
ing itself to compositions with more emphasis on the vertical than
horizontal. Cathedral interiors, or city skylines for example.
5. Wide Sets: An opinion often expressed is that sets don’t need to be as
wide on a 1.85 film as one photographed in 2.40, resulting in savings in
set construction. However, there are many that would argue film format
has no bearing on the width of set construction. As the examples in
Figures 3 and 4 (page 64) point out, it’s possible for 1.85 to require as
wide a set as 2.40, depending on the composition.
6. Least Complex: 1.85 is the simplest format to execute from a mechani-
cal/technical standpoint. The choice of photographic equipment is vir-
tually unlimited, as any standard 35mm camera will accommodate this
format. Plus, you have a choice of very wide and “fisheye” lenses not
available in anamorphic.
7. Expendable Cameras: If a stunt camera mount is required that risks
destroying a camera lens, spherical lenses can be used that are much
more affordable than expensive anamorphic lenses.
8. Video Transfers: While they are effectively not manufactured any
more, many production companies, are still concerned about accom-
modating the 1.33 (4 x 3) aspect ratio of Standard Definition television.
With some effort on the shooting company’s part, the 1.85 composition
can protect for standard definition video so that a simple one-to-one
transfer can be done without panning and scanning. While left and
right image integrity remain virtually intact this way, there is an ap-
proximate 33% increase in the vertical height of the composition.
Although many think it routine to protect the TV area from intruding
objects (e.g., lights, microphones, etc.), it makes the cinematographer
and soundman’s job more difficult, by not being able to bring lights and
microphones down close to the area of composition. This is why many
cinematographers shooting 1.85 will request to shoot with a 1.66:1 as-
pect ratio hard matte. While the same width on the film, 1.66 is slightly
taller than 1.85, closely approximating the height of the 1.33 (4 x 3) TV
frame. This gives the cameraman more freedom to light his subjects
without fear of a light or microphone showing up when transferred to
video.
Yet, in a world where 1.78:1 (16 x 9) aspect ratio video displays are
now the norm, 1.85, for all intents and purposes, drops precisely into
68
|
the HDTV frame. While not precisely the same aspect ratio, the 1.78:1
HDTV frame is only 42 pixels taller than the 1.85 aspect ratio. Meaning,
if you did letterbox a 1.85 image in HDTV, you would have two 21-pixel
lines above and below the image.
9. Sharper Lenses: Many people believe it is an advantage to shoot 1.85
because spherical lenses are sharper than anamorphic 2.40’s lenses.
This is a misconception. It is true that spherical lenses are often sharper
than anamorphic; however, the much greater negative area used with
anamorphic more than makes up for the subtle difference in resolution
from spherical lenses. Properly executed camera tests comparing the
two formats always reach this conclusion.
B. DISADVANTAGeS OF 1.85
1. Negative Area: The principal disadvantage is the actual size of the 1.85
format on the negative. Because of the smaller area, 1.85 is noticeably
grainier than anamorphic 2.40. This is not as noticeable in the original
negative stage and projecting in small screening rooms, but becomes
more pronounced when projected in large theaters.
Graininess can be mitigated with digital grain reduction techniques
in the DI; however, if not applied carefully, such techniques can render
the images looking artificial, a loss of sharpness, and almost video like.
When compared to 1.85, anamorphic 2.40 uses 55% more area on the
negative. This is not insignificant.
2. Composition: Because of the greater height of the 1.85 aspect ratio,
ceilings of sets are more prone to being photographed. This can be a
restriction on how easily a cameraman can light an interior set (visible
ceilings limit where a cameraman can hang lights). On some sets, it may
require additional construction, and has been the experience on films
shooting on studio back lots. Sound can also be compromised by the
microphone having to be farther away.
3. Magnification: When projected, the area of the frame for 1.85 is sub-
jected to much greater magnification on a screen than an anamorphic
frame, resulting in more apparent grain in the image.
4. Jump and Weave: If your motion picture will be projected from a film
print, film projectors always have some level of jump and weave, rang-
ing from annoying to barely detectable. Because of 1.85’s vertical height
on the film frame, it is subjected to a 55% increase in visible jump and
weave over an anamorphic image projected in the same theatre. This
artifact is eliminated with digital projection.
5. 70mm: Not truly compatible with standard 70mm: Although it can be
done, there is a large amount of unused print on the sides when blown
up to 70mm. Also, because of the greater magnification in 1.85/70mm
| 69
prints, grain is much more apparent than in anamorphic blow-ups to

70mm. While not generally an issue any more, this may be a choice
for special events where a film is to be projected on a particularly large
screen (e.g., stadiums or outdoor venues), and 70mm is chosen to
mitigate jump/weave and gain more light on the screen. Less of an issue,
with the proliferation of digital projection, but still an available option.
II. THe 2.40 ASPeCT RATIO

Photographed with Anamorphic (Scope) lenses
The following is a discussion of the 2.40 aspect ratio photographed with
anamorphic lenses. A discussion of Super 35 composed for 2.40 will follow.
Anamorphic 2.40:1 (also known as CinemaScope®, Panavision®, Techno-
vision®, or any other number of brand names), optically squeezes the width
of the image to fit within the 35mm full aperture. While this discussion is
about current state of the art 35mm anamorphic releases, we are now in an
era of digital delivery where use of the area reserved for analog soundtracks
could be employed again. For a film only presented digitally, a cinematog-
rapher could choose to use the entire full aperture width of the frame and
compose for a 2.55 or 2.66:1 aspect ratio as when CinemaScope was origi-
nally conceived.
Figure 6 portrays how anamorphic 2.40 images get from camera to screen.
Many myths persist from the early days of CinemaScope when lenses dis-
torted images, film speeds were slow, and sets required the use of arc lamps
and their associated lighting crews. With today’s sharper anamorphic lenses,
combined with higher speed film stocks and more manageable lighting
packages, the challenges of shooting anamorphically has all but disappeared.
A. ADVANTAGeS of Anamorphic 2.40

1. Negative Area: The most salient advantage is the much larger negative
area. A 55% increase in negative area over 1.85, results in finer grain,
better opticals, and an increase in apparent sharpness (I say apparent,
because while a similar image photographed in 1.85 may be sharper, the
increase in grain and greater magnification actually make it appear less
sharp). This difference becomes most apparent after going through the
dupe negatives.
2. Composition: Allows for complex compositions. Able to do a tight
close up on two individuals simultaneously. Action can be spread across
a wide expanse of the frame, or a subject can be placed at one extreme
edge of the frame with an expanse of negative space emphasizing that
subject. Objects can be used to occlude part of the frame, creating an ar-
tificially smaller aspect ratio for a given scene or camera angle. Gordon
Wills, ASC’s cinematography in Manhattan does this to great effect, as
70
|
Figure 6: Anamorphic 2.40:1

With its roots in CinemaScope, it is
still referred to by many as 2.35:1.
The anamorphic aspect ratio was
actually modified slightly in the
early 1970s to 2.40:1. The image
at left is how a 2.40:1 composition
might appear in the viewfinder of
a camera equipped for anamorphic
cinematography. Also called “Scope,”
and “Panavision.”
Negative: In photography, anamorphic lenses
“squeeze” the image to fit within the 35mm frame.
The frame at left is how the “squeezed” image
appears on the negative. The area to the left of
the image is where the soundtrack is placed in the
release print. In truth, image information is exposed
in the soundtrack area; it is blank in this example
to emphasize it is not normally used in anamorphic
cinematography.
Interpositive (IP): To produce a release print,

the anamorphic negative is first contact printed to
a color interpositive. The interpositive is similar
in appearance to a color negative, yet contains a
positive image.
Internegative (IN): The interpositive is then con-

tact printed to a color internegative. This produces
the dupe negative that enables mass release print
manufacture without risking the original negative.
In a DI workflow, the IN would be recorded directly
out on a laser film recorder, bypassing the IP step
altogether.

printed to color positive print stock to produce the
release print. Upon projection in a theater, anamor-
phic projection lenses “unsqueeze” the image to
appear as originally composed above. The Dolby
Digital tracks are between the track side perfora-
tions; SDDS is outside the perfs; DTS is between
the analog track and the picture area.
© 2012 Robert C. Hummel

| 71
well as using the wide 2.40 composition to photograph a very vertical

city.
3. Reason for Choice: Traditionally the format and aspect ratio of choice
for films with a lot of action or big production values. Conversely, many
successful films have been photographed in this format that were inti-
mate, dramatic interior pieces.
4. Field of View: Most closely approximates the normal field of vision.
5. Lighting: When shooting interiors, ceilings become obscured, poten-
tially giving the cameraman more alternatives for placement of his
lighting. The same applies to exterior city scenes with the possibility
of less vertical set construction. However, these advantages rely on the
chosen composition.
6. Image Quality: A clear advantage is realized when anamorphic nega-
tives are scanned at 4K resolution for projection in digital cinema the-
aters. The area of negative used in anamorphic films means the grain
structure at least equals the resolution of a 4K digital projector, and the
benefits of the 4K scan are realized whether projecting in 4K digital
cinema, 2K digital cinema, or displayed on a standard definition TV set.
The “noisy” grain structure of many 1.85 and Super 35 films is readily
apparent even on 2K projectors.
7. 70mm: While generally only an issue for a special event presentation,
anamorphic is more compatible with 70mm. Because of the original
negative area, there is less of a blow up than 1.85 or Super 35, resulting
in finer grain in the 70mm print. Also, the aspect ratio can more closely
fill the entire 70mm print frame’s 2.2:1 aspect ratio.
It must be said that poor quality anamorphic optical printing lenses
can negate any advantage the anamorphic film element has. It is im-
portant that any lab implementing such a blowup uses state of the art
optics. Or, in the case of digital film recorders, it is important that the
facility executing the blow up uses advanced scaling algorithms.
B. DISADVANTAGeS of Anamorphic 2.40

1. Video Transfer: Compromise of composition when transferred to
standard definition’s video aspect ratio. To extract a 1.33 (4 x 3) video
image directly from the center of the 2.40 frame usually results in odd
compositions and the loss of relevant action.
This is achieved with the “pan and scan” technique (panning the
width of the 2.40 frame, following the most important action). While
the technique is not more expensive to perform when doing the video
transfer, it usually takes longer to perform because of the composition
decisions required for each scene. Most importantly, many people feel it
damages and compromises the original compositions.
72
|
Surprisingly, there still exist m ajor cable channels that transmit a 1.33
standard definition signal and present films panned and scanned.
The alternative is to release videos in letterbox format, where the
2.40 format is maintained by putting black mattes above and below the
frame. This is a common practice in DVD releases of films. Letterbox-
ing a 2.40 image with HDTV’s 1.78 (16 x 9) aspect ratio isn’t objection-
able at all, yet there are still cable outlets that resist respecting a film’s
original composition.
Performing an HDTV 16 x 9 pan and scan can have just as many
compositional compromises as those for standard definition TV.
The difficulty in video transfer is the most often stated disadvantage
of the 2.40 format.
It has always been the position of the ASC that the original com-
positions of motion pictures should always be respected. Therefore,
letterboxing of widescreen films is the preferred practice. When 1.33
motion pictures are presented on widescreen displays, the image should
be presented with “side curtains” or in “pillar-box” format.
2. Expense: It is often said that Anamorphic is more expensive than 1.85.
However, the difference in cost between an anamorphic lens package
vs. a 1.85 lens package is negligible. Panavision ‘C’ series anamorphic
lenses would be less than few thousand dollars more expensive over the
course of a ten-week film schedule.
Also, discussions with a number of prominent cameramen indicate
they wouldn’t increase the size of their lighting package significantly for
the 2.40 aspect ratio. In fact, some say it wouldn’t change at all.
3. Composition: Single close-ups result in wide areas on either side of a
face, with potential for distracting objects in the frame. However, due to
the nature of anamorphic’s longer focal length lenses, usually anything
in the background on either side of a face would be severely out of focus,
tending to emphasize the subject.
4. Set Construction: Many feel that sets need to be built wider because
of the wider aspect ratio. There are also many who feel it doesn’t mat-
ter, and can be accommodated by choosing lenses carefully. See again
Figures 3 and 4 and the discussion under Composition (page 64).
5. Blocking: Some directors have a hard time blocking action within the
wider frame.
6. Extras: Expense of additional extras may be necessary for some crowd
scenes.
7. Valuable lenses: Anamorphic lenses are far too valuable to risk in any
risky stunt situation (Panavision’s Primo lenses for example). Thus al-
ternative forms of capture with a Super 35 type of setup must be used,
risking the material may not cut well with the balance of the anamor-
| 73
phic originated material. Though, if used for a quick action shot, the
difference in image quality will, most likely, not be detectable.
III. SUPeR 35 FORmATS

The Super 35 formats, also sometimes referred to as Super Techniscope,
Super 1.85, and Super 2.35, are all flat, spherical lens formats using similar
equipment to that used in 1.85 photography.
When created in the film lab, all of the Super 35 formats require an optical
step when making dupe negatives for 35mm release prints. When a digital
intermediate is employed, the reformatting happens when a digital negative
is created on a laser film recorder, resulting in a remarkably higher quality
35mm release element than one created optically. In digital cinema presenta-
tions this is not an issue, as the release element is effectively first generation.
At times, some will suggest shooting Super 35 composed for 1.85 (a.k.a.
Super 1.85). The reason for this is the belief that the slight increase in negative
area with Super 1.85 will yield a finer grained image for release. In tests with
absolutely solid, perfect exposures this difference is almost imperceptible,
and is hardly worth the effort for the subtle gain in resolution. The process of
sending the image through interpositive and internegative, and reposition-
ing for standard 1.85 release, negates any benefit from the small increase in
negative area. In practice, when created photochemically, depending on the
scene and how well it has been photographed, Super 1.85 can end up looking
worse than standard 1.85.
In a film lab workflow, Standard 1.85 produces all dupe negatives and
prints with contact printing, while Super 1.85 requires an optical step to
reduce the image into the standard 1.85 area. Contact printing significantly
reduces the appearance of grain, while any optical step precisely focuses the
grain in a negative, effectively enhancing the appearance of grain.
While a digital intermediate eliminates any generational loss of quality in
Super 1.85, the benefits are still not sufficient to justify the effort.
As for arguments that Super 1.85 yields a better 1.85 blow up to 70mm,
the difference is subtle, and only noticeable in a direct A/B or side-by-side
comparison, otherwise it is indistinguishable. If, however, a scene is already
committed to an optical step (i.e., a visual effects shot), Super 1.85 may pro-
vide an improvement in negative area that results in a better image quality
when compared with a standard 1.85 image going through the same VFX
processes.
When people speak of the Super 35 format, they are usually referring to its
use composed for a 2.40:1 aspect ratio, the same ratio as anamorphic 2.40.
Anamorphic 2.40 uses special lenses that squeeze the wide image to fit
within the special anamorphic aperture frame (full aperture less the analog
soundtrack area, a.k.a. the sound aperture). Super 35 composes for 2.40
74
|
Figure 7: Super 35/Super Techniscope 2.40:1

The image at left is how a 2.40:1
aspect ration might be composed in
the viewfinder of a camer equipped
for Super 35 cinematography.
Negative: Super 35 is a full aperture format. This

means the entire frame is exposed with image in-
formation, including the area normally reserved for
the soundtrack. The desired aspect ratio is achieved
by masking off information above and below the
2.40:1 composition. The lines delineating the 2.40
composition are for illustration purposes only; these
do not appear on the negative.
Interpositive (IP): To produce a release print, the

general practice is for the full aperture negative to
be contact printed to a color interpostive. The inter-
positive is similar in appearance to a color negative,
yet contains a positive image. In a DI workflow, this
step would be skipped, and an Internegative would
be directly filmed out on a laser film recorder.
Internegative (IN): In order to make room for the

soundtrack, the ultimate destination of a Super
35 film is an anamorphic release print. It is at this
stage that the 2.40 composition is blown up from
the interpositive and squeezed to conform to a
standard anamorphic frame on the internegative.
This internegative is the dupe negative that enables
mass release print manufacture without risking the
original negative.

printed to color positive print stock to produce
the release print. Upon projection in a theater,
anamorphic projection lenses unsqueeze the image
to appear as originally composed above. The Dolby
digital tracks are between the track side perfora-
tions; SDDS is outside the perfs; DTS is between
the analog track and the picture area.

| 75
with standard lenses, making use of the full aperture area, extending the
width of the frame into that area of the negative traditionally reserved for
the soundtrack; yet the composition is just over 2 perfs high. Although most
cameras already expose picture information in the soundtrack area, it nor-
mally goes unused. Figure 7 portrays how Super 35 composed for 2.40 gets
from camera to the screen.
One version of Super 35 is where the image is centered on the frame of
film, referred to as common center (see Figure 8). As the figure illustrates,
information is exposed over the entire full aperture area of the film. The
filmmaker decides what format he is composing for, and it is that aspect
ratio the film lab or digital intermediate facility will eventually extract from
the frame for theatrical pre-
sentation.
Another method of photog-
raphy for Super 35 is referred 2.40 1.85
to as common topline (see

Figure 9). Common topline
derives its name from the
ground glass of the camera Figure 8: With common center, Super 35 images are
having multiple formats aligned with the vertical center of the full aperture
frame.
scribed on it, all having the
same, or common, topline.
This variant of Super 35
was originally conceived with
the notion that it could be 2.40
a generic film format; shoot 1.85
a movie with the option of
1.33
releasing it in any aspect
ratio you want. The com- Figure 9: With common topline, Super 35 apect
ratios share the same top, or headroom, for their
mon topline was supposed to compositions. Compositions other than 2.40:1 are
lessen the effect of changing obtained by the bottom of the composition.
aspect ratios by maintaining
the headroom and raising or lowering the bottom of the frame. However,
most cinematographers find that composing for multiple formats neutral-
izes the effect of any one composition. In actual practice, it has mainly been
used to mitigate conversion of Super 35 wide screen films to the 1.33 (4 x 3)
aspect ratio of standard television. Yet, even in this application, the change in
composition from 2.40 to television’s 1.78 or 1.33 aspect ratios can rarely be
achieved automatically (close-ups become medium shots, etc.), and usually
requires the same care as an Anamorphic pan and scan.
Experience has shown, as most filmmakers will agree, just modifying a
film’s aspect ratio to fit within the video realm is a creative process. To as-
76
|
sume that a generic format will automatically deliver pleasing compositions

no matter what aspect ratio you choose will not hold up creatively.
That being said, there are several filmmakers that employ the common
topline format with great success and feel it allows them to create a ver-
sion for those venues still displaying a 4 x 3 image that is less objectionable
than panning and scanning. Those most successful at the transformation are
careful to evaluate each scene and camera angle on its own merits, and don’t
attempt a generic framing for an entire film.
Of course, the ideal is to view films in the composition they were origi-
nally intended to be seen, without any alteration in the composition.
The only other technical note with regard to the common topline format,
is that scenes photographed with a zoom in common topline do not au-
tomatically transfer over to another format by just lowering the bottom of
the frame line. Meaning, if a 1.33 extraction is performed, the zoom will
still be centered towards the top of the frame, where the 2.40 composition
is centered, and a 1.33 composition will appear to drift upwards during a
zoom. Any other aspect ratios must be centered on that same point (1.33,
1.85, etc.) when a zoom takes place within a shot. For these shots, common
topline must be panned and scanned.
The rest of this discussion will only deal with Super 35 composed for a
2.40:1 aspect ratio, and apply to either common center or common topline
versions. Compositionally, Super 35’s 2.40:1 aspect ratio has the same com-
positional benefits and constraints as anamorphic 2.40; here we discuss
those qualities that are unique to the Super 35 format.
A. ADVANTAGeS of Super 35 Composed for 2.40 Aspect Ratio

1. Depth of Field: A principal advantage of this format is its greatly in-
creased depth of field over anamorphic 2.40 for the same field of hori-
zontal view. To achieve a width of composition (field of view) the same
as with an anamorphic lens, Super 35 will use a focal length half that of
the anamorphic lens.
For example, to capture the same composition as an anamorphic
50mm lens, a Super 35 setup (from the same nodal point) would use a
25mm focal length. As with all lenses, shorter focal lengths have greater
depth of field; therefore, for a given composition, the Super 35 will have
a greater depth of field vs. anamorphic.
However, as stated in the advantages of 1.85, the potential for greater
depth of field can be minimized if lenses are shot “wide open.”
2. Expense: An often-stated advantage is the production savings in the
lens/camera package over anamorphic. Yet the savings over the course
of a production would amount to less than $20K per camera. This be-
comes even less relevant in a traditional lab workflow, as the expense
| 77
Figure 10: 65mm/70mm 2.2:1

This Image is composed to the 70mm
aspect ratio of 2.2:1. While 70mm
uses a substantially larger negative
area than any of the 35mm formats,
its aspect ratio is not quite as “wide”
as anamorphic’s 2.40:1.
Negative: The negative format

for 70mm is 65mm. The extra 5mm
added later to the print dimension
was to accommodate the addition
of magnetic soundtrack striping to
either side of the perforations.
Interpositive (IP): To produce

a large number of 70mm release
prints, the 65mm negative is contact
printed to a color interpositive. The
interpositive is similar in appearance
to a color negative, yet contains a
positive image.
Internegative (IN): The inter-

positive is contact printed to a color
internegative. This produces the dupe
negative for producing large numbers
of 70mm prints without risking the
original negative.
70 mmRelease Print: The 65mm

internegative is then contact printed
to 70mm color positive print stock to
produce the 70mm release print. The
magnetic striping illustrated here,
is only to indicate where they used
to be. Magnetic tracks are not in
use any more). The area outlined in
yellow is the projection aperture of
© 2012 Robert C. Hummel the 70mm frame.
78
|
of optical Super 35 dupe negatives (needed for 35mm release prints)

negate any cost savings in production.
When a digital intermediate is used for Super 35, the need for lab op-
tical dupes is eliminated because a laser film recorder is able to produce
remarkably high quality anamorphic negatives already optimized for
35mm release.
3. Multiple Aspect Ratios: Able to shoot a film composed for 2.40 and,
if necessary, change directions and release in 1.85 by increasing the top
and bottom of the frame. For most filmmakers, however, this would be
a serious compromise of the original composition (see Figures 5 and 6
and Disadvantage #9 below).
4. Size of Lenses: Lenses are much smaller than anamorphic, resulting in
a smaller, more lightweight and portable camera package. This smaller
size allows the camera to fit in smaller places than some large anamor-
phic optics allow (this is one of the reasons it was chosen for Top Gun;
the cameras were able to fit in the aircraft cockpits).
5. Video Transfer: Claimed to be a simpler 4 x 3 standard definition video
transfer by just doing a full aperture extraction, resulting in a dramatic
increase in the top and bottom areas over the original 2.40 composition
(See Figures 5 and 6). In practice this never works, since a full frame
extraction is such a distortion of the original composition (e.g., close-
ups become medium shots). In actual practice, the best transfers are a
combination of full aperture extractions where appropriate with varia-
tions of panning and scanning when required.
The word of caution here is that adapting a motion picture to the
video domain always requires careful attention, and there are rarely any
quick-fix ways to achieve that.
6. 70mm: Often claimed to be more compatible with 70mm than anamor-
phic. Some have this impression because Super 35 is a straight blowup
to 70mm, while anamorphic has to be unsqueezed when enlarged to
70mm.
This would be true if Super 35 had an equivalent negative area to
anamorphic. As it stands, anamorphic’s greater negative area makes
up for any possible loss of resolution when unsqueezed to 70mm. The
result is, 70mm prints from Super 35 appear noticeably grainier than
those from anamorphic negatives. This holds true for both digital and
optical blow-ups.
B. DISADVANTAGeS of Super 35 Composed for 2.40 Aspect Ratio

1. Negative Area: Most notable is the small negative area. It has only
slightly more negative area than standard 1.85 photography. When com-
pared to anamorphic 2.40’s projection aperture, anamorphic’s negative
| 79
Figure 11: ImAX

IMAX photography photographs
65mm 15 perf film. When using the
full area of the frame, the aspect
ratio is 1.43:1. The dashed yellow
reticle indicates the IMAX projection
aperture. Other aspect ratios can be
achieved with cropping the image in
similar fashion to Super 35, yet with
massively more image information.
While previously limited to the
domain of theme parks and science
museums, films such as The Dark
Knight, Inception, and The Dark
Knight Rises (among others) have
photographed significant portions
of the films in IMAX to great effect.
When presented in IMAX venues,
the balance of material shot in 2.40
or 1.85 are converted as discussed
below and intercut with the full
frame IMAX material.
ImAX Conversions
ImAX Releases of Feature Films
IMAX Releases of Feature Films:
When 1.85 or 2.40 compositions
are released in IMAX, the original
composition can be maintained by
letterboxing the image within the
“taller” IMAX 1.43:1 projection
aperture. In this example, the image
is a 2.40:1 aspect ratio. IMAX also
positions the letterboxed image
slightly lower in the frame for a more
pleasing position for the audience on
the IMAX screen.
ImAX Pan and Scan

IMAX Pan and Scan: There have
been times when films have been
shown on the IMAX screen and
enlarged to fill the entire IMAX
image area. In this example, you can
see how the original 2.40 composi-
tion from above is enlarged to fill
the IMAX projection aperture, which
results in 40% of the image not being
seen. This method has compromises
to composition very similar to “pan-
ning and scanning” for television.

80
|
has over 53% more area. The difference in negative area becomes most
pronounced after 35mm dupe negatives are made. Anamorphic dupe
negs are made with contact printing, which in itself tends to lessen the
appearance of grain. Super 35 dupe negs created in a film lab involve an
optical step where the image is blown up, then squeezed to produce an
anamorphic image for 35mm release prints. Because of this optical step,
grain in the negative tends to be more sharply resolved. Combine that
with magnifying a smaller area of negative to fill the same screen area,
images become noticeably grainier than the same scene photographed
anamorphically.
The most prevalent method for posting a Super 35 is to employ a
digital intermediate. With a DI, the detrimental effects of the film lab’s
optical steps are eliminated. However, if negatives are scanned and
projected at 4K, the differences between anamorphic and Super 35
negatives again become severe, with the anamorphic having much less
apparent grain, greater sharpness and clarity of the image.
2. Opticals: If a film laboratory workflow is employed, all dissolves and
fades must be done with double IP’s or A & B printing, for best image
quality. If traditional optical methods are employed, they become exces-
sively grainy after being subjected to the second optical step required to
produce the film release prints.
Obviously, you will usually be using a digital intermediate, and these
disadvantages vanish.
3. Composite Prints: In a film lab workflow, because of the full aperture
image, composite prints cannot be struck until after the image has been
repositioned into a dupe negative, thus making room for the optical
sound track. If digital or HD previews are employed, this becomes a
nonissue.
4. Prints from Original Negative: In traditional photochemical work-
flows, because of the optical step involved, original negative composite
show prints cannot be struck. Actually, it is technically possible, but can
only be done with complex procedures and such a high risk of failure, it
doesn’t merit subjecting the original negative to the handling involved.
As you may have guessed by now, with a digital intermediate, a
“new” negative will be created on a film recorder that will integrate the
required anamorphic squeeze without introducing the grain found in
traditional optical printing.
5. Previews: More difficult to preview a traditional film print, because of a
special projection mask required for the full aperture work print. Since
Super 35 uses the area reserved for a soundtrack in the workprint stage,
many film theatres cannot be adapted to project the format. A nonissue
for digital previews.
| 81
6. Projected Image Size: Not an issue in digital projection. However, when

projecting the full aperture work print image, the 2.40 composition will
be smaller than the same image projected from an anamorphic print. This
is because the standard practice with Super 35 is to use a 1.85 lens when
projecting the work picture. While possible to get custom lenses that en-
large the image to the same size as an anamorphic print, it isn’t practical
since you would need a special lens for each theatre where you would want
to do this. Also, while a special lens may equal the size of an anamorphic
projected image, the magnified image also means much greater magnifica-
tion of projector jump and weave, and a loss in projection brightness.
7. Editing: On the rare occasion that a flatbed editing console would be
used (KEM), it may require adaptation to reveal the image contained in
the soundtrack area.
8. Choice of Negative: Because of the small negative area, cinematogra-
phers should limit choice of negatives to slower speed stocks, or over-
expose high speed negatives a stop or more for better grain quality,
often negating the advantage of the high speed negative.
Because of the optical step involved, proper exposures are critical.
Super 35 is not forgiving, should a negative be underexposed. In one
instance, a film was so poorly exposed, that the film was altered from
the 2.40 composition to a Super 1.85 composition as the reduction em-
ployed in Super 1.85 was less objectionable than the steps involved with
creating an anamorphic element.
Sometimes a DI can mitigate this problem, but not without some
compromises.
9. Pan and Scan: 4 x 3 video transfers can involve panning and scanning of
the widescreen aspect ratio. This is also a pan and scan of a much smaller
negative area than anamorphic 2.40, resulting in a lower quality video
transfer for 4 x 3 presentations. This becomes most evident in letterbox
versions of a film and particularly in 16 x 9 pan and scans for HDTV.
10. Visual Effects: There is potential for more expensive visual effects, if a
decision is made to have coverage beyond the 2.40 composition, allow-
ing for full frame video transfers. Matte shots, miniatures, etc., might
be compromised on full frame transfers if the image isn’t protected
completely to 1.33 (see Figure 5). However, the expense of coverage
above and below the 2.40 composition for visual effects shots could add
significant cost to those effects, making a pan and scan the more cost-
effective option.
DIGITAl mOTION PICTURe CAmeRAS

Suffice it to say that any aspect ratio is achievable with the current digital
motion picture cameras available.
82
|
Up until the past few years, 1920

the capture format of most
digital cameras has been that
of HDTV (1920 x 1080) or
what is commonly called 2K.4
1080
2.40:1 Scope Aspect Ratio 800
The aspect ratio of all but a few
of these cameras is the HDTV
aspect ratio of 1.78:1 (16 x 9
in video parlance). This 1.78:1
aspect ratio is a result of the Figure 12: In a 1920 x 1080 digital camera, only the
different camera manufactur- center 800 lines are used for a “Scope” or 2.40:1
aspect ratio.
ers leveraging what they have
built for HDTV broadcasting cameras. It’s rare to find a digital camera that
doesn’t owe part of its design to television cameras.
More recently however, Arri, Red, and Sony have built cameras that have
unique designs optimized for motion picture cinematography.
Previously, the photo chemical process of film put all the burden of im-
aging science on the film itself, allowing motion picture camera manufac-
turers to focus their tasks, when building cameras, on the functionality of
everything that was required to get the image to the film; lenses, stable film
transport, ergonomics, etc.
With the world of digital imaging, there has been a paradigm shift. Now
the burden of the imaging science has been put on the camera manufactur-
ers themselves; something once left to the likes of Kodak or Fuji.
When composing for the 2.40:1 aspect ratio, most digital cameras will
capture a letterboxed 2.40 slice out of the center of the imaging sensor,
which, in 1920 x 1080 cameras, results in the pixel height of the image being
limited to only 800 lines.
Cameras with sensor aspect ratios approaching 2.0:1 or 1.33 sensors that
can use anamorphic lenses mitigate the compromise of clarity when com-
posing for 2.40:1.
4. When film resolutions are discussed, and the terms 2K or 4K are used, these refer
to the number of lines that can be resolved by the film. In the case of 2K, that would
mean 2048 lines or 1024 line pairs as photographed from a resolution chart. In the
case of 4K, that would mean 4096 lines or 2048 line pairs (4096 lines) as photo-
graphed from a resolution chart.
In digital imagery the term is applied a bit more loosely. While 2K and 4K still
mean 2048 and 4096, respectively, with digital scanning and photography it refers to
a number of photo sites on the scanner or camera image sensor. Numbers of pixels
does not necessarily translate into actual image resolution. Add to the confusion that
many manufacturers have a curious habit of rounding up; which is how 1920 pixels
gets called “2K.”
| 83
Yet another camera does creative things with the clocking of CCD pixels
so that the entire imaging area is still utilized when shooting a 2.40 image,
with a subtle compromise in overall resolution.
We are already witnessing more and more cameras with imaging sensors
with 4K pixel resolution. So, hopefully, we’ll be back to movies having the
highest fidelity imagery again.
Resolution vs. megapixels

When it comes to determining which digital camera to choose, don’t allow
claims of “megapixels” or numbers of “K” to influence your understanding
of what a camera is capable of. They are terms that are used loosely to indi-
cate the resolution of a digital camera that don’t follow any set guidelines.
There are many factors that would have to be taken into account if you were
going to evaluate the merits of a digital imaging device based on specifica-
tions alone. Two different cameras with precisely the same number of pixels/
photosites can deliver images of dramatically different resolution, tonal
scale, and dynamic range.
The most straightforward way to understand a digital motion picture
camera’s capabilities is to shoot your own rigorous tests and evaluate the
images in a proper viewing environment (see “Screen Heights are Critical”
on page 60). In the same manner when a new film stock has been intro-
duced, the best way to truly understand that emulsion’s capabilities is to test
it, rather than rely on the claims of the manufacturer.
Also, it will be less confusing if you focus your evaluations on the final
image delivered by a given digital camera. Claims about a camera’s imag-
ing sensor can be influenced by marketing. If we concentrate on the final
processed image that is delivered for projection, color correction, and final
presentation, we will be evaluating a camera’s true caliber. This cannot be
achieved looking at an on-set monitor.
Fill Factor
There is one area of a digital camera’s specifications that is most helpful
in determining its sensitivity and dynamic range. This is the statistic that
conveys how much area of an imaging sensor is actually sensitive to, and
captures light vs. how much of a sensor is blind, relegated to the circuitry
for transferring image information. This is called the “fill factor.” It is also a
statistic that is not readily published by all camera manufacturers.
This is an area where not all digital cameras are created equal. The amount
of area a digital imaging sensor is actually sensitive to light (the “fill factor”)
has a direct correlation to image resolution and exposure latitude. With the
currently available professional digital motion picture cameras, you will find
a range of high profile cameras where less than 35% of the sensor’s total area
84
|
is sensitive to light, to cameras where more than 50% of the sensor is light
sensitive.
As film cinematographers, we are used to working with a medium where
it was presumed that the entire area of a 35mm film frame is sensitive to
light; in digital parlance, that would be a fill factor of 100%. When a digital
camera has a fill factor of 40%, that means it is throwing away 60% of the
image information that is focused on the chip. Your instincts are correct if
you think throwing away 60% of image information is a bad idea.
Since the imaging sites on a solid state sensor are arrayed in a regular grid,
think of the 40% sensitive area as being holes in a steel plate. Thus, the image
gathered is basically similar to shooting with a film camera through a fine
steel mesh. You don’t actually see the individual steel gridlines of the mesh,
but it tends to have an affect on image clarity under most conditions.
The fill factor is accurate only when it ignores lenslets, and only reflect
the size of the actual photosite well that captures photons vs. the rest of the
imaging area.
With this statistic, you can quickly compare camera capabilities, or at least
understand their potential.
The higher the fill factor of a given sensor (closer to 100%), the lower the
noise floor will be (the digital equivalent of film grain) and the better the
dynamic range will be.
In addition, solid state sensors tend to have various image processing cir-
cuits to make up for things such as lack of blue sensitivity, etc., so it is impor-
tant to examine individual color channels under various lighting conditions,
as well as the final RGB image. Shortcomings in automatic gain controls, etc.
may not appear until digital postproduction processes (VFX, DI, etc.) begin
to operate on individual color channels.
Sound complicated? Perhaps, but all you need to understand is that
product claims can, and will, be misleading. We’ve lost our way a bit when
thinking that counting pixels alone is a way of quantifying a digital cam-
era’s capability. A return to photographing resolution charts, and actually
examining what these cameras are capable of will serve you much better in
understanding how a given camera will help you tell your story.
In short, do not be seduced by technical specification mumbo jumbo.
Look at images photographed by the camera in question, and evaluate from
a proper distance of screen heights from the image.
Scanner vs. Cameras

To clarify, at the risk of confusing, 2K and 4K film scanners often capture
more information than their camera counterparts; a 2K film scanner will
usually have a CCD array of 2048 x 1556, while a 4K scanner will capture a
4096 x 3112 image. The lesson here is that one’s definition of the dimensions
| 85
Photogate Sensor Photodiode Sensor

Photosites/
Wells
RED GREEN BLUE
BLUE GREEN
RED GREEN BLUE

GREEN RED
Photosites/Wells
Figure 13: At left, a Bayer pattern photogate sensor, where each photosite is transformed
to yield one full color pixel value. At right, A photodiode “macrocell” design, where it
takes six photosites to yield one full color pixel value.
of 2K or 4K can vary; the terms 2K and 4K are only guidelines, your mileage
may vary.
It is important to understand these variables in characteristics, and the
need to be very specific when describing the physical characteristics of
film cameras, digital cameras, scanners and telecines. In the world of film
cameras (and film scanners), 2K refers to an image which is 2048 pixels
horizontally (perf to perf) and 1556 pixels vertically. This image captures the
area of either the SMPTE 59 Style C full aperture frame (.981" x .735") or the
SMPTE 59 Style B sound aperture frame (.866" x .735").
A 4K scan captures the same areas of the film frame as a 2K scan, but the
image yielded is usually 4096 x 3112. With both 2K and 4K scanners, each
individual pixel contains a single unique sample of each of red, green and
blue.
In the digital camera world, 2K often refers to an image that is 1920 pixels
horizontally and 1080 pixels vertically. Again, each individual pixel contains
a single unique sample of each of red, green and blue. This sampling of a
unique red, green and blue value for each pixel in the image is what is called
a “true RGB” image, or in video parlance, a 4:4:4 image. While these cameras
have an image frame size that corresponds to HDTV standard, they provide
a 4:4:4 image from the sensor, which is not to HDTV standard; which is a
good thing, as 4:4:4 will yield a superior picture.
Pixels
Much confusion could be avoided if we would define the term “pixel” to
mean the smallest unit area which yields a full color image value (e.g., RGB,
YUV, etc., or a full grayscale value in the case of black and white). That is, a
“pixel” is the smallest stand-alone unit of picture area that does not require
any information from another imaging unit. A “photosite” or “well” is de-
fined to be the smallest area which receives light and creates a measure of
86
|
light at that point. All current digital motion picture cameras require infor-
mation from multiple “photosites” to create one RGB image value. In some
cases three photosites for one RGB value, in others, six photosites to create
one RGB image value, a then Bayer pattern devices that combine numerous
photosites to create one RGB value.
These definitions inevitably lead us to define the term “resolution” as the
number of pixels that yield a single full color or grayscale image value (RGB,
YUV, etc.).
The images above are just two of many examples of photo sensors used
by digital cameras. One illustrates a Bayer pattern array of photosites using
a photogate sensor, and the other an interline transfer array of photosites
employing a photodiode lenslet design.
© 2012 Robert C. Hummel III
Special Thanks
To Stephen H. Burum, ASC, Daniel Rosen, Garrett Smith, Evans Wetmore, and
Anne Kemp Hummel in helping bring greater clarity to this discussion.
| 87
Anamorphic Cinematography
by John Hora, ASC
A namorphic motion picture photography refers to a process by which

an image is compressed or “squeezed” optically at the time of original
photography. This geometrically distorted image is then expanded or un-
squeezed in presentation by a complementary process of de-anamorphosis.
Beginning in the early 1950s, in response to competitive pressures from
a growing television audience and reduced theater attendance, the motion
picture industry moved to enhance the theatrical experience with large
“widescreen” presentations with an aspect ratio wider than that of the previ-
ous standard of 1.37:1. The 20th Century Fox Corporation revolutionized
the industry by introducing the “CinemaScope” process, which allowed the
continued use of existing 35mm cameras and projectors. The CinemaScope
process consisted of a system of 2X anamorphic compressed wide-angle pho-
tography and projection, a very large, curved, high-gain screen, four-channel
discrete stereophonic sound, and specially perforated print stock. This special
“Fox Hole” stock had smaller perforations, which allowed more area on the
print for the magnetic striping that carried the special sound track.
In contrast to methods that simply masked the top and bottom of the
Academy Aperture and required increased magnification in projection, the
CinemaScope system increased the area of negative available for the photo-
graphic image. CinemaScope was able to use the full height of the available
35mm frame, simultaneously expanding the aspect ratio horizontally to 2.55:1
with anamorphic optics and reducing the magnification required to fill the
large screen. This basic system has continued in use with various modifications.
In accurately controlled demonstrations, it has repeatedly been shown
that the anamorphic photographic process produces an image quality that
is superior to any other widescreen process where original photography is
performed on 35mm film with standard 4-perf production cameras. One
primary reason for this is that, compared to 1:85 or Super 35 (SuperTech-
niscope) methods, anamorphic photography is able to use a considerably
greater area of the original negative, resulting in decreased grain, increased
spatial resolution and increased fineness of tonal delineation.
High-quality anamorphic release prints can also be derived from “flat”
(nonanamorphic) photography originated on larger negatives such as
65mm or VistaVision. This is often the preferred method for visual effects
work. Anamorphic 35mm production photography also yields acceptable
70mm release prints.
88
|
Other systems of production photography such as Techniscope, Super-

Scope and Super 35 (SuperTechniscope), attempt to emulate the wide
aspect ratio of the “scope” image while using considerably smaller negative
areas. In these systems, quality is lost when the image is converted to an
anamorphic release print. The small image area requires slower and finer-
grained film stocks to survive the blow-up, and many of the advantages
sought by using the process are thus nullified. The high resolution use of the
Digital Intermediate process can mitigate these losses.
The anamorphic optical system used for CinemaScope and Panavision
processes are configured to compress the horizontal field of view by a factor
of 2 times while leaving the vertical field unchanged.
There have been numerous methods to achieve this anamorphic capture
of images, and some designs date to the nineteenth century. The original
CinemaScope system was based on designs of Henri Chretien and con-
sisted of a conventional prime lens that formed the image, and a supple-
mentary afocal wide-angle attachment on the front. This attachment was
essentially a reversed Galileon telescope, similar to viewing through a set
of “opera glasses” in reverse. A large front concave negative element is
seen through a positive convex element. This is essentially a supplemen-
tary wide-angle adapter. The anamorphic effect is achieved by grinding
these lenses as sections of a cylinder rather than as sections of a sphere, as
is the case with conventional optics. The power of this attachment is zero
in the vertical axis and ½ in the horizontal, axis resulting in a “squeeze”
factor of 2. Placed in front of a 50mm prime, for example, the lens system
would behave as a 50mm in the vertical plane and as a 25mm in the hori-
zontal plane. Thus, the lens behaves as if it were two different focal lengths
simultaneously.
The anamorphic attachment had to be adjusted to bring the horizontal
field into focus at the same plane as the vertical field which had been brought
into focus by the spherical prime lens. With the introduction of integrated
designs, these two focus adjustments are automatically done by precision
mechanical coupling within the lens assembly. For example, when the focus
is changed to a closer object, the prime lens which forms the image must be
moved forward to obtain focus. Additionally, the anamorphoser’s position
and spacing must be simultaneously adjusted to bring the horizontal rays
into focus at the same plane as the vertical.
However, since the spherical prime lens will have been racked forward
from the film plane by the amount that a 50mm would demand, it will be
forward of the position that a 25mm would need. When the anamorphic
unit is adjusted to obtain correct focus, the effective focal length of this
horizontal lens system will have effectively been increased in the horizontal,
producing a larger image on that one axis. The result is that the squeeze
| 89
ratio is now less than a full 2 times in the horizontal direction when com-
pared to the unsqueezed vertical direction. Since the projection expansion
remains constant at 2 times, close-up objects (such a human face) will be
overly expanded horizontally by the projector lens, resulting in fat faces or
“anamorphic mumps.”
This can be avoided by leaving the lens at a more distant focus, even at in-
finity, and affecting focus by placing a lens of positive Diopter in front of the
anamorphoser. A method of constructing anamorphics is to manufacture
the lens pre-focused at some ideal distance for the particular lens and then
use a spherical focusing lens system in front to do the actual focusing. This
is similar to the methods used to focus most zoom lenses.
Panavision successfully eliminated the “mumps” problem by using
counterrotating cylindrical elements. When the lens was set at infinity,
these two elements were aligned to cancel each other and provided no
cylindrical effect. As focus was placed on closer objects, the prime ele-
ments were moved forward to focus the vertical axis; at the same time, the
horizontal axis would be brought to focus by the power obtained by rotat-
ing the two astigmatizers by appropriate amounts in opposing directions.
With this design, rather than the image “pumping” horizontally with focus
changes, the background will appear to stretch vertically. The lack of dis-
tortion on actor’s faces quickly made the Panavision system universally
favored in Hollywood.
On longer focal lengths, and on zoom lenses, a rear anamorphic unit can
be fitted. This is equivalent to the familiar 2X extender attachment, except
that it is again not a spherical lens but rather a cylindrical section. In this
case, it cannot increase the angle of view of the spherical fore section, but
rather is used to stretch the image vertically by a factor of 2 so that when
expanded horizontally two times by the projection lens, images will have
the correct geometry. Since the field of view is not increased horizontally
but rather decreased vertically, the focal length is described as that which
would be equivalent to a spherical lens with the same vertical field of view,
i.e., equal to two times the lens’ original focal length. Thus, a 20–100mm
becomes a 40–200mm zoom. A 200mm prime would be labeled a 400mm.
This keeps the terminology consistent. Also, the 2X vertical stretch will
result in a loss of light of one stop, which is half what a conventional 2X
extender would have required, since the expansion of the image is only
along one axis, the vertical.
As a rule, short lenses are anamorphosed in front of the spherical compo-
nents, and long lenses in the rear.
Other methods of compressing the image include the use of prisms and
mirrors. Two wedge prisms may be placed before a spherical prime so
that they will compress the image in one direction. This system allows the
90
|
squeeze factor to be varied as the angle between the prisms is increased

or decreased. To focus closer than infinity, a spherical focusing unit must
be incorporated in front of the prisms, and it is possible to add the coun-
ter rotating astigmatizers. Traditionally, such prism systems were usually
applied to lenses with narrow angles of view, and long lenses working at
greater distances, such as those used in projection. Light losses are greater
than with cylindrical lens systems. Prism systems have been used at lower
squeeze ratios for original photography in the Ultra-Panavision and MGM
Camera 65 systems with outstanding results.
Curved reflecting prism-mirror systems of very high quality have been
constructed in the form of a periscope. Known as “Delrama,” these optics
were used with the Technirama system employing a moderate compression
factor of 1.5.
All of these various methods are compatible and can be mixed as long as
the long as the squeeze ratio remains consistent throughout the production
and matches the expansion during the presentation.
Using the Lenses

All anamorphic equipment must be manufactured and maintained to
very close tolerances. As with all lenses, thorough inspection and testing
should be performed before committing the equipment to a production.
Lenses should be inspected on a lens projector and collimation and focus
marks checked on the cameras to be used. Judging focus on the groundglass
of the camera while in production is much more difficult for the camera
operator with anamorphic lenses. The reliability of focus marks also takes
on increased importance.
Film tests should be shot with every lens at full aperture, as well as at
other apertures likely to be used. Due to the high (2X) power required
of the anamorphoser, most anamorphic lenses will improve dramatically
when stopped down from the maximum aperture by several stops. While
very good results should be expected at even the widest apertures, test for
improved quality at f/4 or f/5.6. Observe any differences in quality at the far
right and left sides of the image. It is important to view tests projected on a
large screen under optimal conditions. Anamorphic films are meant to be
presented large.
The quality of the lens may change dramatically when focused at the clos-
est distances, so test the lenses at these distances. In some cases it is possible
to obtain better results using a +¼ or +½ supplementary diopter, rather than
racking the focus to the closest range. Testing will reveal if this is the case.
Also test to observe how the depth of field falls off when focus is brought
forward. The differing designs can give considerably different effects, even
when set at the same focus and same f-stop.
| 91
When mounted on the camera or projector, the azimuth or vertical ori-

entation is especially critical and must be straight north and south relative
to the aperture or strange geometrical distortions will occur. This effect is
sometimes used intentionally and rotating versions have been marketed
as “Mesmerizers.” (In the 1920s Bell & Howell sold just such a device for
16mm home movies.) Looking into the front of a front-anamorphosed lens
while shining a flashlight or pocket laser pointer into it will quickly reveal
if the vertical orientation is off. Lining the camera up exactly level and per-
pendicular to a chart with a vertical and horizontal grid on it, and shooting
a film test, is recommended.
Over the years, differing generations of anamorphic lenses have found
loyal adherents due to the specific “looks” or qualities of the particular
lenses. Matching the lenses within a production is essential. The decision
to choose lenses of a certain series or manufacturer has aesthetic as well as
practical implications.
Anamorphic lenses have lots of glass, and much of it can be expected to
move with changes in focus. Some elements may even rotate in opposing
directions. The types of flare(s) that can be produced may be unique to a
particular design. The rather dramatic horizontal line that can be caused
by a light source such as a flashlight shining into the lens can be used for
dramatic effect. Preparation should involve testing for flare and these pos-
sibly unique effects, whether or not they are desired.
As with all cinema lenses, depth of field should be calculated using the
f-stop, and exposure set using the T-stop. Rather than being able to use a
circle of confusion to calculate depth of field, anamorphic lens systems pro-
duce an “ellipse of confusion.” The vertical and horizontal axes have differing
depths of field. For example, a 50mm lens behaves as a 50mm vertically and
as a 25mm horizontally. Between the vertical axis and the horizontal axis,
the lens will exhibit the depth of field equivalent to various focal lengths
between 25 and 50 millimeters. It is safe to use the most conservative value
for calculating depth of field, which would be that of the spherical prime
component. However, experience may show that the system can be quite
forgiving. Again, testing and familiarity with the lenses being used is the
best answer.
Despite what may at first seem to be formidable complications, anamor-
phic production photography remains the premier method of achieving
large, high-quality images in 35mm production. The system frees the
aspect ratio from the constraint of the shape of a predetermined camera
aperture. The nearly square shape of the scope aperture is optically efficient.
The undesired effects of higher speed films are reduced because of the large
negative area, compensating for any need to work at a smaller aperture.
The results on the screen at dailies is usually the clincher, and that quality
92
|
can be depended on to hold to the release print, as no blowup or optical

conversion is required.
John Hora is a member of the ASC Board of Governors. His many credits as a
cinematographer include the feature films The Howling, Twilight Zone: The
Movie, Gremlins and Honey, I Blew Up the Kid.
| 93
Exposure Meters
by Jim Branch
T he usual final adjustment of a motion picture camera for exposure con-

trol is made with the iris diaphragm in the camera lens. While this is
a very simple adjustment, a great deal depends upon its accuracy. Much
thought has gone into the objectives to be attained by the adjustment of the
diaphragm, and the means to obtain a correct adjustment.
It is recognized that a prime object of exposure control in motion picture
photography is to obtain consistent and uniform images of the principal sub-
jects. It is very important to obtain flesh tones that will be consistent from
one scene to the next. It is undesirable to have flesh tones that are light in one
scene, dark in the next without reason, and light again in the subsequent scene.
Correct exposure control will provide negatives that are consistent from scene
to scene, and that can be printed on a very narrow range of printer lights.
Modern exposure control is based on the use of a good light meter. The
light meter measures the effective intensity of the light and takes into account
the sensitivity of the film in the camera and the exposure time. The exposure
time is a result of the frames-per-second rate at which the camera operates,
and the angle of the shutter opening. Professional cinematographers usually
think in terms of 24 frames per second and a 170- to 200-degree shutter, that
give a basic exposure time of 1⁄50 of a second. The light meter combines all of
the foregoing factors to give an answer in terms of the appropriate camera
lens stop.
Light meters are of two types. Some measure the incident light that illumi-
nates the subject. Others measure the light that is reflected from the scene.
The results obtained from these two different types of meters may be quite
different. It is therefore important to understand the differences between the
two types.
IncIdEnt-LIght MEtErs
These meters are normally used at the location of the photographic
subject. They measure the light that is effective in illuminating the subject,
and provide an answer in terms of f-stop or T-stop for the camera lens. The
camera lens diaphragm opening is then set to match the effective intensity
of the prevailing illumination.
When the film is exposed, the various reflectances presented by the sub-
ject will then each fall into a given place in the film acceptance range. For
example, a face tone of 30% reflectance will fall into the 30% reflectance po-
94
|
Exposure Meters
sition in the film acceptance range. This method thus provides consistently
uniform face tones from scene to scene.
The incident-light meter accomplishes its purpose by doing two things. It
measures the incident light intensity at the location of the photographic sub-
ject. It also takes into account the conditions of illumination geometry—that
is, whether the subject has front key light, side key light, or a back key light.
The meter combines these factors and gives an answer in terms of the correct
setting for the camera’s lens diaphragm.
There are several makes of incident-light meters that use a three-dimen-
sional light collector. The hemispherical light collector allows these meters
to automatically perform the dual function described above.
These incident-light meters are normally used at the position of the prin-
cipal subject, with the hemisphere pointed at the camera lens. The hemi-
sphere then acts as a miniature face of the subject. All illumination that will
be effective on the subject, including key light, fill light, line light, hair light,
eye lights, etc., will be received, evaluated and integrated by the meter. The
meter will then indicate directly the correct f-stop or T-stop for the camera
lens. Incident-light meters are particularly useful because they may be used
on a scene before the principal subject appears. They may also be carried
through a scene, with the hemisphere always pointed at the camera lens, to
detect uneven illumination and particularly hot spots, into which the subject
may move during the action. This allows the scene illumination to be suit-
ably balanced before the principal subject is at hand.
In the case of outdoor photography, it is not always necessary to take the
meter to the location of the principal subject. Under such conditions, the
illumination is usually uniform over considerable areas. If the illumination
is the same at subject location and at camera location, the meter may be used
at camera location. Care should be exercised to point the meter in the proper
direction, as though it were at the subject location.
In general, exposure meters are either analog (with a needle) or digital.
The introduction of the analog incident meter with the 3-D light-collecting
hemisphere revolutionized the method of determining proper exposure for
the cinematographer.
Today, a number of companies throughout the world manufacture expo-
sure meters that employ the basic incident-type principles in their design,
but all due credit for the invention of this meter should be given to ASC
Associate Member Don Norwood, who patented it, and Karl Freund, ASC,
who was instrumental in its development. Most incident meters are provided
with suitable adapters so that they may be converted for use as a reflected-
light meter if the occasion should require it. The reflected-light adapter can
be used in a situation where the cinematographer encounters difficulty in
putting the meter into a position to read either the illumination directly on
| 95
Exposure Meters
the subject, or illumination similar to that on the subject. For example, such
a situation might be encountered when taking a picture out of the window of
an airliner in flight. The reflected-light attachment can also be used in other
situations to evaluate the relative brightness of a background.
spEcIaL EffEcts
When a special effect is desired, the cinematographer may use the incident-
light meter to first determine normal exposure for the subject. He may then
deliberately modify that value, up or down, to achieve the desired effect. This
can be done with considerable confidence, because the incident light meter
will give a firm foundation upon which to base the desired modification.
spEcIfIc sItuatIons
There are some situations, occasionally encountered in outdoor photogra-
phy, that require special attention.
1. Unusually light or dark backgrounds are cause for consideration. When
a scene includes an unusually light background, the cinematographer
may wish to first use the meter as an incident-light meter to determine
the basic exposure for the principal subject in the foreground. Then he
can convert the meter to a reflected-light meter in order to measure the
brightness of the unusual background. The second reading is then used to
modify the basic incident-light reading somewhat. The same procedure
could be followed in the case of an unusually dark background.
2. Outdoor scenes that include a subject in the foreground as well as distant
objects, such as mountains, in the background, usually also include con-
siderable aerial haze. This haze may be invisible or only partly visible to
the eye, but strongly visible to the camera. A frequent photographic result
is a recording of the aerial haze overlaid on the scene background, which
gives the appearance of an overexposed background. In such a situation, a
haze-cutting filter should be used to improve the background. In addition,
use the procedure previously described for the case of an unusually light
background.
3. Scenes consisting of a mixture of sunshine and shade areas, with the principal
subject in a shade area, can be handled by: (a) using the meter in the sunshine
area, or (b) opening up the lens by ½ to ⅔ f-stop from the meter indication.
rEfLEctEd-LIght MEtErs
A spot meter may be used at camera location and aimed at a selected spot
in the scene. The effectiveness of the meter is heavily dependent on the opera-
tor’s judgment in the selection of the spot. The selected spot must be precisely
representative of the particular combination of elements in the composition
of the scene. When using such a meter, the operator must be particularly
96
|
Exposure Meters
careful when confronted with a scene that presents strong contrasts between
the selected spot and the scene background. An example of such a situation
would be a case where a person in the foreground is in front of a very light
background, such as sky or white buildings, etc. In such a situation, the op-
erator should modify the spot reading provided by the meter according to
his own estimate of the situation. When the use of a reflected-light meter is
required, the results of determining the exposure can be greatly improved by
using a “Kodak Neutral Test Card.”
This card is a piece of sturdy 8" x 10" cardboard that is neutral gray on one side
and white on the other. The gray side reflects 18% of the light falling on it, and
the white side reflects approximately 90%. Also, the gray side has a protective
lacquer overcoat that reduces specular reflectance and resists damage due to
fading, fingerprints, soil, etc. To a light meter, an average scene is one in which
the tones, when averaged, form a tone brightness that is equivalent to middle
gray—a tone that reflects 18% of the light illuminating it (the same tone and
reflectance of the gray card). When a scene is not average, the gray card is a
reference that helps you make the proper exposure judgments. A Kodak Gray
Card is manufactured under close tolerances to provide a neutral gray-side
reflectance of 18% (+ 1%) and white-side reflectance of approximately 90%.
tEstIng
Small errors may exist in meters, lens calibrations, emulsion speeds and
development. These small errors will frequently cancel out without undue
harm to the final picture. It is when these errors add up in the same direction
that their cumulative effect is serious. It is therefore wise to test equipment,
film and meters under simulated production conditions so that errors may
be detected and corrected before production begins. It is always a good idea
to “tune up to the variables.”
Much of the material in this section of the manual is basic, but reference
should be made to ASC Associate Member Don Norwood and Eastman
Kodak Co. for the gray card information.
ExposurE MEtErs
gossen starlite
Type: Handheld exposure meter for measuring ambient and flash and incor-
porating both incident and spot meter reading ability.
Light Sensor: 2 silicon photo diodes
Measuring Capability:
Measuring Range: Ambient light: Incident (at ISO 100/21°): EV -2.5 to +18;
Reflected with 1°: EV 2.0 to 18; Reflected light with 5°: EV 1.0 to 18. Flash
Light: Incident (at ISO 100/21°): f/1.0 to f/128; Reflected light with 1°:
f/2.8 to f/128; Reflected light with 5°: f/1.4 to f/128.
| 97
Exposure Meters
Measurement Modes: lx, fc, cd/m2, fL, lxs, cds/m2, flLs

ISO Film Speeds: ISO 3.2/6° to 8000/40° (in 1° DIN increments)
Camera Speeds: 8–128 fps, additional Cine speeds can be adjusted.
Shutter Speeds: 1⁄8000 sec. to 60 min.
Flash Sync Speeds: 1 to 1⁄1000 sec.
F-Stops: f/1 .0 to f/128.
Power Source: 1.5V AA battery or 1.2V rechargeable.
Dimensions: 6½" x 2½" x 1"
Weight: Approximately 6.4 ounces.
gossen color-pro 3f
Type: Handheld digital 3-color meter for ambient and flash; determines
photographic color temperature of light sources and filtration required.
Light Sensor: 3 balanced silicon photodiodes for ambient and flash.
Measuring Range: 2000 to 40,000 degrees Kelvin.
Light Balancing Filters: -399 to 475 Mired Scale, switchable to correspond-
ing Kodak Wratten filters.
CC filter Values: 0 to 95 Magenta and 0 to 06 Green.
Power Source: 9V MN1604 or equivalent.
Dimensions: 5" x 2¾" x 1"
gossen Luna-star f2
Type: Handheld exposure meter for measuring ambient and flash in both
incident and reflected light (with 5° Spot attachment).
Light Sensor: sbc photodiode, swivel head
Measuring Range: Ambient light (at ISO 100/21°): EV -2.5 to +18; Flash
Light (at ISO 100/21°) f/1.0 to f/90.
Measuring Angle in Reflected Mode: 30°
ISO Film Speeds: 3/6° to 8000/40°
Camera Cine Speeds: 8–64 fps, as well as 25 fps and 30 fps for TV.
Flash Sync Speeds: 1 to 1⁄1000 sec., as well as 1⁄90 sec.
F-Stops: f/1.0 to f/90.
Power Source: 9V battery.
Dimensions: 2¾" x 5" x 1"
gossen Luna-pro digital

Type: Handheld exposure meter for measuring incident and reflected light.
Light Sensor: sbc photodiode
Measuring Range: Incident Light (at ISO 100/21°): EV -2.5 to +18.
98
|
Exposure Meters

Camera Cine Speeds: 8–64 fps, as well as 25 fps and 30 fps for TV.
Power Source: 1.5V battery.
Dimensions: 2½" x 4⅝" x ¾"
gossen Luna-pro digital f

Type: Handheld exposure meter for measuring ambient and flash light.
Light Sensor: sbc photodiode.
Measuring Range: Incident Light (at ISO 100/21°): EV -2.5 to +18; Flash Light
(at ISO 100/21°) f/1.0 to f/90.
Camera Speeds: 8–64 fps, as well as 25 fps and 30 fps for TV.
Flash Sync Speeds: 1 to 1⁄1000 sec., as well as 1⁄90 sec.
F-Stops: f/1.0 to f/90
Power Source: 1.5 V battery.
gossen Luna-pro f
Type: Handheld analog exposure meter for measuring ambient and flash light.
Measuring Range: Incident Light (at ISO 100/21°): EV -4 to +17; Flash Light
(at ISO 100/21°) f/0.7 to f/128.
ISO Film Speeds: 0.8/0° to 100,000/51°
Camera Cine Speeds: 4.5–144 fps.
Shutter Speeds: 1⁄4000 sec. to 8 hours.
Flash Sync Speeds: 1⁄60 sec.
gossen Luna-pro s
Type: Handheld analog exposure meter for measuring ambient sun and
moon incident and reflected light.
Light Sensor: Photoresistance (CdS)
Measuring Range: Incident Light (at ISO 100/21°): EV -4 to +17.
Measuring Angle in Reflected Light Mode: 30° (with Tele attachment).
| 99
Exposure Meters
ISO Film Speeds: 0.8 to 25,000

Shutter Speeds: 1⁄4000 sec. to 8 hours.
Flash Sync Speeds: 1⁄60 sec.
Power Source: Two 1.5V batteries.
Dimensions: 2¾" x 4⅓" x 1⅜"
Weight: Approximately 6 ounces.
gossen ultra-spot 2
Type: Handheld Spot meter for measuring ambient and flash light.
Measuring Range: Ambient Light (at ISO 100/21°): EV -1 to +22; Flash Light
(at ISO 100/21°) f/2.8 to f/90.
Measuring Angle of Reflected Light: Viewfinder (15°), metering field (1°).
ISO Film Speeds: 1/1° to 80,000/50°
Camera Speeds: 8-64 fps, as well as 25 fps and 30 fps for TV.
Shutter Speeds: 1⁄8000 sec. to 60 min, as well as 1⁄90 sec.
Flash Sync Speeds: ⅛ sec. to 1⁄1000 sec., as well as 1⁄90 sec.
Dimensions: 3½" x 2¼" x 7½"
gossen Mavolux 5032 c/B

Type: Handheld exposure meter.
Light Sensor: Silicon cell.
Measuring Modes: fc, lx, cd/m2, and fL.
Measuring Range: .01 to 199,900 lx in 4 ranges (MR I: 0.01 lx/0.001 fc;
MR II: 1 lx/0.1 fc; MR III: 10 lx/1 fc; MBR IV: 100 lx/10 fc.), 0.01 to
19,990 fc.
Luminance Measurement with Luminana attachment: 1 to 1,999,000
cd/m2, 0.1 to 199,900 fL.
Power Consumption: Approximately 75 hours.
Power Source: 1.5V alkaline manganese AA (IEC LR 6).
Dimensions: Meter: 2½" x 4¾" x ¾"
Sensor: 1¼" x 4⅛" x 1⅛"
Minolta cinemeter II
Type: Handheld digital/analog incident meter.
Light Sensor: Large area, blue enhanced silicon photo sensor. Swivel head,
270 degrees.
100
|
Exposure Meters
Measuring capability: Direct readout of photographic exposures in full

f-stops or fractional f-stops. Also measures illuminance level in foot-
candles and Lux.
Measuring Range: Direct-reading multiple-range linear circuit incorpo-
rates a high quality CMOS integrated amplifier whose bias current is
compensated against drift up to 70E° C.
Dynamic Range: 250,000 to one. Digital f-stop: f/0.5 to f/90 in 1⁄10 stop
increments. Analog f-stop: f/0.63 to f/36 in ⅓-stop increments. Photo-
graphic illuminance: 0.20 to 6400 footcandles, 2 to 64,000 Lux.
Display: Vertical digital/analog bar graph that consists of 72 black liquid
crystal bars (6 bars per f-stop), that rise and fall depending on the light
intensity. The scale can be used in three different display modes (Bar,
Floating Zone and Dedicated Zone), and in three different measure-
ment modes (f-stops, footcandles and Lux).
Display Modes:
1. Bar mode is similar to a needle-reading meter, except that the move-
ment is up and down instead of left to right.
2. Floating Zone mode: a single flashing bar forms a solid bar that
graphically indicates the range of illumination in the scene. It can
also be used for the measurement of flickering or blinking sources.
3. Dedicated Zone mode is used to save up to five separate measure-
ments.
Display Range:
ISO film speed: 12 to 2500 in ⅓-stop increments.
Camera speed: 2–375 fps
Shutter Angle: 45E° to 90E° in 1⁄9-stop increments, 90E° to 205E° in
1⁄12-stop increments
Filter factors: ⅓-stop to 7 f-stops.
Resolution: Digital: 1⁄6 stop. Analog: 1⁄6 stop.
Accuracy: Digital 1⁄6 stop.
Additional Functions: Memory store and recall.
Lamp: Electroluminescent backlit liquid-crystal display.
Power consumption: Operating reading 5 mA with backlite on.
Power Source: One 9V battery.
Dimensions: 6⅝" x 3" x 13⁄16"
Minolta auto Meter IV f

Type: Handheld exposure meter for measuring ambient and flash light.
Measuring Modes: Ambient, flash.
Ambient Range (at ISO 100): Incident: EV 2.0–19.9.Reflected: Viewfinder
5 degrees EV 2.5–24.4
| 101
Exposure Meters
Display Range: ISO : 3–8000 in ⅓-stop increments

Shutter Speed (ambient): 1⁄8000 sec.–30 min. in ½-stop increments.
Shutter Speed (flash): 1⁄500 sec.–1 sec. in ½-stop increments.
Camera Speeds: 8–128 fps.
F-Stops: f/1.0 – f/90 in 1⁄10-stop increments.
EV: -7.8 – 31.5 in 1⁄10-stop increments.
Power Source: 1.5V battery.
Minolta auto Meter V

Measuring Modes: Ambient, Flash
Ambient Range (at ISO 100): EV-2.0 TO 19.9
Reflected: View Finder 5° EV 2.5 to 24.4.
Display Range: ISO 3 to 8000 in ⅓-stop increments.
Shutter Speed (Ambient): 30 min. to 1⁄1600 sec.
Shutter Speed (Flash): 30 min. to 1⁄1000 sec.
Camera Speeds: 8, 12, 16, 18, 24, 25, 30, 32, 64, 128 fps.
F-Stops: f/0.7 to f/90 +.09
EV: -11.8 to 35.5
Power Source: 1.5V battery
Repeatability: ±0.1 EV
Minolta auto Meter Vf/Kenko KfM-1100

Same stats as Auto Meter IVF - Except:
Measuring Range: EV: -17 to 40.8 in 0.1 stop increments
Minolta Luminance: ft-1E°, nt-1E° & nt-⅓E°
Type: Reflex-viewing spot-reading automatic/manual luminance meter.
Light Sensor: Silicon Photovoltaic cell with 1E° (⅓E° in model nt-⅓E°)
angle of acceptance.
Viewing System: Focusing through-the-lens reflex type. Objective lens
85mm f/2.8. Angle of view: Circular 9E° with central 1E° (⅓E° in model
nt-⅓E°) marked circle.
Magnification: 2.96X focused at infinity.
Measuring Capability: Direct readout of illuminance level in footlamberts
or candelas.
Measuring Range: Model ft-1E°: 0.01 to 99900 ft-L (0.01step); Model nt-1E°:
0.1 to 99900 cd/m2 (0.1 step); Model nt-1⁄3E°: 1.0 to 99900 cd/m2 (0.1
step).
Display Range: Red (+) LEDs at the right of the number display indicates
10X and 100X the display reading.
Accuracy: Within +4% of C.I.E. standard +1 digit in last display position.
Screen-flicker accuracy: Within 1% of average luminance with projection
cycle of more than 72 Hz and duty of 7% (projector at 24 fps).
102
|
Exposure Meters
Analog Output: Output voltage: 1v over full scale. Output impedance:

10 kilo-ohms.
Power Consumption: 6 mA in analog mode. Meter can monitor changes in
luminance for a period up to 40 hours.
Power Source: 9V battery (Eveready 216 or equivalent).
Estimated Battery Life: Approximately 1 year with normal use.
Dimensions: 2⅞" x 6⅜" x 411⁄16"
Weight: 18⅛ ounces, without battery.
Minolta 1-degree spot Meter

Type: Spot-reading reflex-viewing exposure meter for ambient or flash light.
Measuring Method: Reflected light by silicon photo cel detector masked for
1 degree angle of acceptance.
Measuring Range: Ambient: EV 1.0 to EV 22.5; Flash: f/2 to f/90.
Accuracy: ±1⁄10-stop repeatability.
F-Stops: F-numbers: f/0.7 to f/90 +0.9 stop in 1⁄10-stop increments.
EV numbers: -43 to +28 in 1⁄10-stop increments.
Brightness difference: -9.9 to +9.9 stops in 1⁄10-stop increments.
ISO range: 12 to 6400 in ⅓-stop increments.
Ambient Exposure Time: 30 min. to 1⁄8000 sec. in 1-stop increments (cine
1⁄50 sec.).
Flash Exposure Time: 1 to 1⁄1000 sec. in 1-stop increments.
Note: F-number, EV-number and brightness difference shown in both ex-
ternal and finder displays.
Memory: 2-measurement capacity, both indicated by pointers on analog
display; digital recall possible.
Exposure Time Calculation: Analog/digital readout and recall of highlight,
shadow or averaged (midtone) exposures automatically calculated for op-
timum correspondence of brightness range of subject with film latitude.
Power Source: 1.5V AA alkaline manganese (Eveready E91 or equivalent),
carbon zinc or 1.2V nickel cadmium (Ni-Cad) cell.
Size: 1⅞" x 5⅞" x 39⁄16" (48 x 150 x 89mm).
Weight: 8½ ounces (240g) without battery.
Minolta color Meter III-f/Kenko KcM-3100

Type: Three-color digital color meter for color photography; deter-
mines filtration required and photographic color temperature of
light sources.
Film Type Settings: Daylight film balanced to 5500°K; Type-A tungsten
film balanced to 3400°K; Type-B tungsten-balanced to 3200°K.
Measuring Range (ISO 100): Ambient: EV 3–16.3; Flash: f/2.8–180 (in two
ranges).
| 103
Exposure Meters
Shutter Speed Setting Range (for flash measurements): 1 to 1⁄500 sec. in

1-stop increments.
Display Modes: LB index and CC index; LB filter number and CC index;
photographic color temperature.
Display Range: LB index: -500 to 500 mireds; CC index: 200G to 200M;
LB filter number: 80A + 80D to 85B + 81EF; Photographic color tem-
perature: 1600 to 40,000°K.
Repeatability: LB index: 2 mireds; CC index: 2 digits.
Power Source: 2-1.5V AA batteries.
pentax spotmeter V
Measuring Range: EV 1–20 (100 ASA).
Film Speeds: ASA 6-6400
Shutter Speeds: 1⁄4000 sec.-4 min.
F-Stops: f/1 to f/128.
EV Numbers: 1-19⅔; IRE 1-10.
Measuring Angle: 1 degree.
Measuring Method: Spot measuring of reflected light; meter switches on
when button pressed; EV direct reading; IRE scale provided.
Exposure Read Out: LED digital display of EV numbers (100 ASA) and up
to 2 dots (each of which equals ⅓ EV).
Photosensitive Cell: Silicon Photo Diode.
Power Source: 6V silver-oxide battery.
pentax digital 1 degree spot Meter

Measuring System: Spot measuring of reflected light; meter switches on
when button pressed.
Exposure Read Out: LED digital display of EV numbers (100 ASA) and up
to 2 dots (each of which equals ⅓ EV).
Photosensitive Cell: Silicon Photo Diode.
Measuring Angle: 1 degree.
Measuring Range: (ASA/ISO 100) EV 1-128 (EV 20 displayed as “0”).
Film Speeds: ASA 6-6400.
Shutter Speeds: 1⁄4000 sec.–4 min.
F-Stops: f/1 to f/128.
Camera Speed: 24 fps.
IRE: 1–10.
Power Source: 6V silver oxide battery (A544, PX28L or PX28).
sekonic L508c
Type: Handheld exposure meter for ambient and flash incorporating both
incident and spot meter reading ability.
104
|
Exposure Meters
Measuring Range: Incident Light: EV (-) 2 to EV 19.9 @ 100 ISO

Reflected Light: EV3 to EV 19.9 @ 100 ISO
Incident Reading Head: 270 Swivel Head
Measurement Modes: Footcandle 0.1 to 99,000, Lux 0.63 to 94,000, Foot-
lambert 3.4 to 98,000
ISO Film Speed: ISO 3 to ISO 8000 (⅓-stops)
Camera Speed: 1, 2, 3, 4, 6, 8, 12, 16, 18, 24, 25, 30, 32, 36, 40, 48, 60, 64, 72, 75,
90, 96, 120, 128, 150, 200, 240, 256, 300, 360, 375, 500, 625, 750, 1000 fps
Shutter Angle: 5 to 270 at 5 stops + 144 and 172 degrees.
Shutter Speeds: 30 min. to 1⁄8000 sec. (full, ½- or ⅓-stops)
F-Stop: f/1.0 to f/128.9 (full, ½- or ⅓-stops)
Accuracy: ±0.1 EV or less
Additional Functions: Digital f-stop and shutter speed readout in view-
finder; Parallax-free rectangular 1–4 spot zoom; Retractable incident
Lumisphere for dome or flat disc readings; Auto Shut-Off 20 min.
Lamp: Electro-luminescent auto illumination at EV6 and under for 20 Sec.
Power Source: 1.5V AA battery (alkaline, manganese or lithium)
Dimensions: 3.3" W x 6.1" H x 1.6" D (82mmW x 161mmH x 39mmD)
Weight: 8½ ounces (240g)
sekonic L608c
Type: Handheld exposure meter for ambient and flash incorporating both
incident and spot meter reading ability.
Light Sensor: Silicon photo diodes (incident and reflected)
Measuring Range: Incident Light: EV (-) 2 to EV 22.9 @ 100 ISO
Reflected Light: EV3 to EV 24.4 @ 100 ISO
Measurement Modes: Footcandle 0.12 to 180,000, Lux 0.63 to 190,000;
Cd/m2 1.0 to 190,000; Foot-lambert 0.3 to 190,000
Display Mode: Digital f/0.5 to f/128.9 (in ⅓-stops); Analog f/0.5 to f/45
(in ⅓-stops)
ISO Film Speed: ISO 3 to ISO 8000 (1⁄3-stops)
Camera Speed: 1, 2, 3, 4, 6, 8, 12, 16, 18, 24, 25, 30, 32, 36, 40, 48, 50, 60, 64,
72, 75, 90, 96, 100, 120, 125, 128, 150, 200, 240, 250, 256, 300, 360, 375,
500, 625, 750, 1000 fps
Shutter Angle: 5–270 at 5 stops + 144 and 172
Shutter Speeds: 30 min. to 1⁄8000 sec. (full, ½- or ⅓-stops)
F-Stops: f/0.5 to f/128.9 (full, ½- or ⅓-stops)
Filter Factors: 85, -n.3, -n.6, -n.9, -A3, -A6, -A9
Memory Function: 9 readings on analog scale (f/stop and shutter speed)
with memory recall and clear feature
Accuracy: ±0.1 EV or less
Additional Functions: Digital f-stop and shutter speed readout in viewfind-
| 105
Exposure Meters
er; Parallax-free rectangular 1–4 spot zoom with digital display. Shutter
speed and aperture are displayed in viewfinder; Retractable incident
Lumisphere for dome or flat disc readings; Digital Radio Transmitter
Module that eliminates the need for an additional external transmit-
ter at the meter’s position; 12 Custom Function Settings for advanced
preferences and features.
Power Source: 3.0V (CR123A lithium battery)
Dimensions: 3.5" W x 6.7" H x 1.9" D (90mmW x 170mmH x 48mmD)
Weight: 9½ ounces (268g)
sekonic L-308BII
Measuring System: Incident or reflected for flash and ambient light; Silicon
photo diode.
Measuring Modes: Ambient and flash (cord, cordless) – incident and
reflected (40 degrees)
Receptor Head: Nonrotating, noninterchangeable.
Aperture/Shutter Priority: Shutter speed priority.
Display Read-out: Digital LCD
ISO Range: ISO 3 to 8000 in ⅓-stop increments.
F-Stops: f/0.5–f/90 9/10
Shutter Speeds: Ambient: 60 sec.–1⁄8000 sec.; Flash: 1 sec.–1⁄500 sec.
EV Range: (ISO-100) EV(-) 5 to EV 26.2
Power Source: 1.5V AA battery.
Dimensions: 4.3" x 2.5" x .9" (110 x 63 x 22mm) WDH
Weight: 2.8 ounces (80 g) without battery.
sekonic L-398M
Measuring System: Incident light type, reflected light measurement is also
possible
Measuring Modes: Ambient incident and reflected
Receptor Head: Rotating, interchangeable receptor.
Display Readout: Indicator needle
ISO Range: 6 to 12,000; Measuring Range: EV4-EV17 (for incident light)
EV9-EV17 (for reflected light)
F-Stops: f/0.7–f/128
Shutter Speeds: Ambient: 1⁄8000 to 60 sec.; Flash: None.
EV Range: (ISO-100) EV 4 to 17
Camera Speeds: 8–128 fps
Power Source: Selenium photocell (no battery needed)
Dimensions: 4.4" x 2.3" x 1.3" (112 x 58 x 34mm) WDH
Weight: 6.7 ounces (190 g)
106
|
Exposure Meters
sekonic L-358
Measuring System: Incident: Dual retractable lumisphere, Reflected: with
included reflected light attachment; Silicon photo diodes
Measuring Modes: Ambient and flash (cord, cordless, multi flash)—
incident and reflected (54 degrees).
Receptor Head: Rotating 270 degree with built-in retractable lumisphere.
Aperture/Shutter Priority: Aperture and shutter priority
Display Readout: Digital LCD plus LCD analog, (auto-backlit LCD at EV 3
and under for 20 sec.)
ISO Range: Dual ISO settings: 3 to 8000 (⅓-steps)
F-Stops: f/1.0 to f/90.9 (full, ½- or ⅓-steps)
Shutter Speeds: Ambient: 1⁄8000 sec. to 30 min.; Flash: 1⁄1000 sec to 30 min.
EV Range: (ISO-100) EV -2 to 22.9
Camera Speeds: 2–360fps
Exposure Memory: Capable of nine exposure measurement readings
Shadow/Highlight Calculation: Yes
Brightness Difference: Displays the difference in 1⁄10-stop increments
Flash To Ambient Ratio: Yes
Multiple Flash: Yes, unlimited
Exposure Calibration: ±1.0 EV
Power Source: One CR123A lithium battery
Dimensions: 2.4" x 6.1" x 1.46" (60 x 155 x 37mm) WHD
Weight: 5.4 oz (154 g)
sekonic L-558
Measuring System: Dual function retractable incident lumisphere; 1° spot
viewfinder; Two silicon photo diodes (SPD).
Measuring Modes: Ambient and flash (cord, cordless, multi-flash) – inci-
dent and spot (1°).
Metering Range: Ambient Incident Light: EV -2 to EV 22.9; Reflected Light:
EV 1 to EV 24.4; Flash Incident Light: f/0.5 to f/161.2; Reflected Light:
f/2.0 to f/161.2
Receptor Head: Rotating 270 degrees; with built-in retractable lumisphere.
Aperture/Shutter Priority: Aperture, shutter priority and EV metering value
Display Readout: Digital LCD plus LCD analog, (Auto-backlit LCD at EV
6 or under for 20 seconds)
ISO Range: 3 to 8000 (in ⅓-stop steps)
F-Stops: f/0.5–f/128.9 (full, ½- or ⅓-stops); Under and Overexposure indi-
cation.
Shutter Speeds: Ambient: 30 min. to 1⁄8000 sec. (full, ½- or ⅓-stops, plus 1⁄200
and 1⁄400); Flash: 30 sec. to 1⁄1000 sec. (Full, ½- or ⅓- stops; Special flash
speeds: 1⁄75, ⅛0, 1⁄90, 1⁄100, 1⁄200, 1⁄400)
| 107
Exposure Meters
EV Range: (ISO-100) EV -9.9 to 46.6 (in 1⁄10-stops).

Camera Speeds: 2–360 fps, (fps at a 180°).
Exposure Memory: Up to nine readings on analog scale with memory recall
and clear feature.
Shadow/Highlight Calculation: Yes
Brightness Difference: ±9.9EV (in 1⁄10-stops) flash or ambient light evaluation.
Flash To Ambient Ratio: Yes; Displays percentage of flash in total exposure
in 10% increments.
Multiple Flash: Unlimited readings.
Exposure Calibration: ±1.0EV for incident and reflected independently (in
1⁄10-stops); Exposure Compensation ±9.9EV for incident and reflected
independently (in 1⁄10-stops); Filter compensation ±5.0EV for incident
and reflected independently (in 1⁄10-stops).
Power Source: Lithium type, One CR123A; Auto “shut-off ” after 20 min-
utes; Battery power displayed with a symbol in three status levels.
Dimensions: 3.5" x 6.7" x 1.9" (90 x 170 x 48mm) WDH
Weight: 9½ ounces (268 g)
sekonic L-778 dual spot f

Type: Multi-function design light meter.
Incident: 270-degree swivel head with dual function retractable incident
lumisphere.
Reflected: 1-degree to 4-degree parallax-free zoom spot viewfinder.
Photo Cells: 2 silicon photo diodes (SPD).
Measuring Range: Ambient incident: EV 0.2–EV 19.9; ambient reflected:
EV 3–EV 19.9.
Film Speed: ISO 3-8000 in ⅓-stop increments.
F-stops: f/1.0–f/128 in 1⁄10-stop increments.
Shutter Speed Range: DIP Switch preselectable full speeds or ½-stop
range.
Camera Speeds: 1⁄75, 1⁄80, 1⁄90, 1⁄100, 1⁄200, 1⁄400, 1, 2, 3, 4, 6, 8, 12, 16, 18, 24,
25, 30, 32, 36, 40, 48, 60, 64, 72, 96, 120, 128, 150, 200, 240, 256, 360
fps
EV Range: 0.9–36.1
Exposure Compensation: ±9.9EV in 1⁄10-stop increments.
Power Source: 1.5V AA penlight battery or 1.5V alkaline manganese or
lithium battery.
Repeat Accuracy: ±0.1 EV or less.
Auto Shut Off: 20 minutes.
Tri Pod Socket: ¼".
Weight: 7.4 ounces (210 grams).
Dimensions: 3.3" x 6.1" x 1.7" (84 x 156 x 40mm).
108
|
Exposure Meters
spectra cinespot 1E spot Meter

Type: Through-the-lens viewing, spot-reading auto-matic/manual lumi-
nance meter.
Light Sensor: Silicon Photovoltaic cell with 1E° angle of acceptance.
Viewing Optics: 1.6x magnification: erect system with focusing eyepiece.
Measuring Capability: Direct readout of illuminance level in footlamberts
or candelas.
Measuring Range: Low Range 0–30 fL (or 0-100cd/m2) readings legible
down to 0.5fL. High Range 0–300 fL (or 0–1,000 cd/m2), upper limit
may be increased by use of accessory 10X or 100X attenuators.
Spectral Response: Within +4%(by area) of CIE Photopic Luminosity
Function.
Accuracy: +1% of full scale or +5% of reading (whichever is greater).
Error Due To Chopped Light: +0.5 % at 24 cycles /second.
Power Source: 6V battery (Eveready 544 or equivalent).
Dimensions: 5” x 2” x 6.4”.
Weight: 15 ounces.
spectra cinespot footlambert Meter

Measuring Capability: Footlamberts or candelas/square meter; also mea-
sures illuminance with accessory reflectance standard.
Sensitivity: Low range, 0 to 30 footlamberts (0 to 100 cd/square meter on
metric version); readings legible to 0.5 footlamberts. High Range, 0–300
footlamberts (0–1000 cd/square meter on metric version); accessory
10X and 100X attenuators increase upper limit to 3000 or 30,000 foot-
lamberts (10,000 or 100,000 cd/square meter).
Angle of Coverage: 1-degree measuring field, 21-degree diagonal viewing
field.
Viewing Optics: 1.6X magnification; focusing eye-piece; illiminated scale
for low light reading.
Photodetector: Silicon photovoltaic cell.
Spectral Response: Matched to the human eye response (CIE 1931 photopic
luminosity function) within +5% (by area), using computer-selected
glass filter.
Accuracy: ±1% of full scale or ±5% of reading, whichever is greater, when
measuring blackbody sources: traceable to the National Bureau of Stan-
dards.
Dimensions: 5” x 2” x 6.4”
(127 x 52 x 162mm).
Power Source: 6V battery (A544 or equivalent).
Warranty: One year, limited.
| 109
Exposure Meters
spectra professional IV
Type: Handheld exposure meter for measuring incident and reflected light.
Light Sensor: Silicon Photovoltaic cell, computer-selected glass filters tai-
lored to spectral response of the film. Swivel head, 270 degrees.
Measuring Capability: Direct readout of photographic exposures. Also
measures illuminance level in foot-candles and Lux.
Measuring Range: One million to one (20 f-stops) direct-reading multiple-
range linear circuit controlled by microcomputer.
Display Range: ISO film speed: 3 to 8000 in ⅓-stop increments.
Resolution: Digital: 0.1 f-stop. Analog: 0.2 f-stops.
Accuracy: Digital: 0.05 f-stop.
Additional Functions: Memory store and recall.
Lamp: Optional electroluminescent lamp for backlit liquid-crystal display.
Power Consumption: Operating (reading) 5mA. Data retention 5uA.
Power Source: 6V battery (A544, PX28L or PX28).
Dimensions: 5½" x 2½" x 2".
spectra professional IV-a

Advanced Digital Exposure Meter (ADEM)
Type: Measures both incident and reflected light; reflected light angle of ac-
ceptance is 42 degrees square.
Technology: Advanced front-end op amp, custom-sealed hybrid electronic
circuitry, multiple-range linear-circuit-controlled by advanced micro-
computer and custom liquid-crystal display with aviation green backlit
electro-luminescent lamp.
Light Sensor: Silicon photovoltaic cell, computer-selected glass filters tai-
lored to spectral response of the film; swivel-head assembly turns 270
degrees.
Measuring Capability: Direct readout of photographic exposures; measures
contrast ratios and average in f-stop and illuminance modes; measures
illuminance level in footcandles (fc) and lux (lx).
Measuring Range: One million to one (20 f-stop), widest sensitivity range
available; digital f-stops: f/0.35 to f/128 in 1⁄10-stop increments; analog
f-stop from f/0.7 to f/45 in ⅓-stop increments.
Photographic Illuminance: Footcandles 0.1 to 70,000 and lux 1 to 100,000.
Contrast Ratios: Contrast ratio of two readings – key plus fill to fill light
alone from 1:1 to 999:1; f-stop difference: the difference in f-stop be-
tween a new reading and the last standard/memorized measurement is
displayed in 1⁄10-stop increments up to ±12 f-stops.
110
|
Exposure Meters
Averaging: Average of two readings and or continuous averaging in f-stop

and footcandle modes are made possible.
Film Speed: ISO 3 to 8000 in ⅓-stop increments.
Exposure Time: 1⁄8000 sec.–30 min.
Accuracy: Digital: 0.05 f-stop; calibration is traceable to National Institute of
Standard Technology, Washington, D.C.
Memory Recall: Recalls stored reading in memory (M1), second reading in
memory (M2) and/or the ratio, average and f-stop of the two readings.
Instant Recall of All Measurement and Calculations: By pressing the
recall (RCL) switch, the first stored/memorized reading (M1), the
second stored\memorized reading (M2), contrast ratio or bright-
ness (f-stop) difference, average footcandle or average f-stop can be
recalled.
Immediate Update of All Exposure Calculations: If the film speed (ISO),
fps/cine (fps) or exposure time/still (time) settings are changed, the
f-stops are immediately recalculated and displayed in normal, f-stop
difference or average modes.
Memory Erasure: You can erase the stored/memorized readings in memory
(M1) and memory (M2) by pressing the store (STR) and recall (RCL)
switch simultaneously.
Camera Shutter Angle: Complete list of shutter angles and equivalent com-
pensation table conveniently provided.
Bright Green Display: Electro-luminescent lamp for viewing at very low
light levels.
Power Consumption: Operating 5mA/reading; data retention 5uA.
Power Source: 6V battery, #s A544, PX28L, PX 28.
Battery Life: Approximately one year with normal use.
Dimensions: 5½" x 2½" x 2".
Weight: 6 ounces.
Supplied Accessories: Photosphere (for ambient light); photodisc (for key
or directional light); leather carrying case; neck strap; 6V battery; in-
struction manual; warranty card (two-year limited warranty).
| 111
Lenses
by Iain A. Neil
I n the realm of cinematography, lenses form one part of a three-part

camera system that comprises:
i) a light (or radiation) collecting and imaging device – the lens
ii) a light (or radiation) detecting medium – the film1
iii) a storage device – the film itself 2
The above sequence i), ii) and iii) follows the path of light from the object to
the image and storage of many images over time (for later viewing). In essence,
it technically describes the basic function of not only 100-year-old and modern
film camera systems, but also future electronic camera systems. Interestingly, all
of these past, present and future camera systems rely singularly on one image-
forming device—the lens—which predates film and will probably outlast cur-
rent silicon electronics (see Future Lenses section later in this chapter).
Yet the cine lens, with its centuries-old history of development, is gener-
ally not well understood. Indeed, the pedigree of contemporary cine lenses,
bearing a descriptive title that is usually meaningless to anyone but the lens
designer, has become a marketing tool. In a similar vein, “techno” terms
such as modulation transfer function (MTF), which is common in defining
military lenses, electronic and optical system performance, are now used to
describe the overall performance of cine lenses. However, MTF is only one
criteria for measuring cine lens performance; its real value lies in the lens
designer/manufacturer domain.
In the field of cinematography, lens performance characteristics, both tech-
nical and artistic or aesthetic, are quite well known and appreciated to some
degree, particularly by experienced and accomplished cinematographers.
Therefore, considerable effort has been concentrated on providing useful,
practical information regarding cine lenses in this chapter; for example, rules
of thumb, tricks of the trade, special considerations, as well as some basic lens
theory will be covered.
IntroductIon to Lenses In GeneraL

Before getting into any details about cine lenses, it is worth addressing
some of the most commonly asked questions about lenses in general, which
1. Or, in the future, digital image detector, such as a charge-coupled device (CCD).
2. In the future, magnetic tape or magnetic/optical disk containing digital data.
112
|
Lenses
apply both to cine and still-photographic lenses. The most common ques-
tions pertain to the basic formulation and function of lenses. For example:
i) What is a lens?
ii) What does it do?
iii) What is it made of?
iv) Where and how does it fit into a cine camera system?
To answer such questions, Figure 1 depicts a simple lens configuration,
which may be referred to in the following brief discussion of the pertinent
questions and other points of interest.
Essentially, a cine camera objective lens, or “taking lens,” is tasked with
collecting light (or radiation) emanating from object points in an object
space (within a given field of view, which is dependent on lens effective focal
length and size of image format) and forming image points in an image
space; hence, the name objective lens.
Before we answer the above questions, some definitions are required.
Because cine lenses are all tasked with collecting light (or radiation) from
an object space and relaying this to an image space, and because the object
and image are real and the light or radiation is almost always in the visible
spectrum (as seen or nearly seen by the human eye), the following discus-
sion of lenses refers to visible waveband objective lenses. Therefore, all cine
lenses may be classified as objective lenses that collect visible light from a
real object in front of the lens (anywhere from close to the lens to infinity
distance) and form a real image of the light somewhere after the lens.
A lens is an image-forming device, normally refractive but sometimes
reflective, which collects light emanating from a real object and, by virtue of
its refracting or reflecting properties, forms a real image. Lens systems may
be refractive or reflective or a combination thereof. Systems that are either
partly or totally reflective, which are quite popular in long focal-length still-
photography lenses and astronomical telescopes because of their compact-
ness or efficiency at collecting light, are uncommon in cine lenses for one
major reason. A reflective or partly reflective system (with coaxial optics)
depends on at least two mirrors to change or reverse direction of the light
from object space before it reaches a real image (see Figure 2a). To achieve
this, such an optical system, with mirrors aligned on a common optical axis,
must involve a central obscuration so that, in the central portion of light
beams, light is vignetted and not transmitted to the final image (see Figure
2a). At first, this condition might not seem important to cinematography;
however, the aesthetic result can be quite unacceptable to a cinematogra-
pher. To explain this, imagine a night scene with two street lamps, one at
six feet and in focus, and one at 20 feet and considerably out of focus. With
a refractive lens system, the result is as expected: one lamp sharp, and one
lamp soft with a blurred image. However, in the case of a reflective or partly
| 113
Lenses
OBJECT SPACE LENS IMAGE SPACE

+h
+
n n'
h'
u'
u
h
-
-h Where: h = object height, h' = image height
n = refractive index (in object space)
n' = refractive index (in image space)
u = tan (angle) in object space
u' = tan (angle) in image space
And: Lagrange invariant = hnu = h'n'u'

Figure 1a. Single lens definition
CAMERA
FACEPLATE
OBJECT SPACE STOP/IRIS
FILM
OBJECT AT
INFINITY
BACK FOCAL
LENGTH
FOCAL
LENGTH
Figure 1b. Single lens illustration

114
|
Lenses
AXIAL BEAM OF
LIGHT FROM
OBJECT SPACE
IMAGE
SURFACE
Obscuration
blocks central
portion of
light beam(s)
Figure 2a. Reflective lens layout
reflective system, one lamp is sharp and the other is soft, but in the form of a
blurred, donut-shaped image that does not look normal or realistic (see Fig.
2b). Sometimes such a result is visually acceptable, maybe even appealing.
However, for the majority of filming situations, this characteristic or effect of
reflective (sometimes referred to as catadioptric3 lens systems makes them
unappealing to cinematographers).
The question “What is a lens made of?”(now restricted to refractive lens
systems) is answered as follows: Objective lenses, including cine lenses,
comprise one or more lens elements in a series that refract or “bend light” at
their air-to-element surface interfaces. The lens element refractive materials
can be glass, plastic, crystalline, liquid or even chemically vapor-deposited
(CVD) materials (such as zinc sulfide, which transmits light from the visible
to far infrared wavebands).
For the most part, glasses are the predominant refractive medium or
substrate, mainly because their optical and mechanical characteristics are
superior and consistently precise; this is extremely important in the field
of cinematography, where lens-performance requirements are very de-
manding. Therefore, cine lenses are almost always made of glass elements,
3. Catadioptric still-photography lenses that utilize mirror surfaces (flat, spherical or

aspherical), in addition to refractive surfaces, offer the intrinsic advantage of folding
the optical path back on itself twice, making the lens compact in length, which is par-
ticularly attractive in narrow-angle, long focal-length lenses that otherwise would be
large and cumbersome (see Figure 2a). Wide-angle, short focal-length catadioptric
lenses are uncommon because of the presence of large field-dependent aberrations
that are difficult to correct in a mirrored lens system.
| 115
Lenses
SPHERICAL REFRACTIVE LENS SPHERICAL REFLECTIVE LENS

WITH CO-AXIAL SURFACES
Figure 2b. Appearance of out-of-focus bright point objects at night
although occasionally they utilize calcium fluoride4, a crystalline material

(e.g., long focal-length telephoto Canon primes).
To answer the remaining lens question, “Where and how does it fit into
a cine camera system?,” it is first necessary to describe a couple of aspects
about modern vs. old camera systems and how they, together with changing
lens technology (both design and manufacture), have influenced the pro-
gressive design of cine lenses.
In the “old days,” say, the first half of the twentieth century, all lens designs,
including cine ones, had to be kept simple, employing up to five lens ele-
ments or five doublet components (i.e., two elements cemented together).
This was simply due to the fact that anti-reflection coatings did not exist,
thus causing tremendous light loss through a lens. For example, a 50mm
f/2.8 focal-length lens containing five single lens elements and 10 refractive
surfaces would typically experience a 5% loss per surface (i.e., 95% trans-
mission or 0.95 normalized transmission), meaning that for 10 surfaces the
overall transmission might be 60% (i.e., 0.9510). Since T-stop = f-stop ÷ √
(normalized transmission), this 50mm f/2.8 lens would have a T-stop = 3.6.
A f/2.8 lens working at T3.6 does not sound too bad, but consider a 10 or even
20 element f/2.8 lens with corresponding T-stops of 4.7 and 7.8. Fortunately,
cine lenses of this era had one major advantage over later lenses: the film
cameras they were attached to were predominantly nonreflex. Therefore, the
lens could be placed quite close to the film, which made the lens-design task
easier and the lenses less complicated. In fact, it is interesting to note that in
4. The use of calcium fluoride is best avoided in telephoto cine lenses because it is
highly sensitive to temperature changes of even a few degrees Fahrenheit (or Celsius)
and can be expected to produce significant defocusing that may be troublesome in
obtaining and maintaining sharp focus of objects over a short period of time (1–5
minutes).
116
|
Lenses
the case of old wide-angle, short focal-length lenses, their back focal length
was normally smaller than their focal length, which would make them in-
compatible with modern reflex cameras. However, all of the light that is lost
has to go somewhere, and even in the best lens designs, some of it would,
by successive lens-element surface reflections, head toward the film, causing
ghosting and/or veiling glare. To aggravate matters even more, these slow
lenses of T3.6–T5.6 full aperture, coupled with the insensitivity of film stock,
say ASA 50, meant that huge amounts of light were required to illuminate a
scene—good for the lighting supplier but trouble for the cinematographer,
especially in terms of ghosting and veiling glare. Still, the cinematographer
benefitted from one great, indeed overwhelming advantage—a larger depth
of field than he/she is accustomed to now. So these early cine lenses got close
to the film, were necessarily simple in construction (no coatings), and due
to their lack of speed (aperture) performed well (because of good aberration
correction at their full aperture, albeit with careful lighting). A sampling of
these old lens forms is depicted in Figure 3, which includes their well-known
technical or inventor names.
Of course, modern cine cameras are virtually all reflex because of their
need to provide continuous line-of-sight, through-the-lens viewing to the
camera operator. What this means for the lens is that its rear element must
be located some distance in front of the film as predicated by the reflex mir-
ror design of the camera. Fortunately, by the 1950s the previously discussed
transmission problem had been remedied by the introduction of thin-film
technology that ushered in anti-reflection coatings. More complex lens con-
figurations, containing anywhere from ten to twenty elements, were now
considered practical, and the fixed focal-length lens (or prime) suddenly had
a partner—the zoom lens. Both lens types still had to deal with a large back
focal-length distance, but this was now easily managed because complex
lens arrangements were feasible. Even those troublesome wide-angle lenses,
now sitting at a film distance mostly exceeding their focal lengths, could be
relatively easily constructed.
Even though the post-1950s cine lenses were substantially better than their
predecessors, they had one additional demand—faster speed, i.e., greater
aperture. Although film-stock sensitivity had gradually improved, low-light
filming situations had increased, thus requiring cine lenses of full aperture
T1.3-T1.4 and sometimes T1.0 or less. Fortunately, or perhaps with good
timing due to demand, glass technology started to improve substantially in
the 1960s. The first major effect on cine lenses was the realization that those
fast-aperture lenses were now possible due to high refractive index glasses.
However, aberration correction was still limited, especially at T1.3-T1.9 aper-
tures. By the early 1980s, glass technology had improved so much that aber-
ration correction, even in lenses of T1.9 full aperture and, to a lesser extent,
| 117
Lenses
Basic Cooke Modified Cooke
Focal length F less than lens length L Focal length F less than lens length L
Basic Double Gauss Modified Double Gauss
Focal length F less than lens length L Focal length F less than lens length L
Telephoto (or Dyalite) Inverse Telephoto (or Retrofocus)
Focal length F more than lens length L Focal length F much less than lens length L
Petzval
For each lens, object space is to the left,
image space is to the right
(light travels left to right),
and image is formed at surface S.
Lens size or scale is arbitrary.
Focal length F less than lens length L
Figure 3. Basic “old” lens forms

118
|
Lenses
T1.3, was approaching the maximum theoretical limit, even after allowing
for all other lens design constraints such as length, diameter, weight, cost, etc.
Perhaps more significantly, zoom lenses could be designed to perform as
well as prime lenses but were still of greater size, weight and cost. Of course,
it is easy to draw comparisons with the still-photography market, but this is
misleading because the performance requirements of that market are nor-
mally lower than than those of the cine arena. Nevertheless, advancements
in the still-photography lens market are a good indication of where cine
lenses might go. One important area of distinction between still and cine
lenses is in the mechanical design. Whereas still lenses are intended for a
consumer (amateur) market, cine lenses address an industrial (professional)
market. The mechanical requirements placed on the latter dictate greater ac-
curacy, reliability and higher cost than is necessary for the former. Precision
lead screws (or threaded mated parts) have, for some time, been the norm
in prime cine lenses, but they are slowly being supplanted by linear-bearing
technology in some primes and many zooms. Zooms are the main benefac-
tor of linear-bearing technology because they have at least two moving zoom
groups and one focus group, all of high optical power requiring precision
alignment and maintenance thereof. Just like in the still-photography mar-
ket, the cost of all the technologies so far described means that in the field
of cine lenses, zooms are likely to eventually dominate over prime lenses,
except in extreme applications such as very wide-angle, fisheye, or long
focal-length lenses.
Another optical technology in its infancy is the design and manufacture
of cine lenses utilizing aspherical surfaces. These axially rotational, sym-
metrical, nonspherical surfaces, which have been used in infrared waveband
military systems (e.g., thermal imagers) since the 1970s, are only now being
introduced in cine lenses. Manufacturing and assembly techniques have im-
proved to the extent that several cine zoom lenses employing one aspherical
surface, and even one cine zoom lens employing two aspherical surfaces,
are now available for use by the cinematographer. The aspheric technology
utilized in cine lenses should not be confused with that used in inferior-per-
formance still-photography lenses. Extremely high-precision, ground and
polished glass aspherical surfaces are needed for cine lenses to achieve the
high-performance imaging expected, but the much-advertised, aspherically
surfaced still-photography lenses depend on essentially low-quality, low-
cost, molded and replicated lens elements. Many other optical technologies
that are perhaps relevant to cine lenses could be described, such as gradient
index glasses (i.e., GRINS), diffractive or binary surfaces, and holographic
elements, but for the time being aspherical-surface technology is the most
promising (at least until, the next edition of the ASC Manual). Figure 4 il-
lustrates several modern cine-lens optical designs.
| 119
Lenses
To summarize what has been discussed so far, optical technology, along

with camera equipment and customer requirements, have yielded the cine
lenses commonly available today. Getting back to the question, “Where and
how does a cine lens fit into a camera system?,” the modern cine lens, be it
prime or zoom, houses up to twenty-five lens elements and comprises high-
technology glasses, coatings and mechanical components. Focal lengths
from 6mm to 1000mm at working apertures from about T1-32 are avail-
able. All of these lenses fit onto reflex camera systems and generally perform
admirably given all the physical constraints (length, diameter, weight) and
cost imposed on their designs. Derivatives and other embodiments of these
technology-driven lenses are available today in many forms. We will later
discuss them in terms of function instead of technology via movie examples
that indicate their purpose, versatility and limitations.
KInds and types of Lenses

In cinematography, three kinds of lenses are commonly used, providing
wide, medium and narrow fields of view. For the dominant spherical (or
so-called flat) format, 1.85:1, wide lenses have focal lengths from 10mm (or
less) to 35mm, medium lenses have focal lengths from 35mm to 100mm,
and narrow lenses have focal lengths from 100mm to 1000mm (or more).
The above categorization is somewhat arbitrary, but it encompasses not
only the preferred fields of view, but also the majority of lens constructional
forms currently employed.
For the wide-angle lenses (as with the others), the exact constructional
form varies depending on the source (designer, manufacturer, supplier).
However, the form is almost always based on an inverse telephoto (i.e., ret-
rofocus) arrangement of lens elements, which means that the front group-
ing overall is negatively powered and the rear grouping overall is positively
powered. This arrangement is a prerequisite for obtaining the necessary back
focal length or lens-to-film interface distance. Many variations within this
power arrangement have been derived, but they all effectively do the same
thing—suitably increase the back focal length while ensuring image-quality
performance. A modern wide-angle lens design is shown in Figure 4a.
For the medium-angle lenses, the constructional form may vary from an in-
verse telephoto to a double Gauss and even a weak telephoto construction (Fig-
ure 3). This means that the power groupings may be quite different depending
on the exact focal length, as well as size, weight, full aperture, cost, etc. Figure 4b
shows a modern, medium-angle, double Gauss derivative lens design.
The narrow-angle lenses are dependent on “stretched” double Gauss and
telephoto lens power constructional forms, with the latter dominating focal-
length designs of 200mm or longer. Figure 4c gives an example of the former
and Fig. 4d gives a zoom-lens example of the latter.
120
|
Lenses
Anamorphic lenses contain a hybrid construction comprising either a

spherical prime or zoom lens and either a front- or rear-integrated anamor-
phic lens usually containing mostly cylindrically surfaced lens elements.
Due to the size of front anamorphic optics, they are most commonly used on
prime lenses of focal length 25–200mm with full apertures of T2–T2.8; rear
anamorphic optics, being relatively small in size, are usually found on large
telephoto prime and zoom lenses with full apertures of T2–T4. With respect
to anamorphic lens image-quality performance, a good rule of thumb is that
front anamorph lenses are invariably better than rear anamorph counter-
parts even though they tend to be faster, offering full apertures of T2–2.8. In
terms of focal length, anamorphic lenses usually afford twice that of spheri-
cal lenses. However, due to their horizontal squeeze, they still provide the
same vertical field of view, per focal length, as spherical lenses.
Zoom lenses are highly complex, usually having upwards of fifteen ele-
ments. They may be telephoto or inverse telephoto in constructional form,
but many times their form is not distinguishable. Figure 4d shows a modern,
long focal-length, slightly telephoto zoom lens containing twenty-four ele-
ments. An even more complex inverse telephoto, macro-focus zoom lens is
illustrated in Figure 7.
formats and Lenses

There are three film sizes currently in use: 35mm, 16mm and 65mm; the
former two are the most popular. In terms of image formats, there are many,
e.g., 16mm, Super 16mm, 1.85:1 35mm, 2.40:1 (anamorphic) 35mm, 1.33:1
(TV) 35mm, 8-perforation (VistaVision) 35mm, 65mm, 15-perforation
(IMAX) 65mm, to name just a few of the more common ones. All of them
require the film to be transported vertically through the camera (i.e., perfo-
rations traveling in a vertical direction) except for the 8- and 15-perforation
ones, which require the film to travel horizontally through the camera (i.e.,
in a direction perpendicular to the other formats). The direction of move-
ment is in itself not important to the lenses used, but the large difference
in image format or negative area profoundly affects various aspects of the
lenses to be used. In fact, the difference in area, or more exactly the diagonal
dimension of any image format, is the parameter that directly relates to what
the lens needs to cover in terms of field of view.
It must be unequivocally stated that the anamorphic format is defined as
2.40:1 and not 2.35:1, which has been popularized by well meaning but incor-
rect individuals or companies. To be exact, the universally used anamorphic
film format is 2.39:1 at the film, which translates to 2.40:1. This anamorphic
ratio has been around for nearly 30 years, and it is remarkable that other for-
mat ratios like 2.35:1 keep popping up to describe widescreen anamorphic
movies. To be absolutely clear, the 2.35:1 anamorphic format first appeared in
| 121
Lenses
OBJECT SPACE IMAGE SPACE
INVERSE
a. TELEPHOTO WIDE ANGLE LENS 21mm, T1.9
40mm
MODIFIED MEDIUM TO LONG

b. DOUBLE GAUSS FOCAL LENGTH LENS 75mm, T1.9
40mm
HIGHLY MODIFIED MODERATELY LONG

c. DOUBLE GAUSS FOCAL LENGTH LENS 150mm, T1.9
40mm
LONG FOCAL LENGTH CLOSE FOCUS

d. TELEPHOTO ZOOM ZOOM LENS 135mm - 420mm, T2.8
100mm
FOCUS
ZOOM ZOOM
FOCUS
Figure 4. Variety of modern lens designs (29mm image size)

122
|
Lenses
the mid-1950s but died out by the early 1970s. Just consider that over the last
thirty years, the vast majority of 35mm format anamorphic movies have been
shot using anamorphic lenses with a 2X horizontal squeeze covering a 1.2:1
film negative format, giving 2.40:1 presentation. The originator of the ana-
morphic format, CinemaScope, in the early 1950s defined it as an even larger
2.55:1, but this quickly disappeared when the 2.35:1 format was introduced.
By the early 1970s, these anamorphic formats and others were superceded by
the widely adopted 2.40:1 format, which became and is presently the de facto
standard in anamorphic cinematography (see SMPTE 59-1998).
Before getting into format and lens specifics, it should be mentioned
that detailed information about format image size, area, etc., can be found
elsewhere in this manual (see Cinematographic Systems chapter). Also, to
discuss the effect of specific formats on lenses, it is necessary to explain some
elementary theory about film formats and lenses. Referring to Figure 5a, it
can be seen that if the same focal-length lens, set at a constant aperture, is
used in three widely differing image format diagonals, then the fields of view
are entirely different. Now, let’s say the focal lengths of the lenses (still at a
constant aperture) are selected for constant fields of view as shown in Figure
5b. Then, upon projection of each image (after processing to a print) on a
constant-size viewing screen, it would be apparent that the in-focus objects
would look the same. However, for out-of-focus objects it would be clearly
apparent that the depths of field are quite different. This result is extremely
important to the cinematographer, not only because of the artistic impact,
but also because apart from changing the lens focal length and hence field
of view and perspective, nothing can be done to the lens design to alter this
result. The optical term “Lagrange invariant” (an unalterable law of phys-
ics), has been defined (see Figure 1), and the aforementioned result is a
direct consequence of it. In Figure 5b, its controlling effect on field of view
(perspective), focal length and depth of field vs. image format size are self-
evident. Only one real option is available to the cinematographer to alleviate
this condition or even solve it—change the lens aperture. This seems quite
simple until the practicalities of actual shooting are fully considered. How
can you shoot in the 65mm format or, for that matter, the 16mm format and
achieve the same look as for the 35mm format? Some remedies (not cures)
can be implemented, and they are best understood by taking examples from
old and new 65mm-format feature films. It should be understood that be-
cause the 65mm format intrinsically has less depth of field than the 35mm
format for lenses of equivalent field of view, an abundant use of lighting
combined with stopping down the lens enables a similar image to be realized
(see Lawrence of Arabia, Dr. Zhivago and Ryan’s Daughter all shot by Freddie
Young BSC). Also, diffusion filters can help to increase the apparent depth
of field, albeit with some loss of image sharpness. Another option to help the
| 123
Lenses
a. Constant FOCAL LENGTH and f/2 aperture
½ Horizontal Focal Spherical

Field of View Length Film Format
(degrees) (mm) (mm)
½ Horizontal Focal Spherical
50 Field of View Length Film Format
16 (degrees) (mm) (mm)
6.0˚ 65
50
27.8˚
50 35
11.8˚
b. Constant FIELD OF VIEW and f/2 aperture
1/2 Horizontal Focal Spherical

Field of View Length Film Format
(degrees) (mm) (mm)
50
16
6.0˚
100
35
6.0˚
250
65
6.0˚
Figure 5. Formats and lenses.

124
|
Lenses
65mm-format depth of field in certain scenes are slant-focus lenses (see the
bar-top bar scene in Far and Away, shot by Mikael Salomon, ASC). In com-
parison, for the 16mm format the greater depth of field is more difficult to
correct since this implies even faster lenses, which are not available because
there is a minimum f-stop that cannot be gone below, f/0.5. Therefore, the
only real solution for the 16mm format is to forego the preferred field of
view and corresponding perspective by changing the focal length of the lens
or working at lesser object distances. Using hard lighting is another approach
that helps somewhat with 16mm format depth of field, but the overall look
may suffer. Electronic enhancement in postproduction is another possibility,
but again, the overall look may suffer.
To conclude, it is fair to say that for 35mm and 65mm film formats, just
about anything can be successfully shot as long as one is willing to accept
the costs involved for, say, lighting. For smaller formats, such as 16mm film
or high-definition video cameras (with ⅔-inch detectors), the main limita-
tion is too much depth of field in low-light-level situations where the lens
aperture cannot realistically and practically be less than T1.2–T1.5. Only
faster lenses or a retreat to larger formats, be they film or electronic, will
completely solve the depth of field issue. Of course, what is or is not deemed
acceptable in terms of the depth of field or look of the picture has so far
been determined by what is expected. In other words, it is highly influenced
by past results. Future results, especially with digital-video cameras and
lenses, might look different, and over time might become quite acceptable.
So maybe the depth of field concerns will disappear. In the meantime, the
Lagrange invariant, just like Einstein’s theory of relativity, cannot be broken,
so lens depth of field, perspective and look is inextricably linked to and gov-
erned by the format size.
Anamorphic vs. spherical depth of field will be covered later in this chap-
ter. Also, the deliberate omission of the circle of confusion in the preceding
discussion about depth of field is because it has no bearing on different film
formats that have similar resolution capabilities, especially when viewed
eventually on a cinema screen. The circle of confusion is a purely mathemati-
cal value used to determine an estimate of expected or effective or apparent
depth of field, but that’s all, and it should only be used for that purpose.
spherIcaL vs. anamorphIc Lenses

Here we will discuss some of the differences between spherical and ana-
morphic lenses. There will be no attempt to suggest or imply which of these
formats is better. Only lenses for the 35mm cine format will be described, but
the same observations apply to other formats. Unlike the Super 35 pseudo-
anamorphic format, both the true spherical (1.85:1) and anamorphic (2.40:1)
35mm formats require no anamorphic opticals (i.e., special printer lenses)
| 125
Lenses
in the process of film negative to release print. Since the late 1950s, the ana-
morphic film format has been about 59% greater in negative film area than
the spherical 1.85:1 film format. An often-asked question is, what happened
to the original CinemaScope anamorphic lenses? Interestingly, the word
“scope” has survived to this day, even though the terms spherical (i.e., flat)
and anamorphic (i.e., widescreen) are best suited to describe the format dif-
ference. There are many reasons, mostly economic or business-related, as to
why CinemaScope lenses disappeared by the mid-1960s. Some aspects of the
early lenses did have technical deficiencies, and these are worth expanding
upon.
Early anamorphic lenses produced one particularly disconcerting, focus-
related image characteristic, which caused several problems with both actors
(especially actresses) and the camera crew. The problem was “anamorphic
mumps,” a well-known term coined by movie production people. A good
example of this is to consider an actress (speaking her lines) walking from,
say, 20 feet to 5 feet (i.e., full-body to facial shot) while the camera assis-
tant or focus puller does a follow focus to keep her in focus at all times.
Assuming a high-quality anamorphic prime lens was used (CinemaScope
circa 1955-1965), the face of the actress would naturally increase in size as
she approaches the camera. However, due to lens breathing through focus
(explained in detail later), and more specifically anamorphic lens breathing,
not only did the face of the actress increase in size, but also the face would
become fatter at the close focus. So the breathing effect, or increase in size of
the object, is much greater in the horizontal direction as compared to verti-
cal direction. Obviously, actors were unhappy about this phenomenon, so
early anamorphic pictures had close-up shots at 10 feet instead of 5 feet (to
alleviate the effect). Production companies and camera crews, particularly
the cinematographer, did not like this, because with the old, slow film stocks,
a ton (for lack of a better word) of lighting was required, and the perspective
of the shot was not as it should be. Heat from the vast lighting required also
produced problems, like sweat on the actors’ faces and makeup meltdown.
In 1958, a U.S. patent was issued for an anamorphic lens design that virtu-
ally eliminated this problem, and anamorphic lenses utilizing the patented
invention have been used continuously for more than forty years. They are
the predominant anamorphic lenses used to shoot the majority of widescreen
movies to this day. The importance of these new anamorphic lenses was ex-
emplified by the fact that Frank Sinatra the main actor, in the movie Von
Ryan’s Express, shot by William H. Daniels, ASC, demanded that these lenses
be used. Before leaving this subject, an interesting piece of historical informa-
tion: The first prototype anamorphic prime lenses with reduced “anamorphic
mumps” were used in the 65mm-format film (2.75:1 with 1.25x anamorphic
squeeze) Ben Hur, released in 1959 by MGM and shot by Robert Surtees,
126
|
Lenses
ASC. A little-known anecdotal fact about these anti-anamorphic mumps

lenses is that they can be specially designed to squeeze, or thin, an actor’s face
in close-ups; indeed, this was soon requested by a particular actress (who
shall remain nameless) whose face had fattened with age.
An often-asked question about anamorphic vs. spherical lenses is, “What is
the depth of field for each?” For a spherical lens, the answer is simple: look up
published depth of field tables (provided in this manual; see page XXX). For
an anamorphic lens, the answer is complicated; firstly, there usually are not
any published depth of field tables, and, secondly, the depth of field is different
depending on whether it is measured in the vertical or horizontal directions
of the film format. What does this mean in reality? Taking an example of,
say, two 50mm focal-length lenses, one spherical and one anamorphic, both
set at the same aperture, two anamorphic scene compositions can be cre-
ated, whereas only one is available spherically. Rather than dissect all possible
compositions of a scene, a good rule of thumb is to allow for half the depth
of field when comparing anamorphic to spherical lenses of equal vertical
or horizontal fields of view and aperture. Although this covers a worst-case
scenario of fitting an anamorphic scene horizontally into a spherical-format
width, it does provide a good safety margin for all filming scenarios.
In terms of absolute or theoretical image quality and overall aberration
correction, there is no doubt that spherical lenses are capable of superior
performance over anamorphic lenses. However, in terms of practical image
quality or what will be eventually viewed on the screen or on smaller presen-
tation mediums (e.g., television), both easily provide adequate performance.
For theatrical presentation on a fixed-width cinema screen, the anamorphic
format will have a distinct advantage over the spherical format because there
is more depth of focus at the film print in the projector, which means that
constancy of film print position in the projector is less critical. Another
consideration relating to image quality or residual aberration correction dif-
ferences in spherical as opposed to anamorphic lenses is integration of visual
effects (optical and computer-generated). Part of the reason why the Super 35
format (spherical-lens origination, anamorphic release) has recently become
popular, even though the film negative is small—65% smaller than pure
anamorphic 2.40:1—is because the aforementioned low residual aberrations
may aid the effects community. Considering various aspects of spherical vs.
anamorphic-lens residual aberrations, the latter lens tends to produce more
field curvature, astigmatism and distortion. Because of this, the spherical lens
has a sweet spot (i.e., excellent image quality) with a diameter roughly equal
to 90% of the width of the format. In comparison, the anamorphic lens has
an elliptical sweet-spot area bounded vertically within 80% of the format and
bounded horizontally within 90% of the format (see Figure 6a). What this
means practically for the cinematographer is that objects of principal impor-
| 127
Lenses
SPHERICAL 1.85:1 FORMAT ANAMORPHIC 2.40:1 FORMAT
90% of total 90% of total

width width
80%
Sweet of total Sweet
Spot height Spot
a. Spherical and Anamorphic Lens Sweet Spots
OR
b. Spherical and Anamorphic Out-of-Focus Bright Object Points at Night
Figure 6. Color sampling as a compression of image data
tance are best kept within these sweet spots (at lens apertures approaching
full aperture). It should also be noted that all lenses, spherical and anamor-
phic, tend to perform best beginning at an aperture stopped down by at least
one from their maximum aperture opening and up to, say, an aperture of T11
to T32 depending on lens focal length (i.e., T11 for very short focal-length
lenses, T32 for very long focal-length lenses).
Until quite recently, a particular problem associated with anamorphic
prime lenses has been their limited ability to provide good-quality imaging at
close focus distances. In fact, all lenses, spherical and anamorphic, are usually
designed to perform best at one distance, and gradually lose image-quality
performance toward infinity focus and especially at close focus. Modern zoom
lenses, spherical and anamorphic, are less afflicted by this problem because
they incorporate complex, usually multiple, internal focus lens groups (de-
scribed later). In the case of anamorphic prime lenses (as opposed to spherical
prime lenses, which do quite well in this respect), there has always been a
trade-off with regard to lens size, weight and image quality at close focus (6 feet
to 3 feet). Indeed, compact, lightweight close-focusing anamorphic lenses of
fairly low image quality have been around since the 1960s, but until recently
no anamorphic lenses, large or small, could provide good image quality over
128
|
Lenses
a focus range from infinity to 2–3 feet with low veiling glare characteristics. In
the mid-1980s, technological advances in lens coatings and fabrication tech-
niques brought spherical prime lenses with the above attributes to the mar-
ketplace. Later, cinematographer (now director) Jan DeBont, ASC suggested
that the combination of these spherical prime lenses with the best anamorphic
optics might produce anamorphic prime lenses of high image quality and
low veiling glare that would compete favorably with the best available spheri-
cal prime lenses. Such anamorphic lenses were produced and first used on
the movie Flatliners, shot by DeBont. Still, the close-focus anamorphic lens
image-quality problem had to be solved. The source of the solution turned
out to be using developments in spherical zoom-lens cam technology. Using
precision cams in modern anamorphic lenses, still based around that 1958
anamorphic lens patent invention, high-quality-imaging anamorphic prime
lenses with substantially reduced veiling glare and close focusing down to 21⁄2
feet or less were produced and are commonly used in the marketplace. They
are still somewhat large and heavy, but they provide the main and preferred
image-quality characteristics required for modern filmmaking.
Since anamorphic (widescreen) and spherical (flat) movies are no longer
identified before being shown in theaters, and many times are not even pre-
sented in their full format ratio, what sets them apart lens-wise? Depth of field
differences previously mentioned can be looked for. Streaking of hot (bright)
objects, especially point objects such as a kick of sunlight off the chrome trim
of a car, is usually more pronounced when anamorphic lenses have been
used. Two good examples of films that purposefully use this characteristic to
intensify the action are Close Encounters of the Third Kind,5 shot by Vilmos
Zsigmond, ASC, where the small alien spaceships with bright lights fly low
along a twisty road at night; and Speed, shot by Andrezej Bartkowiak, ASC,
where the underground runaway train with bright headlights hurtles along
the tracks just before crashing. In both of these scenes, the streaking can be
seen as blue, red, purple and white bright lines, mainly horizontally spread
across the picture but also at other angles. To reduce streaking where bright
point sources are unavoidable, one technique is to introduce a tiny amount
of diffusion, say, by single fog or black Pro-Mist types of filters. Perhaps the
definitive giveaway of an anamorphically lensed movie appears in a scene
shot at night with a streetlamp well out of focus. In a spherical movie the
lamp will be blurred and circular in shape, but in an anamorphic movie the
lamp will be blurred and elliptical in shape. (See Figure 6b.)
In summary, the pace of development in spherical vs. anamorphic prime
lenses has been greatest in the former, and yet the latter has benefited greatly
5. In some scenes, the streaking was introduced by visual effects using 65mm format
lenses.
| 129
Lenses
from some of the former’s developments. The image-quality performance

differences and depth of field considerations have been reduced from a tech-
nical level to that of artistic or aesthetic preference, which is good, because
it involves interpretation of the storyline. To elaborate on this, it is worth
considering a movie example concerning the practical issue surrounding
anamorphic vs. spherical-lens depth of field. Dean Semler, ASC, ACS shot
the feature film Dead Calm in the anamorphic format. Given the confined,
if not claustrophobic, circumstances of filming almost entirely on a yacht on
the open sea with many close focus distances, one might question the reason
for choosing the anamorphic format. However, Semler had good reason: the
cinematographer wanted to emphasize what was happening on the yacht
during the movie, and not distract the viewer by drawing attention to things
in the distance or not relevant to the scenes being enacted.
To conclude, it may seem that an excessive amount of attention has been
placed on spherical vs. anamorphic lenses, but we wanted to thoroughly ex-
plain why the choice is artistic as well as technical.
prIme Lenses vs. Zoom Lenses

Earlier in this section, we discussed various technological developments
relating to lenses in general. All of these and other developments have been
instrumental in bringing zoom lenses to a point where today they rival or
exceed prime lenses in virtually every aspect, be it cost, performance or
versatility; their only drawbacks are still size and weight, but even those are
being gradually eroded. In cinematography, where optical and mechanical
performance requirements and expectations are traditionally higher, zoom
lenses have seen a much slower acceptance than in other applications,
(e.g., still photography—especially point-and-shoot cameras—photocopi-
ers, video camcorders, etc). In the late 1950s to early 1960s, zoom lenses
gained popularity with cinematographers mainly for their intrinsic zooming
capability. Today, the most technologically advanced cine zoom lenses have
become popular not only because of their zooming capability, but more
importantly for their performance and versatility. They are used equally for
zooming and as variable focal length primes.
Prime cine lenses, having a huge range of discrete focal lengths from about
6mm (fisheye) to 1000mm, full apertures starting at about T1.2 and close
focusing to 1 or 2 feet, are commonplace. The overall performance of these
modern prime cine lenses is usually excellent. In general, if you want more
from a prime lens you can get it; you just have to pay more. The performance
that can be achieved now closely approaches limits set by the laws of physics.
Of course, good prime cine lenses still need proper optical design, mechani-
cal design, manufacturing, assembly, testing and calibration, no matter how
much development money is invested. Zoom lenses are controlled by the
130
|
Lenses
same factors. However, zoom lenses, apart from their zoom capabilities, can
and already do offer considerable advantages over even the best prime lenses.
Primarily due to their greater cost but far better return on investment, as
well as their greater overall complexity, zoom lenses can readily incorporate
advanced features, such as close-to-macro focusing (with virtually constant
aperture throughout focus and zoom) and reduced breathing at short focal
lengths. In particular, the breathing control offered in some modern cine
zoom lenses is important to many cinematographers.
Optical breathing is a phenomenon peculiar to cameras that continually
record images over time, i.e., film or video. It is not present in still photog-
raphy. Breathing is well illustrated by considering a scene containing two
persons talking intermittently with each other, one at 6 feet and one at 20 feet
focus. Let’s say the person at 20 feet (in focus and at the edge of the scene)
first talks to the person at 6 feet (slightly out of focus but quite discernable,
centered in the scene). Then let’s say that during the conversation, the person
at 6 feet (now in focus by refocusing the lens) starts talking to the person at 20
feet (now slightly out of focus), but the person at 20 feet, due to refocusing the
lens, moves out of the scene. This means that the person at 6 feet is talking to
nobody in the scene, thus ruining the take. In other words, breathing, through
change in field of view during focusing, has moved objects at the edge of the
scene into and out of the scene. Patents of zoom-lens inventions dating back
to the late 1950s have addressed this problem, and several modern cine zoom
lenses, sometimes using complex internal focusing arrangements, have suc-
cessfully minimized this effect—especially at short focal lengths, where the
larger depths of field make this effect more noticeable.
Modern cine zoom lenses are available in a variety of forms—lightweight
(2-3:1 zoom ratio), variable prime (2.5:1 zoom ratio) and conventional.
The latter comprises medium ratio (3:1–5:1), wide-angle and telephoto
zoom lenses, e.g., 14.5–50mm T2.2, 17.5–75mm T2.3, 20–100mm T2.8,
135–420mm T2.8, 150–600mm T6.3, and large-ratio (10:1-11:1) wide to long
focal length zoom lenses, e.g., 24–275mm T2.8, 25–250mm T3.5–4.0.
To conclude, prime and zoom lenses complement each other in the field of
cinematography, and virtually all movies now feature prime lenses and one
zoom lens. Many movies use prime lenses and more than one zoom lens;
some movies use only a few prime lenses and are shot almost entirely with
zoom lenses.
specIaLty Lenses and systems

In addition to prime and zoom lenses, the cinematographer has at his or
her disposal a large variety of other lenses or lens systems (which contain
primes or zooms) that provide different imaging characteristics or other
features: specialty lenses. Some of these specialty lenses may be just prime
| 131
Lenses
OBJECT SPACE IMAGE SPACE
FOCUS GROUPING ZOOM GROUPING
FRONT FOCUS REAR FOCUS FRONT ZOOM REAR ZOOM

GROUP GROUP GROUP GROUP
OBJECT TO
FRONT EFL (mm)
VERTEX AT INF.
DISTANCE FOCUS
(MAGNIFICATION) STOP/IRIS
14.5
INFINITY
(INF:1)
COMPENSATOR VARIATOR
LOCUS LOCUS
23.9
208mm
(12:1)
ASHPHERICAL
SURFACE AUXILIAY GROUP WITH
ASHPHERICAL SURFACE
50.0
60mm
(3:1)
50mm

132
|
Lenses
lenses with unusual features, and some may involve an optical system that
accepts attachment of a variety of primes or zooms.
The Cine Lens list, starting on page 65 3 in this manual, contains some
of the best-known specialty lenses and systems and identifies their specific
characteristics and possible applications. Some of them are dependent on
folded optical configurations utilizing mirrors or prisms. They are all unique,
but some have overlapping properties. None of them can be construed as
front or rear lens attachments, because they attach directly to the camera.
By far the most significant aspect of these lenses and optical systems is
their ability to achieve in-camera real-time shots not possible with regular
primes and zoom lenses. Other advantages include provision of large depth
of field, extreme close or even macro focusing, and maneuvering among
objects (e.g., miniatures, models, forced perspective).
Some good examples of their shot-making capability can be seen in the
following movie and TV sequences. In Titanic, cinematographer Russell
Carpenter, ASC and visual-effects supervisor Erik Nash used a Panavision
/Frazier lens system and camera, each under motion control, to shoot the
beginning of the last sequence in the movie. Shortly after the woman drops
the gemstone into the ocean from the back of the research ship, a dry-for-
wet scene commences with the camera system approaching a model of the
sunken Titanic hulk (in dark blue lighting), then traversing over the bow of
the ship heading toward the port side, then entering and traveling through
a covered outside walkway, and eventually slowing to a halt after turning
left to see two doors with unlit glass windows, which then are lit and open
to reveal a congregation of people collected together to toast the lead actor
and actress. In this shot, which is far too complicated to describe fully (al-
though it is worth noting that CGI and a Steadicam rig are also involved),
the large depth of field, image-rotation control and pointing capability of
the Frazier lens system are utilized up to the point where the doors open.
Another movie, The Rock, shot by John Schwartzman, ASC, exemplifies the
variety of shots that can be accomplished with the Frazier lens system and
some other specialty lenses. Many of the shots are seen toward the end of
the movie, when Nicolas Cage is being chased on the Alcatraz prison walk-
ways while carrying the deadly green marbles and accidentally dropping,
then grabbing, them on the parapet of the lighthouse tower. In this shot,
the macro focus (close-up of his feet) to the infinity focus (San Francisco
distant skyline) carry of focus (i.e., huge depth of field with no follow focus)
is clearly illustrated. For periscopic specialty lenses, a good example can be
seen in the introduction to public television’s Masterpiece Theatre series,
where the point of view given has the lens working its way through table-top
memorabilia in a Victorian-era drawing room, eventually halting in front of
a hardbound book with the title of the play about to be presented on its front
| 133
Lenses
cover. This particular shot combines the maneuvering, pointing, role and
close-focus capabilities of the Kenworthy snorkel.
The above examples relate to specialty lens systems where different objec-
tive or taking lenses, primes and zooms, can be attached to an optical unit
which basically relays the light through mirrors and prisms to form a final
image, and wherein pointing and image-rotation means are housed. Many
other lenses or lens systems, some under remote control, such as the pitch-
ing lens, can be used to achieve similar results.
Other specialty lenses, such as slant-focus lenses and bellows-type lenses,
typically have fewer features than those afforded by the aforementioned
lens systems. However, they do offer the cinematographer opportunities
to capture other distinctive views of a scene. For example, the slant-focus
lens permits reduced depth of field limitations to be overcome in low-light-
level scenes where objects span across the field of view with continuously
increasing focus distance or with continuously decreasing focus distance,
such as is found in the typical car scene with the driver and passenger in
conversation and the camera looking into the car from either the passenger
or driver window. Bellows-type lenses can produce shots similar to slant-
focus lenses, but because of their bellows dependency cannot easily be ad-
justed in shot. However, the real advantage of bellow lenses is their ability to
produce a distorted field of view or perspective. In other words, even though
the film-format shape remains constant, objects can be highly distorted and
defocused differentially across the scene. For example, the well-known THX
rectangular-shaped credit could be made trapezoidal in shape, with each
corner defocused by different amounts.
All of these specialty lenses are currently very popular with cinematog-
raphers. Their attributes are quite well-suited to movie making, and they
are used to great effect in TV commercials to get that difficult-to-obtain or
different look.
Lens attachments
The most common cine lens attachments are placed before or after (and
occasionally within) a cine camera lens, be it a prime or zoom, spherical or
anamorphic.
Front lens attachments include diopter and split diopter close-focusing
lenses, image stabilizers, image shakers, distortion optics and inclining
prisms. Their main duty is to maximize the versatility of the lens to which
they are attached. Most of these front attachments lose some light transmis-
sion and, in some instances, image quality, but their overall effect proves
them to be worthwhile even though they have deficiencies.
Rear lens attachments mainly include extenders (to increase or decrease
the overall lens focal length) and mesmerizers (rotatable anamorphic optics).
134
|
Lenses
Extenders are well established in still photography. Their main property is to

increase (or even decrease) the overall focal length of the lens. There are
three disadvantages when using them: overall image quality may be some-
what degraded; aperture is reduced at the rate of one stop per each 1.4x mul-
tiplier (e.g., x2.0 extender = 1.4x1.4x, giving a two-stop loss); and the focus
scale of the lens will be incorrect by approximately the physical length of the
extender. In practical terms, the rule of thumb is that a good high-quality
1.4x rear extender will not substantially affect the overall image quality, but
a 2x or greater extender will probably cause a significant reduction in image
quality. Technically, rear extenders should only be employed if no alterna-
tive prime or zoom lens is available. Mesmerizers, due to their anamorphic
nature, may degrade image quality even further. In TV commercials this is
not a problem, but in moviemaking some caution should be exercised, such
as stopping down the aperture at least one stop.
One front lens attachment, which is quite commonplace in the video-lens
marketplace but has only recently been introduced in cine lenses, is the sin-
gle or double aspherically surfaced element. The idea here is to take a single,
negatively powered lens element with one or two nonspherical surfaces and
attach it to the front of a wide-angle prime or zoom lens so that the focal
length is decreased to 80% (or even as far as 66%) of its original value; cor-
respondingly, the field of view is proportionally increased, the image quality
is little changed. It should be pointed out that such attachments only work
properly with prime or zoom lenses that have a minimum full field of view
of about 75°. These front lens attachments offer an economical way to pro-
vide increased wide-angle, short focal-length lenses not available in a lens
series and, by virtue of their aspherically surfaced lens design prescription,
astigmatism, distortion and lateral chromatic aberrations are kept under
control. The only real disadvantage of this wide-angle lens approach is the
loss of focus capability, because the front attachment requires the principal
lens to be set at the end of its close-focus range to reach the hyperfocal-focus
distance of the overall system. However, such wide-angle lenses’ large depth
of field does not really warrant any focusability. Examples of these front lens
attachments currently being used include a 6mm T2 prime covering the
Super 16mm film format, which becomes a 4.5mm prime (using a 0.75x
front lens attachment), and a 14.5mm T1.9 prime covering the Academy
35mm film format, which becomes an 11.5mm prime (using a 0.8x front
lens attachment). Due to the simplicity of these front lens attachments, no
significant stop loss is associated with their use.
format sIZe and Lens attrIbutes

Earlier, in Formats and Lenses, lens depth of field was discussed. There are
some other important implications resulting from format size on lens size
| 135
Lenses
(volume and weight), image quality and aperture, as well as reflex mirrored
camera considerations.
Perhaps the most noticeable effect is the tendency for lens size to increase
almost linearly with each format size, assuming a constant focal length and
constant full aperture. Of course, this must partly happen in reflex camera
systems because the reflex mirror, size and position dictate the lens back-
focal length. Since the size and distance of the mirror from the film (or
image plane) increases mainly according to film-format size, this means that
65mm-format lenses tend to be larger than 16mm-format lenses, at least
at their rears. However, the greater difficulty in covering a larger format,
combined with the need for larger depth of field, makes 65mm-format
lenses slower, T2.8–T5.6 full aperture, whereas 16mm format lenses are
faster T1.4–T2.0. Unlike in still photography, where a larger-format lens is
chosen to provide a larger print, in cinematography the 65mm, 35mm and
occasionally 16mm prints are invariably shown on similarly sized cinema
screens, roughly 20–50 feet wide. In summary, 65mm-format cine lenses
are optically less complex, slower and have slightly less resolution per mil-
limeter of film negative than 35mm- or 16mm-format lenses. Nevertheless,
65mm lenses perform as well as they need to, based on the size of the format.
The key to the final presentation quality using large-format lenses, including
65mm, VistaVision and 35mm anamorphic, is, as it always has been, the
larger format size and corresponding area of the negative.
specIaL appLIcatIons of Lenses

Special applications fall into the category of visual effects, which include
special effects, computer-graphics imaging (CGI), digital effects, animation,
etc. Most of these applications have one thing in common: they require
principal photography with lenses and film to achieve about 2,000 pixels
of usable information (and more recently up to 4,000 pixels) across the film
format. Today, the most-used and best cine lenses perform adequately and
can meet this requirement. Even anamorphic lenses, which get a bad name,
are not pressed in meeting this requirement. In fact, until quite recently, film
stocks, because of their chemical image processing for enhanced contrast in
the negative, have proved to be more problematic than lenses, especially in
blue or green matte screen applications.
Although modern cine lenses may be considered precision optical instru-
ments that were designed, manufactured, tested and calibrated in a highly
scientific manner, many of their characteristics, features and specific proper-
ties have not been made available publicly or accessible to people working in
special applications. In the last five to 10 years, books or manuals containing
some of this information have appeared; however, further dissemination of
lens information is necessary.
136
|
Lenses
To explain what some of this lens information is, a good method is to

examine a recently made movie with many special effects, such as Titanic.
After camera equipment was supplied to the production, the crew requested
certain lens information. In particular, variations of distortion and field of
view (see breathing earlier described) through focus and zoom and lens re-
peatability were topics that kept arising. Eventually, this and other kinds of
information were released so that the desired special effects could be realized.
With the ever-increasing utilization of visual effects, a tremendous amount
of lens information will be not only required, but also indispensable in the
making of movies. Most of this information is not yet publicly available.
Lens depth of fIeLd

Depth of field lens information is worth a special mention because until
quite recently, its method of calculation, especially at close- to medium-
focus distances (say 2–6 feet), has been somewhat inaccurate for wide-angle
prime and zoom lenses; and its interpretation for wide-angle lenses in gen-
eral needs clarification.
For modern, high-quality, wide-angle cine zoom lenses that are typically
about 6–12 inches long, the lens length cannot be ignored when calculating
the depth of field. To properly calculate the depth of field, the first nodal, or
conveniently entrance pupil position of a wide-angle lens must be known.
In most wide-angle zoom and prime lenses they are normally a few inches
back from the front of the lens. So assuming that properly calculated depth
of field tables for the lens in question are not available, a good rule of thumb
for a short focal-length lens is to subtract the physical lens length (from front
to flange/mounting point) away from the object distance (i.e., object to film),
and then look up the depth-of-field table. It goes without saying that a lens
at a close focus distance of 4 feet (object to film), after the subtraction of, say,
one foot, will produce a significant difference in the depth of field from that
expected—just look at any contemporary depth of field table to compare 4
feet and 3 feet at T2 for a lens of any focal length.
One common complaint about the depth of field observed for wide-angle
primes or zooms: when the focus distance is 6 feet and the expected depth
of field (according to the tables) is 4 to 9 feet, why is the defocused object
at 4 feet almost as sharp as the focused object at 6 feet, while the defocused
object at 9 feet looks comparatively very soft? The answer has nothing to do
with the lens performance, the validity of the depth-of-field tables, or even
lens physical length; it simply has to do with the magnification of objects at
different focus distances. Since the object at 4 feet is only 11⁄2X larger than
the object at 6 feet, but the object at 4 feet is 2¼X larger than the object at
9 feet, then for the same object, (say, a person’s face at all three focus dis-
tances, and all faces lit in a similar manner), it is to be expected that the
| 137
Lenses
aforementioned result will appear. In addition, it is worth noting that the

introduction of even the weakest diffusion filters, as well as soft lighting, will
further compound and exaggerate this condition.
Lens desIGn and manufacture

To the various suppliers of cine lens products, the design and manufacture
(including assembly, test and calibration) of cine lenses have been, and still
is, considered proprietary, as in intellectual property. A survey of cine-lens
patents from the 1950s through today (few in number) confirms the secre-
tive nature attached to cine-lens products by various suppliers. Quite a few
technical and scientific publications on some of the above aspects of cine
lenses are available and are listed under References toward the end of this
manual. Most of them concentrate on the optical design of cine lenses.
Over the years, several different approaches have been used to provide suit-
able cine lenses to the cinematographic marketplace. More often than not, the
best approach has been to design custom cine lenses from scratch. In this case,
two substantially different lens development approaches have been chosen.
The first, which for the sake of discussion will be called the old-world method,
entails a company performing virtually all tasks in-house. The second, called a
new-world method, has the design and development process controlled at one
company (with the final assembly, testing and calibration performed there),
but a large number of specialized suppliers being subcontracted for partial
design and component manufacture—sort of like the way modern jet aircraft
are produced. Both approaches have advantages, the first having everything
totally controlled under one roof, and the second being better suited to exploit
new technologies. In terms of which approach produces the best cine lenses,
only the results can provide an answer. However, the latter approach is best
suited to providing faster development of new lenses.
In such a limited space it is not realistic to discuss all the aforementioned
aspects of cine lenses. The references we have mentioned, although mainly
containing theoretical information, such as MTF performance, are very use-
ful in gaining a fairly comprehensive understanding of optical aberrations
and numerous other optical (e.g., veiling glare) and mechanical (e.g., use of
linear bearings) considerations in the design of cine prime and zoom lenses.
We suggest that the reader peruse the listed references for a more detailed
account of this subject.
Lens erGonomIcs
Unlike most still-photography lenses, which are purely driven by cost vs.
image-quality requirements, cine lenses have a strong ergonomic compo-
nent. This component is indirectly dependent on what the cinematogra-
pher expects of a cine lens, but is directly related to what his or her crew,
138
|
Lenses
specifically camera assistants, must do to obtain the desired shot. Since the
assistant’s job may depend on ergonomic aspects of the chosen cine lens, this
little-discussed lens aspect is certainly worth noting.
It is often believed that the quickness of sharp focus being attained in a
cine lens by rotation of the focus gear means that a sharp in-focus image
of the object has been achieved. However, the opposite is true, because it
is best (and this has been conclusively tested and will be verified by most
camera assistants) for the lens to have a lengthened focus scale with many
spaced-apart focus calibration marks (e.g., infinity, 60, 30, 20, 15, 12, 10, 9
feet, etc., continuing down to close focus) so that the sharpness of focus may
appear more difficult to reach, but the likely error in focus is reduced. Cam-
era assistants say that this ergonomic lens aspect of providing an extended,
well-spaced and clear focus scale, usually on a large lens barrel, is crucial to
providing a well-focused, sharp image.
The aperture or T-stop scale is also important, because it should have clear
markings (e.g., T2, 2.8, 4, 5.6, etc.) spaced well apart and preferably linearly
spaced for equal stop differentials. The ergonomics of the latter condition
have become more important lately due to motion-control-powered iris
practices including variable, aperture and frame-rate shots.
Zoom scales were until recently largely ignored in terms of the focal-length
markings and spacing on the lens zoom scale. On modern zoom-lens scales,
the normal cramping of long focal-length marks has been avoided through
optimization of zoom cam data so that long focal-length zoom mark spacings
have been expanded and short focal-length zoom markings have been com-
pressed (but not to adversely affect selection of short focal-length settings).
Most modern prime and zoom cine lenses can be expected to offer ex-
panded focus, aperture and zoom scales on both sides of the lens, i.e., dual
scales. Also, all of these scales should not be obscured by matte boxes, mo-
tors or other ancillary devices.
Some other ergonomic or perhaps more practical aspects of cine lenses are
worth identifying. In prime or zoom lenses, mechanical precision and repeat-
ability of focus marks are important, especially in 35mm-format lenses of short
focal length, say 30mm or less. Typically, in a high-quality, wide-angle prime
lens, where the entire lens is moved to focus, the repeatability error due to me-
chanical backlash, slop, high spots, etc., should be less than a half-thousandth
of an inch (i.e., 12 microns metrically). Correspondingly, in wide-angle prime
or zoom lenses where internal or “floating” lens elements are utilized, a similar
tolerance, sometimes smaller, needs to be achieved. The eventual tolerance
required in this case may be larger or smaller than that mentioned, depending
upon the focal length (or power) of the movable focus group(s).
Maintenance of line of sight or “boresight” in primes, but more so for
zooms, is a major ergonomic consideration, especially in visual-effects and
| 139
Lenses
motion-control applications. The best quality zoom lenses today can be ex-
pected to provide a maximum, or, worst case, line of sight variation through
zoom of less than one inch at a 12-foot focus distance.
In addition to what has been described, the camera assistant or any other
lens user demands that a cine lens not only provide accurate focus, aperture
and zoom control, but also the right feel. An analogous way to explain this is
to consider the turning of the frequency-control knob on a radio. Any move-
ment of the knob should be smooth, not rough or sticky, and yet require some
force. The same applies to the feel of the focus, aperture, zoom or other gears
on a cine lens. For the best cine lenses, this feel is taken for granted, but for
many others an uncertainty and lack of confidence about the lens is built up
in the camera assistant’s mind, and this is not conducive to good filmmaking.
future Lenses
From conception to final use, all modern cine lenses are heavily depen-
dent on computers. The complexity of modern cine lenses, whether optical,
mechanical, electronic, some combination thereof, or otherwise, is so great
that computers are not only needed, they are fundamental to producing
high-quality cine lenses. The advancement and gradual inroad made by
zoom lenses in many fields, including cinematography, is a testament to the
importance of the computer in the whole zoom-lens design process, from
initial idea to final application. Without computers, most modern zoom
lenses would not exist.
A full series of prime lenses will still continue to be available for some time,
but increasingly, the prime lens will need to become more complex, with fea-
tures such as continuous focusing from infinity to close or even macro object
distances; internal, optically generated filtration effects; and so on, so that
their design and manufacturing costs, i.e., return on investment, remain eco-
nomically attractive. Individual prime lenses do offer one valuable advantage
over zooms: they fill niche or special lens requirements better, e.g. very wide
angle and fisheye field of view, very long focal lengths, slant focus, bellows,
etc. Therefore, in the future, cine lens complements will still include primes
as well as zooms and specialty lenses, but the mix will undoubtedly change.
Technological advancements in raw glasses, aspherical surfaces, thin films
(i.e., coatings), diffractive optics and other aspects of lenses will continue to
fuel lens development. A recently introduced compact, wide-angle, macro-
focus (continuously focusable from infinity down to almost the front lens
element), constant-aperture (through focus and zoom) cine zoom lens in-
dicates what is to come. This lens, which employs two aspherical surfaces,
some of the most exotic glasses ever produced, five cams, two movable focus
groups, two movable zoom groups, a movable iris and a total of twenty-three
lens elements, is a good example of what is already possible (See Figure 7).
140
|
Lenses
Given the high performance level achievable by such a zoom, whether it be

image quality, breathing control, macro focusing or many other features, it is
easy to understand why zoom lenses have gained and will gain ground over
traditional primes.
A good footnote is to consider where lenses have come from and where
they are likely to go in comparison to other technologies. Lenses were here
before film and will be here after film. Lenses were here before silicon-based
electronic computer processors and will be here after they are long gone!
When optically based, electronic-like computer processors arrive in the not
too distant future, lenses will still be around, and their development will still
be thriving.
Lens data tabLes

The lens data tables provided in this manual have been compiled for many
35mm, 16mm, VistaVision and 8mm spherical cine lenses purely as a guide
to what might be expected in terms of lens depth of field, hyperfocal distance
and other characteristics. It is extremely important to note that only a real
film test will determine the actual characteristics of any lens.
For lens depth of field, a conservative circle of confusion of .0010 inches
(1⁄1000) has been consistently used throughout the lens data tables. However,
with modern cine equipment, lesser circles of confusion may be necessary.
There are two reasons for this. Firstly, most modern spherical cine lenses
produce sharp images mainly due to their high-contrast capabilities, and,
secondly, most modern film negative stocks exhibit similar high-contrast
tendencies. So, for such a lens and such a film stock, with hard lighting and
no filtration, the effective or apparent depth of field falls within a smaller
range about the chosen focus distance than shown in the lens data tables.
Although this lens depth of field scenario may be considered extreme, it
seems to arise more and more often in modern filmmaking, and as most
cinematographers and their film crews will attest, it is usually better to play it
safe when dealing with lens depth of field. Therefore, it is recommended that
a reduced depth of field, up to half of what is shown in the lens data tables,
be adopted when extreme conditions arise.
In order to accommodate these needs (as indicated in the tables), the
tables enable you to calculate a circle of confusion of .0005 (5⁄10,000), or half
the size and twice as critical as listed under a given f-stop. Once you match
your focusing distance to the chosen f-stop, merely move two columns to
the left in order to see the depth of field value for a 5⁄10,000 circle of confusion.
The hyperfocal distances provided in the lens depth of field tables in this
manual indicate the focus distance to which a lens must be set so that the
(defocused) image quality performance is nearly equal for objects placed at
infinity and half the hyperfocal distance.
| 141
Lenses
To conclude, it must be emphasized that no matter how rigorous the

mathematical calculations of lens properties or characteristics (including
depth of field) are, there will always be nonlens contributions, such as filters,
film stock, lighting, etc., and these must be considered in the final outcome.
Only the results of an actual film test can determine the depth of field that is
acceptable to the eye of the beholder.
Iain A. Neil is an Optical Consultant based out of Switzerland. His company

ScotOptix contracts globally with optical technology companies, providing techni-
cal, business and intellectual property expertise with specialization in zoom lenses,
multi-configuration optical systems and new technology implementation. He has
more than 100 worldwide optically related patents issued and applied for, has
published and edited more than thirty papers and books and has garnered eleven
Academy Awards, two Emmys and the Fuji Gold Medal. He has been active in the
optics industry for over thirty-six years and is currently a fellow member of SPIE
and SMPTE, an associate member of ASC, a member of OSA and a voting member
of A.M.P.A.S.
142
|
Lenses
| 143
Camera Filters
by Ira Tiffen
C amera filters are transparent or translucent optical elements that alter

the properties of light entering the camera lens for the purpose of
improving the image being recorded. Filters can affect contrast, sharpness,
highlight flare, color and light intensity, either individually or in various
combinations. They can also create a variety of special effects. It is im-
portant to recognize that even though there are many possibly confusing
variations and applications, all filters behave in a reasonably predictable
way when their properties are understood and experienced. Most of these
properties relate similarly to filter use in both film and digital imaging. The
following will explain the basic optical characteristics of camera filters, as
well as their applications. It is a foundation upon which to build by experi-
ence. Textual data cannot fully inform; there is always something new to
discover for yourself.
In their most successful applications, filter effects blend in with the rest of
the image to help get the message across. Use caution when using a filter in
a way that draws attention to itself as an effect. Combined with all the other
elements of image making, filters help make visual statements, manipulate
emotions and thought, and make believable what otherwise would not be.
When used creatively, with an understanding of their abilities, they can re-
ally get the viewer involved.
Changing TeChnology
More than ever, changes in technology have fostered new applications,
considerations and formats for filters. Digital technology requires a new
array of spectrum and sensitivity concerns as well as opening up new im-
aging opportunities, especially post-capture. Look for references to such
changes throughout this section.
FilTer Planning
Filter effects can become a key part of the look of a production, if con-
sidered in the planning stages. They can also provide a crucial last-minute
fix to unexpected problems, if you have them readily available. Where pos-
sible, it is best to run advance tests for preconceived situations when time
allows.
144
|
Camera Filters
SizeS, ShaPeS and

on-lenS MounTing TeChniqueS
Lens-mounted filters are available in round and rectangular shapes and
in many sizes. Round filters generally come supplied with metal rings that
mount directly to the lens. Frugal filter users might find it preferable to use
adapters allowing the use of a set of filters of a single size with any lenses of
equal or smaller sizes. Round filters can also be supplied with self-rotating
mounts where needed, as for polarizers or Star effects. They can be readily
stacked in combination. Rectangular filters require the use of a special filter
holder or matte box. They offer the additional benefit of allowing slidability
for effects that must be precisely aligned within an image, such as graduated
filters. In all cases, it is advisable to use a mounting system that allows for
sturdy support and ready manipulation. In addition, the use of a lens shade
at the outermost mounting position (from the lens) will minimize the effect
of stray off-axis reflections.
FilTer FaCTorS
Many filter types absorb light that must be compensated for when calcu-
lating exposure. These are supplied with either a recommended filter factor
or a stop value. Filter factors are multiples of the unfiltered exposure. Stop
values are added to the stop to be set without the filter. Multiple filters will
add stop values. Since each full stop added is a doubling of the exposure, a
filter factor of 2 is equal to a one-stop increase. Example: Three filters of one
stop each will need three additional stops, or a filter factor of 2 x 2 x 2 = 8
times the unfiltered exposure.
When in doubt in the field about compensation needed for a filter that you
have no information on, you might use your light meter with the incident bulb
removed. If you have a flat diffuser, use it; otherwise just leave the sensor bare.
Aim it at an unchanging light source of sufficient intensity. On the ground,
face-up at a blank sky can be a good field situation. Make a reading without
the filter. Watch out for your own shadow. Make a reading with the filter cover-
ing the entire sensor. No light should enter from the sides. The difference in
the readings is the compensation needed for that filter. You could also use a
spot meter, reading the same bright patch, with similar results. There are some
exceptions to this depending on the filter color, the meter sensitivity, and the
target color, but this method is often better than taking a guess.
Published filter-factor information should be taken as a starting point.
Differing circumstances may call for deviations from the norm.
FilTer gradeS and noMenClaTure

Many filter types are available in a range of grades of differing strengths. This
allows the extent of the effect to be tailored to suit various situations. The grade
| 145
Camera Filters
numbering range can vary with the effect type; generally, the higher the num-
ber, the stronger the effect. Unless otherwise stated, there is no mathematical
relationship between the numbers and the strengths. A grade 4 is not twice
the strength of a grade 2. A grade 1 plus a grade 4 doesn’t add up to a grade 5.
Another possible source of confusion is that the various filter manufac-
turers often offer filters with similar names, using terms such as “fog” and
‘diffusion,’ which may have different characteristics.
The oldest standard for naming filter colors was developed early in the
20th century by Frederick Wratten and his associate, C.E. Kenneth Mees.
While at Kodak they sought to make early film capabilities more closely re-
spond to customer requirements. In doing so, they created specifications for
a series of filter colors that correspond to particular applications. They gave
each color a number, what we have since called Wratten numbers. These
will be referenced later in this article. Kodak makes gel filters that are the
defining standard for the Wratten system. Other manufacturers may or may
not reference the Wratten designations, even if they otherwise use their own
numbering system.
Contact the various manufacturers for additional details about their filter
products and nomenclature.
CaMera FilTerS For BoTh Color

and BlaCk-and-WhiTe
ultraviolet Filters
Film often exhibits a greater sensitivity to what is invisible to us: ultra-
violet light. This is most often outdoors, especially at high altitudes, where
the UV-absorbing atmosphere is thinner, and over long distances, such as
marine scenes. It can show up as a bluish color cast with color film, or it can
cause a low-contrast haze that diminishes details, especially when viewing
far-away objects in either color or black-and-white. Ultraviolet filters absorb
UV light generally without affecting light in the visible region.
It is important to distinguish between UV-generated haze and that of
airborne particles such as smog. The latter is made up of opaque matter that
absorbs visible light as well as UV light, and will not be appreciably removed
by a UV filter.
Since the amount of UV encountered changes often, some cinematogra-
phers find that they can obtain more consistent color outdoors through the
regular use of a filter that totally removes UV, along with their other filters.
Some manufacturers will offer combination UV absorbers with other effects
on special order for this purpose.
Ultraviolet filters come in a variety of absorption levels, usually measured
by their percent transmission at 400 nanometers (nm), the visible-UV wave-
length boundary. Use a filter that transmits zero percent at 400nm for aerial
146
|
Camera Filters
Images by art adams/ProVIdeoCoalItIon.Com

Figure 1a. NO FILTER: Digital cameras can be overly sensitive to light in the far red and
infrared. Certain objects have the property of reflecting a disproportionate amount of light
in this spectral region, which can result in dark objects like the four black fabric samples
in the center and at the left of the chart appearing with a distinct reddish tint while the two
samples on the right remain neutral.
and far-distant scenes; one that transmits in the 10%–30% range at 400nm is
fine for average situations.
Digital imaging sensors typically are less sensitive than film to UV light
and have less need of UV-attenuating filters.
infrared Filters and Film imaging

Certain special situations call for the use of black-and-white or color
infrared-sensitive films. For aerial haze penetration, recording heat effects
and other purposes, they are invaluable. Their color and tonal renditions are
very different, however, from other film types (consult film manufacturers
for further details). Various filters are used to reduce unwanted visible light.
Red, orange and yellow filters, as used for panchromatic black-and-white
film, can enhance contrast and alter color.
Filters that absorb all visible light, transmitting only infrared, such as the
Wratten #87 or #89, can also be useful. Bear in mind that you cannot see
through these filters, which makes their use more complicated. Results will
vary with film type and other factors.
Since most lenses focus infrared wavelengths slightly differently than
other colors, you may need to use a different focusing scale on your lens
when using infrared film.
In general, for black-and-white imaging, heat-reflecting surfaces will ap-
pear lighter than they might ordinarily; for color imaging, the image will
consist of artificial colors corresponding to the various levels of reflected
heat. Prior testing for most situations is a must.
| 147
Camera Filters
Figure 1b. WITH FAR RED/INFRARED FILTER: The addition of a filter that absorbs or
reflects an appropriate amount of light in the far red/infrared region can eliminate the
reddish cast and visually restore the objects to their proper neutral appearance, as seen
in this image. Digital cameras differ by design so that the correction filter required by
each will vary. It is best to check with the camera and filter manufacturers for what will
work best with any given camera.
infrared Filters and digital imaging

Digital imaging sensors have a greater sensitivity to far red and infrared
than film. This part of the spectrum plays a more important part in the rendi-
tion of proper color, particularly for skin tones, than with film. However, this
also raises some difficulties that can be mitigated by certain optical filters.
Black objects in particular can reflect varying amounts of far red and infra-
red light while often retaining a similar appearance to the eye. The difference
is due to variations in the dyes used to color the objects. The extra sensitiv-
ity of the digital sensor to this light, though, can impart a color tint, often
brownish-red, to the rendition of certain of the objects (and not to others)
as captured by the camera. This can produce unacceptable color variations.
Using a filter that reflects infrared light, known as a hot mirror, is often
the solution to this problem. Its dichroic coating equalizes the amount of
far red and infrared light reflected from each object to a point where it does
not exhibit a reddish color cast. Colors otherwise unsuitably tinted appear
natural with this filter.
Some digital cameras already incorporate varying degrees of hot mirror ca-
pability in their internal design. It will be important to know what level, if any, of
such attenuation of far red and infrared light is needed by the camera in use to
avoid both the above effect and introducing any other unwanted color artifacts.
Another technique is the combination of certain color temperature con-
trol filters with hot mirrors. Examples include the 80C and the 80D Wratten
colors. This has been found to be helpful for certain cameras.
148
|
Camera Filters
neutral-density Filters
When it is desirable to maintain a particular lens opening for sharpness
or depth-of-field purposes, or simply to obtain proper exposure when con-
fronted with too much light intensity, use a neutral-density (ND) filter. This
will absorb light evenly throughout the visible spectrum, effectively altering
exposure without requiring a change in lens opening and without introduc-
ing a color shift.
Neutral-density filters are denoted by (optical) density value. Density
is defined as the log, to base 10, of the opacitance. Opacitance (degree of
absorption) of a filter is the reciprocal of (and inversely proportional to) its
transmittance. As an example, a filter with a compensation of one stop has a
transmittance of 50%, or 0.5 times the original light intensity. The reciprocal
of the transmittance, 0.5, is 2. The log, base 10, of 2 is approximately 0.3,
which is the nominal density value. The benefit of using density values is
that they can be added when combined. Thus, two ND 0.3 filters have a
density value of 0.6. However, their combined transmittance would be found
by multiplying 0.5 x 0.5 = 0.25, or 25% of the original light intensity.
Neutral-density filters are also available in combination with other filters.
Since it is preferable to minimize the number of filters used (see section on
multiple filters), common combinations such as a Wratten #85 (daylight-
conversion filter for tungsten film) with an ND filter are available as one
filter, as in the 85N6. In this case, the two-stop ND 0.6 value is in addition to
the exposure compensation needed for the base 85 filter.
There are two types of neutral-density filters in general use. The most preva-
lent type uses organic dyes to attenuate light. For situations where it is neces-
sary to obtain the most even control from near-ultraviolet through the visible
spectrum into the near-infrared, a metallic vacuum-deposition coating, often
a nickel alloy, is ideal. For most situations, though, the silvered-mirror appear-
ance of these filters imparts internal reflections that need to be addressed in use.
Special metallic coatings can also be employed for recording extreme
bright-light situations, such as the sun during an eclipse. These filters are
very dense and absorb substantially through the IR and UV range, as well as
the visible, to reduce the potentially blinding level of light. The best have a
density value of about 5.7, which allows less than 0.001% of the overall light
through. Caution: Do not use any filter to aim at the sun unless it is clearly
labeled as having been made for that purpose. Follow the manufacturer’s
directions and take all possible precautions to avoid accidental (and poten-
tially permanent) blindness.
Variable neutral density Filters

While available for some time for use with film, the growth of digital imag-
ing with its inherently greater light sensitivity has fostered a growing need for
| 149
Camera Filters
the variable ND filter. The most common version combines two polarizers in
one assembly which, when rotated relative to each other, produce the desired
variation in overall light transmission while retaining a mostly neutral color
rendition. While you can produce your own version of this effect using two
separate polarizers, particularly a circular one mounted on the lens and a
linear one mounted onto that (away from the camera), there are reasons to
obtain a purpose-built unit. Since polarizers can affect light of different wave-
lengths somewhat differently, they can generate a color cast, often deep blue,
at the lowest transmission point, which is reached when both polarizers are
positioned so their polarization axes are perpendicular to each other. Proper
selection of polarizer foil characteristics for use by the manufacturers of vari-
able ND filters will produce more neutral color balance, but even then, using
these filters is usually better when not set for the lowest possible transmission.
For most current cameras that require circular polarization, it is impor-
tant that the variable ND has a quarter-wave retarder on the side facing the
camera, as with a traditional circular polarizer. In addition, if the variable
ND assembly allows not only one filter element to rotate relative to the
other (necessary to achieve the variable density effect) but also allows both
elements to rotate together (while retaining their positions relative to each
other) then the assembly can also function similarly to a traditional polar-
izer, where reduction of polarized reflected glare improves images of the
sky, surfaces of bodies of water, and when imaging through windows. When
using these remember to separately make rotational adjustments for both
overall transmission as well as for polarized glare reduction.
infrared-neutral density Filters

Digital camera sensors often have greater sensitivity to light than most
film. This has prompted the need for increasing use of denser versions of
neutral density filters for exposure control and for manipulating depth-of-
field. An issue with these relates to the preponderance of infrared sources
and reflecting surfaces as well as the manner in which neutral density filters
are made, in reference to how ND filters handle light in the far-red and in-
frared regions of the spectrum.
Traditional ND filters made using dyes do not absorb light in the far red
quite the same as elsewhere in the visible spectrum. When such a one-stop
(0.3) ND filter is used it will allow a slightly greater percentage of light in the
far red through than would be dictated by color neutrality. This percentage
increases as the ND density increases, but is usually acceptable up through
ND 0.9. Beyond that, the incremental increases in the relative amount of far
red light passing through imparts an unwanted brownish-red color tint effect.
This can be handled by certain dyes in the glass that absorb specifically
in the far red region. Another approach is to use strong dichroic coatings
150
|
Camera Filters
Images by brIan austad

Figure 2a. NO FILTER: Midday scene as it appears without a filter. The underexposed fore-
ground appears unnaturally dark when the exposure is set to properly render the sky.
that reflect such light, like the hot mirrors mentioned earlier. These can be
made to various transmission percentages to match a particular camera’s
requirements. A characteristic of such a coating is that it is angle-dependent,
meaning the color of light passing through it is different depending on the
incident angle. In practical terms, light going straight through the center
of the filter will be slightly differently tinted than light entering at a greater
angle at the filter’s edge. The stronger the effect of the coating, the greater
this disparity between center and edge. In addition, this effect becomes espe-
cially prominent at wider lens angles. In some situations, you may find that
this color shift is unacceptable.Testing is recommended.
Ultimately, what you need to know is that there are infrared-neutral den-
sity (IR-ND) filters made using various combinations of these techniques,
the dichroic hot mirror coatings and the far-red-absorbing dyes, that are of-
fered by various filter manufacturers that each work with a particular range
of applicable cameras. A key concern to be aware of is that they may or may
not be able to be stacked in combination depending on the specific filter
types and application. The matching filter specifications can change with
each new camera introduction. Keep abreast of developments in this area
with your suppliers of cameras and filters.
A relatively new entry offering a unique approach in this field is the Tes-
sive Time Filter. Essentially a controllable liquid crystal shutter panel that fits
in a standard matte box, its primary use is to eliminate certain undesirable
motion artifacts. Since it also acts as an electronically variable ND filter that
attenuates IR it can function as a variable IR-ND filter.
| 151
Camera Filters
Figure 2b. GRADUATED NEUTRAL DENSITY 0.6 FILTER: Taken at the same midday time as
the unfiltered image, the filter absorbs two stops from the sky, allowing it to appear cor-
rectly while the exposure is adjusted to properly render the foreground through the clear
half of the filter. The soft transition between the clear and the ND halves of the filter allow
the effect to blend well together and make for a more balanced image.
graduated nd Filters
Often it is necessary or desirable to balance light intensity in one part of
a scene with another, namely in situations where you don’t have total light
control, as in bright exteriors. Exposing for the foreground will produce a
washed-out, overexposed sky. Exposing for the sky will leave the foreground
dark, underexposed
Graduated ND filters are part clear, part neutral-density, with a smoothly
graded transition between. This allows the transition to be blended into the
scene, often imperceptibly. An ND .6-to-clear, with a two-stop differential,
will sometimes compensate the average bright-sky-to-foreground situation.
These filters are also available in combination colors, where the entire
filter is, for example, a Wratten #85, while one half also combines a graded-
transition neutral-density, as in the #85-to-85N6. This allows one filter to
fulfill the need for two
Graduated filters generally come in three transition types. The most com-
monly used is the soft-edge graduation. It has a wide enough transition area
on the filter to blend smoothly into most scenes, even with a wide-angle
lens (which tends to narrow the transition within the image). A long focal
length, however, might only image in the center of the transition. In this
case, or where the blend must take place in a narrow, straight area, use a hard
edge. This is ideal for featureless marine horizons. For situations where an
152
|
Camera Filters
extremely gradual blend is required, an attenuator is used. It changes density

almost throughout its length.
The key to getting best results with a graduated filter is to help the effect
blend in as naturally as possible. Keep it close to the lens to maximize transi-
tion softness. Avoid having objects in the image that extend across the tran-
sition in a way that would highlight the existence of the filter. Don’t move the
camera unless the transition can be maintained in proper alignment with
the image throughout the move. Make all positioning judgments through a
reflex viewfinder at the actual shooting aperture, because the apparent width
of the graduation is affected by a change in aperture.
Graduated filters are best used in a square or rectangular format in a rotat-
ing, slidable position in a matte box. This will allow proper location of the
transition within the image. They can be used in tandem; for example, one
affecting the upper half, the second affecting the lower half of the image. The
center area can also be allowed to overlap, creating a stripe of the combina-
tion of effects in the middle, most effectively with graduated color filters (see
section on Graduated Color Filters).
Circular gradient and Circular/elliptical Center Spot nd Filters

Replicating the look of early lenses, which tended to go dark at the edges, or
for other situations where such an effect is desirable, circular gradient filters
can be made; they have a clear circular center that gradually darkens in a cir-
cular fashion out toward the edges. Center spot ND filters are similar in hav-
ing a clear center, but there is no graduated transition: a clear circle or ellipse
(or other shape, to order) with a razor-hard edge is surrounded by an even
border of neutral density. In addition, subtly brightening a particular area
of the image with these filters (relative to the darker edges) helps direct the
viewer’s eye toward action in the brighter area. This is especially useful when
positioning the filter using a slidable, rotating mount such as a matte box so
that the brighter area can be located almost anywhere within the frame.
Conversely, when using certain lenses with a design that produces a
brighter center and darker edges, there are circular gradient filters that are
darker in the center and clear toward the edges to compensate. These filters
are usually offered by the same companies that manufacture these lenses.
Polarizing Filters
Polarizers allow color and contrast enhancement, as well as reflection
control, using optical principles different from any other filter types. Most
light that we record is reflected light that takes on its color and intensity from
the objects we are looking at. White light, as from the sun, reflecting off a
blue object appears blue because all other colors are absorbed by that object.
A small portion of the reflected light bounces off the object without being
| 153
Camera Filters
absorbed and colored, retaining the original (often white) color of its source.
With sufficient light intensity, such as outdoor sunlight, this reflected glare
has the effect of washing out the color saturation of the object. It happens
that for many surfaces, the reflected glare we don’t want is polarized, while
the colored reflection we want isn’t.
The waveform description of light defines nonpolarized light as vibrating in
a full 360-degree range of directions around its travel path. Polarized light in its
linear form is defined as vibrating in only one such direction. A (linear) polar-
izing filter passes light through in only one vibratory direction. It is generally
used in a rotating mount to allow for alignment as needed. In our example
above, if it is aligned perpendicular to the plane of vibration of the polarized
reflected glare, the glare will be absorbed. The rest of the light, the true-colored
reflection vibrating in all directions, will pass through no matter how the po-
larizing filter is turned. The result is that colors will be more strongly saturated,
or darker. This effect varies as you rotate the polarizer through a quarter-turn,
producing the complete variation of effect from full to none.
Polarizers are most useful for increasing general outdoor color saturation
and contrast. Polarizers can darken a blue sky, a key application, on color as
well as on black-and-white film, but there are several factors to remember
when doing this. To deepen a blue sky, the sky must be blue to start with,
not white or hazy. Polarization is also angle-dependent. A blue sky will not be
equally affected in all directions. The areas of deepest blue are determined by
the following rule of thumb: when setting up an exterior shot, make a right
angle between thumb and forefinger; point your forefinger at the sun. The area
of deepest blue will be the band outlined by your thumb as it rotates around
the pointing axis of your forefinger, directing the thumb from horizon to ho-
rizon. Generally, as you aim your camera either into or away from the sun, the
effect will gradually diminish. There is no effect directly at or away from the
sun. Do not pan with a polarizer without checking to see that the change in
camera angle doesn’t create undesirable, noticeable changes in color or satura-
tion. Also, with an extra-wide-angle view, the area of deepest blue may appear
as a distinctly darker band in the sky. Both situations are best avoided. In all
cases, the effect of the polarizer will be visible when viewing through it.
Polarizers need approximately 1½ to 2 stops exposure compensation,
generally without regard to rotational orientation or subject matter. They are
also available in combination with certain standard conversion filters, such
as the 85BPOL. In this case, add the polarizer’s compensation to that of the
second filter.
Certain camera optical systems employ internal surfaces that also polar-
ize light. One example is the use of a videotap. Using a standard (linear)
polarizer may cause the light to be further absorbed by the internal optics,
depending on the relative orientation. This may interfere with the normal
154
|
Camera Filters
operation of the equipment. The solution in these instances is to use a cir-

cular polarizer. The term “circular” does not refer to its shape. Rather, it is
a linear polarizer to which has been added, on the side facing the camera, a
clear (you can’t see it, you can only see what it does) quarter wave retarder.
This corkscrews the plane of polarization, effectively depolarizing the light
(after it has been through the linear polarizer, which will have already had
its effect on enhancing your image), eliminating the problem. It is of critical
importance, then, when using a circular polarizer that it be oriented in the
proper direction. That is, the retarder layer must be on the side facing the
camera. The filter must be either clearly labeled by the manufacturer as to
the correct direction, or mounted in a ring that only threads on one way.
Be careful when using a filter that mounts on the rear of a lens, as in some
wide-angle designs. Some lenses will allow the threaded filter ring to mount
in the wrong orientation. You can ensure the correct direction by seeing that
the direction the camera views through the filter is the one where the filter
functions as a normal polarizer. If turned the other way, it will not produce
the polarization effect. The circular polarizer otherwise functions in the
same manner as a standard linear one.
Polarizers can also control unwanted reflections from surfaces such as
glass and water. For best results, be at an angle of 33 degrees incident to the
reflecting surface. Viewing through the polarizer while rotating it will show
the effect. It may not always be advisable to remove all reflections. Leaving
some minimal reflection will preserve a sense of context to a close-up image
through the reflecting surface. For example, a close-up of a frog in water may
appear as a frog out of water without some telltale reflections.
For certain situations, it may be desirable to use a pair of polarizers to
create a variable-density filter. Although the transmission will vary as you
rotate one relative to the other, you should be aware of the resultant light
loss, even at its lightest point.
For relatively close imaging of documents, pictures and small three-di-
mensional objects in a lighting-controlled environment, as on a copy stand,
plastic polarizers mounted on lights aimed at 45 degrees to the subject from
both sides of the camera will maximize the glare-reducing efficiency of a
polarizer on the camera lens. The camera in this case is aimed straight at the
subject surface, not at an angle. The lighting polarizers should both be in the
same perpendicular orientation to the one on the lens. Again, you can judge
the effect through the polarizer.
Polarization Filters and 3-d

The development of optical systems for capturing right and left eye stereo-
scopic views for 3-D productions uncovered new applications for polariza-
tion. A staple of 3-D production today, the beam-splitter rig positions two
| 155
Camera Filters
cameras at right angles to each other, each aimed at opposing sides of a par-
tially silvered mirror. One camera sees through the mirror, the other images
off the reflection. The reflected light will be partially polarized by the mirror;
the transmitted light will only exhibit the partial polarization inherent in
the scene that both cameras are recording. The differences in the two light
paths can cause troublesome differences in brightness of particular details in
the image that can disrupt the stereoscopic effect. One solution to this is to
position a quarter-wave retarder plate, essentially a clear piece of glass that
twists the polarization axis of the incoming light, in front of the mirror in
the beam-splitter rig, effectively neutralizing the negative polarization effect.
SPeCial eFFeCT FilTerS

general information
The following filter types are available in a wide range of grades useful in
both color and black-and-white imaging. They have no recommended filter
factors, but may require exposure compensation based on several things.
Filters that lower contrast or create flare, where contrast and/or light inten-
sity is higher, will do more for any given grade. The more light they have to
work with, the more they can do. The same filter in two different lighting
conditions may produce two different effects. With diffusion, or image-
softening filters, higher contrast scenes appear sharper, needing more dif-
fusion than scenes of lower contrast. Diffusion requirements will also vary
with other conditions. Smaller film formats will allow less diffusion, as will
large-screen projection. Color may allow less diffusion than black and white.
Producing for television may require a greater degree of diffusion to survive
the transition. These relationships should cause you to choose exposure and
filter grade based on the situation and personal experience. Prior testing is
always recommended.
diffusion Filters
Many different techniques have been developed to diffuse image-forming
light. Strong diffusion can blur reality for a dream-like effect. In more subtle
forms, diffusion can soften wrinkles to remove years from a face. The optical
effects all involve bending a percentage of the image-forming light from its
original path to defocus it.
Some of the earliest portrait diffusion filters still in use today are nets. Fine
mesh, like a stocking, stretched across the lens has made many a face appear
youthful, flawless. This effect can now be obtained through standard-sized
optical glass filters, with the mesh laminated within. These function through
“selective” diffusion. They have a greater effect on small details, such as wrin-
kles and skin blemishes, than on the rest of the image. The clear spaces in the
mesh transmit light unchanged, preserving the overall sharp appearance of
156
|
Camera Filters
the image. Light striking the flat surface of the net lines, however, is reflected
or absorbed. A light-colored mesh will reflect enough to either tint shadows
lighter, which lowers contrast, or its color while leaving highlight areas alone.
The effect of diffusion, however, is produced by the diffraction of light that
just strikes the edges of the mesh lines. This is bent at a different angle, chang-
ing its distance to the film plane, putting it out of focus. It happens that this
has a proportionately greater effect on finer details than on larger image ele-
ments. The result is that fewer wrinkles or blemishes are visible on a face that
otherwise retains an overall, relatively sharp appearance.
The finer the mesh, the more the image area covered by mesh lines and
the greater the effect. Sometimes, multiple filters are used to produce even
stronger results.
As with any filter that has a discrete pattern, be sure that depth of field
doesn’t cause the net pattern to become visible in the image. Using small
apertures or short focal length lenses make this more likely, as will using a
smaller film format such as 16mm vs. 35mm, given an equal field of view.
Generally, midrange or larger apertures are suitable, but test before critical
situations. When in need of net diffusion in circumstances where mounting
it in front of the lens will cause the pattern to show, try mounting the filter
in a suitable location behind the lens (if the equipment design allows). This
should reduce the chance of the pattern appearing. Placing a glass filter be-
hind the lens may alter the back-focal length, which may need readjustment.
Check with your lens technician. A test is recommended.
When diffusing to improve an actor’s facial appearance, it is important not
to draw attention to the presence of the filter, especially with stronger grades,
when diffusion is not required elsewhere. It may be desirable to lightly dif-
fuse adjacent scenes or subjects not otherwise needing it to ensure that the
stronger filtration, where needed, is not made obvious.
In diffusing faces, it is especially important that the eyes do not get overly
soft and dull. This is the theory behind what might be called circular diffusion
filters. A series of concentric circles, sometimes also having additional radial
lines, are etched or cast into the surface of a clear filterThese patterns have the
effect of selectively bending light in a somewhat more efficient way than nets,
but in a more radial orientation. This requires that the center of the circular
pattern is aligned with one of the subject’s eyes—not always an easy or possible
task—to keep it sharp. The rest of the image will exhibit the diffusion effect.
A variation on the clear-center concept is the center-spot filter. This is a
special-application filter that has a moderate degree of diffusion surrounding
a clear central area that is generally larger than that of the circular diffusion
filter mentioned previously. Use it to help isolate the main subject, held sharp
in the clear center, while diffusing a distracting background, especially in
situations where a long lens and depth-of-field differentiation aren’t possible.
| 157
Camera Filters
Figure 3a.
Images by Ira tIffen
NO FILTER –
STANDARD EXPOSURE.
Midday scene as it appears
without a filter.
Figure 3b.
BLACK PRO-MIST 1
STANDARD EXPOSURE
Midday scene with highlight haze
more visually suggestive of the
sun’s heat and the humidity by
the lake.
Figure 3c.
SUNRISE 3 GRAD PLUS
BLACK PRO-MIST 1,
one stop under.
Combining the hazy atmosphere
of the Black Pro-Mist with the
color of the Sunrise Grad, under-
exposure produces a visual sense
of early morning.
Figure 3d.
TWILIGHT 3 GRAD,
two stops under.
The cool colors of the Twilight
Grad plus a two-stop underexpo-
sure produces a visual sense of
early evening.
158
|
Camera Filters
Another portrait diffusion type, long in use, involves the placement of

small lenslets or clear refracting shapes dispersed on an otherwise clear op-
tical surface. They can be round, diamond-shaped or otherwise configured.
These are capable of more efficient selective diffusion than the net type and
have no requirement to be aligned with the subject’s eye. They don’t lower
contrast by tinting shadows, as light-colored nets do. These lenslets refract
light throughout their surface, not just at the edges. For any given amount of
clear space through the filter, which is relative to overall sharpness, they can
hide fine details more efficiently than net filters.
The above types of filters, though most often used for portrait applica-
tions, also find uses wherever general sharpness is too great and must be
subtly altered.
Some diffusion filters, notably called “dot” filters, can effectively combine
image softening with the appearance of mild highlight flare and a reduction
in contrast. Although perhaps more akin to mist and fog effects, they also
fall into the category of diffusion filters.
Sliding diffusion Filters

When attempting to fine-tune the application of diffusion within a se-
quence, it can be invaluable to be able to vary the strength of the effect while
filming. This can be accomplished by employing an oversized filter that has
a graduated diffusion effect throughout its length. It is mounted to allow
sliding the proper grade area in front of the lens, which can be changed on
camera. When even more subtle changes are required, maintaining consis-
tent diffusion throughout the image while varying the overall strength, a
dual opposing-gradient filter arrangement can be used.
optical diffusion in the digital Camera

Unlike film, digital imaging records on discrete elements known as pixels.
These are positioned as an array over the surface of the imaging sensor. The
pattern of pixels can produce a distractingly animated moiré effect with
details in the scene being recorded. To address this, digital cameras are
typically provided with an optical low pass filter, or OLPF, placed in the light
path just before the sensor. This alters the position and pattern of fine details
striking the sensor so that the risk of generating unwanted moiré artifacts is
substantially reduced or eliminated.
The OLPF will result in a typically minimal amount of image diffusion,
but also a much improved image due to its effect on moiré.
Fog, double Fog and Mist Filters

A natural fog causes lights to glow and flare. Contrast is generally lower, and
sharpness may be affected as well. Fog filters mimic this effect of atomized water
| 159
Camera Filters
droplets in the air. The soft glow can be used to make lighting more visible,
make it better felt by the viewer. The effect of humidity in, say, a tropical scene
can be created or enhanced. In lighter grades, these filters can take the edge off
excess contrast and sharpness. Heavier grades can create unnatural effects, as
for fantasy sequences. In general, however, the effect of a strong natural fog is
not produced accurately by fog filters in their stronger grades. Their look is too
soft, with too much contrast, to faithfully reproduce the effect of a thick natural
fog. For that, double-fog or graduated-fog filters are recommended.
Double-fog filters have milder flare and softening characteristics than
standard fog filters while exhibiting a much greater effect on contrast, espe-
cially in the stronger grades. A very thick natural fog will still allow close-up
objects to appear sharp. So will a double fog filter. The key to the effect is the
much lower contrast combined with a minimal amount of highlight flare.
Graduated-fog filters, sometimes called “scenic,” are part clear or light fog
and part denser fog effect. Aligning the clear or weaker half with the fore-
ground and the stronger half with the background will render an effect more
like that of a natural fog, accumulating density with distance.
Mist filters generally produce highlight flare which, by staying closer to
the source, appears more as a halo than will the more outwardly extended
flare of a fog filter. They give an almost pearlescent glow to highlights. The
lighter grades also find uses in toning down the excessive sharpness and
contrast of modern film and lens combinations without detracting from the
image. Black Pro Mist-type filters also create moderate image softening and
modest-to-strong highlight flare, but without as much of a lightening effect
on shadows.
Contrast Control Filters

There are many situations, such as bright sunlit exteriors, where proper
contrast is difficult to maintain. Exposing for either highlights or shadows
will leave the other severely under- or overexposed. Low-contrast filters cre-
ate a small amount of localized flare near highlight areas within the image.
This reduces contrast by lightening nearby shadow areas, leaving highlights
almost unchanged. A variation, called Soft Contrast by its manufacturer,
includes a light-absorbing element in the filter that, without exposure
compensation, will reduce contrast by also darkening highlights. Use the
Soft Contrast filter when lighter shadows are not desired and highlights are
too bright. In both cases, the mild flare produced from bright highlights is
sometimes used as a lighting effect.
Another type of filter reduces contrast without any localized flare. As
named by Tiffen, the Ultra Contrast filter series uses the surrounding
ambient light, not just light in the image area, to evenly lighten shadows
throughout. Use it where contrast control is needed without any other effect
160
|
Camera Filters
on sharpness or highlight flare being apparent.

Star and Flare effects
Making points of light in the scene more interesting has long been the pur-
pose of star filters. These typically have patterns of parallel lines chemically
etched into the filter’s surface which act like cylindrical lenses, refracting the
light into bright lines that appear to extend outward from the light source
in a direction perpendicular to the lines on the filter. There will be two star
points for each direction of parallel lines. Spacing the lines closely makes for
a stronger effect; conversely wider spacing makes it more subtle.
A related flare effect originated with anamorphic lenses. When aimed into
a bright point light source, it was possible to achieve a broad cyan-blue flare
streak, often simply referred to as a “blue streak,” due to the internal reflec-
tions in the nonspherical lens. More recently, lens accessories have been
developed that achieve this effect with spherical lenses when desired. One
is a mechanical device mated to a matte box that can produce two different
strengths of this effect. Another approach embeds cylindrical lenses into the
traditional glass filter format for use with standard matte boxes. This latter
version also facilitates creating streaks in several strengths and a variety of
colors besides blue.
FilTerS For BlaCk-and-WhiTe

Tone Control Filters
Black-and-white imaging records only tonal differences between colored
objects, which appear as black, white or different shades of gray. Proper ren-
dition depends on your own desires and, for film, the differences between
film sensitivity to colors and that of the eye. The latter is due to the fact that
most panchromatic emulsions are more sensitive to blue, violet and ultravio-
let than to other colors. Therefore, blue appears as lighter on film than it does
to the eye. This can make a blue sky light enough to appear a similar shade
of light gray as the clouds that are in it, making the clouds disappear. A more
“correct” cloud presence is obtained through the use of a yellow filter, such as
a Wratten #8, which can absorb blue light, darkening the sky to more closely
match what the eye would see. The #8 also acts as a general compensator for
most subjects, giving a tonal rendition similar to that of the eye. Deeper col-
ors, closer to the red end of the spectrum, such as Wratten #15 deep yellow,
#16 orange, and #25 and #29 red filters, will produce progressively deeper
and artificially more dramatic renditions of blue sky.
Remember that since these filters act on color differences to produce tonal
differences, the required colors must be present. The part of the sky you
are recording must be blue to be affected. Areas of sky closer to the sun or
nearer the horizon are generally less blue than elsewhere. Use of a graduated
neutral-density filter can darken a sky relative to the foreground, but will not
| 161
Camera Filters
increase contrast between a blue sky and the clouds.

Using filters for contrast control can be a matter of artistic preference or of
necessity. It is possible for two disparate colors, say, a certain orange and blue,
to record as the identical tone, eliminating any visible difference between
them. Filters will lighten objects of their own color and darken those of their
complement. Complementary color pairs are: green-red, orange-blue, and
violet-yellow. An orange filter in the above case will darken the blue and
lighten the orange; a blue filter will perform the reverse.
A green filter, such as Wratten #11, can be used to lighten green foliage to
show more detail. It may also be used to provide more pleasing skin tones
outdoors, especially against blue sky.
FilTerS For Color

general information
Recording color involves knowing more about light sources than is nec-
essary for black-and-white imaging. Sunlight, daylight, exterior lighting at
different times of day, incandescent, fluorescent, as well as other artificial
sources, all have color characteristics that vary significantly. We see images
through our eyes only after they are processed by our brain, which has the
ability to make certain adjustments to the way we see color. White will still
appear white to the eye in various types of lighting, as long as we don’t have
more than one type visible at a time. Film has no such internal compensa-
tion. It is designed to see only a certain type of light as white—all others
will appear different to the extent of their difference. Filters are required to
provide the necessary fine-tuning.
Knowing that light is a form of energy, we can theoretically view it as en-
ergy emitted from a hot object, usually termed a “black body,” that gives off
light as a function of its temperature. The color of that light can be measured
in degrees Kelvin (°K). The normally encountered types of light can be cat-
egorized by certain anticipated color temperatures or can be measured with
a color temperature meter.
Color in a digital World

Optical filters for controlling color in digital imaging can yield different
results than they do with film. The relationship between the light sensitive
components and the techniques used to produce full-color images is suf-
ficiently different between the two recording formats that it will help to
learn more from their respective manufacturers about the important color
rendition characteristics of the particular film stock or digital camera in
use.
Film may be said to be chemically coordinated to expect light at a certain
color temperature—using optical color filters will alter the spectral balance
162
|
Camera Filters
of incoming light in a manner that works well with this format.

Digital imaging combines optical as well as electronic manipulation of
incoming light and its subsequent color rendition that may or may not react
well to optical color filters depending on the particular design parameters of
the equipment in use. It will be best to check with the camera manufacturer
to determine their recommendation.
The following discussion of color control references those systems, film
and digital, that react well to the use of optical color filters for attenuating the
spectrum to yield desired color effects.
Color-Conversion Filters
Color-conversion filters are used to correct for sizable differences in color
temperature between the film and the light source. These are comprised of
both the Wratten #80 (blue, as used for daylight film in tungsten lighting)
and the Wratten #85 (amber, as used for tungsten film in daylight) series of
filters. Since they see frequent outdoor use in bright sunlight, the #85 series,
especially the #85 and #85B, are also available in combination with various
neutral-density filters for exposure control.
light-Balancing Filters
Light-balancing filters are used to make minor corrections in color tem-
perature. These are comprised of both the Wratten #81 (yellowish) and the
Wratten #82 (bluish) series of filters. They are often used in combination
with color conversion filters. Certain #81 series filters may also be available
in combination with various neutral-density filters for exposure control.
Color-Compensating Filters
Color-compensating (CC) filters are used to make adjustments to the red,
blue or green characteristics of light. These find applications in correcting
for color balance, light source variations, different reversal film batches and
other color effects. They are available in density variations of cyan, magenta
and yellow, as well as red, blue and green filters.
decamired® Filters
Decamired (a trademark of the manufacturer) filters are designed to
more easily handle a wide range of color-temperature variations than the
previously mentioned filters. Available in increments of both a red and a
blue series, Decamired filters can be readily combined to create almost any
required correction. In measuring the color temperature of the light source
and comparing it to that for which the film was designed, we can predict the
required filtration fairly well.
A filter that produces a color-temperature change of 100° K at 3400° K will
| 163
Camera Filters
produce a change of 1000° K at 10,000° K. This is because the filter relates to

a visual scale of color. It will always produce the same visible difference. A
color change of 100° K at the higher temperature would hardly be noticed.
To allow simple calculation of such differences, we convert the color
temperature into its reciprocal, that is, to divide it into 1. Then, since this
is usually a number with six or more decimal places, we multiple it by 106,
or one million, for convenience. This is then termed the “mired value,” for
micro reciprocal degrees. It identifies the specific change introduced by the
filter in a way that is unrelated to the actual temperature range involved.
To see this more clearly, let’s look at the following changes in color tem-
perature from both the degree and mired differences. Numbers are degrees
Kelvin, those in parentheses are mireds:
9100 (110) to 5900 (170) = difference of 3200 (60)
4350 (230) to 3450 (290) = difference of 900 (60)
4000 (250) to 3200 (310) = difference of 800 (60)
From this, you can see that although the degree differential varies as the
range changes, the actual filtration difference for these examples, in mireds,
is the same.
To use this concept, subtract the mired value of the light source from that of
the film (the mired value of the Kelvin temperature the film is balanced for).
If the answer is positive, you need a reddish filter; if negative, use a bluish fil-
ter. Mired-coordinated filters are termed “Decamireds.” Mired value divided
by 10 yields Decamireds. The 60 mired shifts (above) would be produced by
an R6 filter, where the higher values were that of the lighting. Sets of such
filters come in values of 1.5, 3, 6 and 12 Decamireds, in both B (bluish) and R
(reddish) colors. These numbers are additive; that is, a pair of R3s produces
an R6. An R6 plus a B6 cancel each other out to produce a neutral gray.
Fluorescent and other discontinuous Spectra lighting Correction

Since filters never actually add color but only absorb certain wavelengths
to increase the relative proportion of others, the original light source must
have the colors you want to start withSome sources are totally deficient in
certain wavelengths, which cannot be added back using only filters. This is
particularly true of many types of metal-halide lighting. With other light-
ing types, such as fluorescent, color-temperature measurements may not
provide the correct filter requirements, because color-temperature theory is
based on having a continuous spectrum, meaning light at all wavelengths. It
is possible for a light source to have a sufficient spectral distribution to emu-
late a correctable color temperature when so measured, but its effect on film
can be very different. (See section on Lighting for further details page 375.)
Fluorescent lighting generally produces a greenish color overcast. Each
of the many lamp types varies in color, and it can be difficult to know the
164
|
Camera Filters
Images by Ira tIffen

Figure 4a. NO FILTER: Standard exposure/midday scene as it appears without a filter.
precise correction. Similarly named lamp types can vary with manufacturer,
and lamps will also vary over time in use. Special filters are available, how-
ever, that are designed to yield good-to-excellent color under fluorescents,
without the need for a color- temperature meter and a variety of CC filters.
Mixed lighting Situations

A question often arises about what to do when there is more than one type of
lighting used in a scene. The key to this is to first try to make all the light sources
behave the same. That is, choose one that predominates, correct the camera
for that, and correct the other lighting with gel filters made for them. You can
convert daylight coming through a window with a gel placed over the window,
which will make it a similar color temperature as the predominant tungsten, or
even fluorescent, lighting inside. Then correct the camera for that type of light.
There are many such combinations that will work; which one to choose is often
a matter of economics. Filtering a factory of fluorescents with gels or filter tubes
may cost far more than just gelling up the occasional window.
If there is no way to correct all the lights for one color temperature, try to
minimize the intrusion of those for which you cannot correct to the camera.
Sometimes, this can be used to advantage. The cool blue light from outdoors
through a window can make the tungsten-lit interior seem that much warmer
and cozier. Once “normal” color is within reach, frequently a variation can
be even better. It all depends on your purpose and the story you are telling.
graduated Color Filters

Similar to graduated ND filters, graduated color filters are also produced
| 165
Camera Filters
Figure 4b. TAKEN AT THE SAME MIDDAY TIME AS THE UNFILTERED IMAGE: The filter
absorbs two stops and is shown here with an additional one stop underexposure – a total
of three stops under. The cool tones and underexposure produce a visual sense of evening
under moonlight illumination.
in a wide range of standard and custom colors, densities and proportions

for many applications. A blue-to-clear filter can add blue to a white, hazy
sky without affecting the foreground. An orange-to-clear filter can enliven a
tepid sunset. Color can be added to the bottom of the scene, as with a green-
to-clear filter used to enrich the appearance of a lawn.
Stripe filters are another type of graduated filter, having a thin stripe of
color or neutral density running through the center of the filter, graduating
to clear on either side. These are often used to horizontally paint various
colors in layers into a sky, as well as for narrow-area light balancing.
When seeking to gradually alter the color or exposure of a scene on-
camera, an extra-long graduated (attenuated) filter, mounted to slide in front
of the lens, can accomplish what might otherwise be a complicated task. For
example, going from the appearance of daylight to that of twilight can be
synthesized with a day-for-night filter in combination with a neutral density,
both gradually altering density in the one filter. With careful lighting, this can
achieve a realistic change-of-time look with a minimum of difficulty, as both
exposure and color are altered together in a controllable manner
Coral Filters
As the sun moves through the sky, the color temperature of its light changes.
It is often necessary to compensate for this in a variety of small steps as the day
progresses, to match the appearance of different adjacent sequences to look as
166
|
Camera Filters
though they all took place at the same time. Coral filters are a range of graded
filters of a color similar to a #85 conversion filter. From light to heavy, any ef-
fect from basic correction to warmer or cooler than normal is possible. Corals
can also compensate for the overly cool blue effect of outdoor shade.
Sepia Filters
People often associate sepia-toned images with “early times.” This makes
sepia filters useful tools for producing believable flashbacks and for period
effects with color film. Other colors are still visible, which is different from
original sepia-toned photography, but appear infused with an overall sepia tint.
didymium Filters
The Didymium Filter is a combination of rare earth elements in glass. It
completely removes a portion of the spectrum in the orange region. The ef-
fect is to increase the color-saturation intensity of certain brown, orange and
reddish objects by eliminating the muddy tones and maximizing the crim-
son and scarlet components. Its most frequent use is for obtaining strongly
saturated fall foliage. It also enlivens brick and barn reds, colors that aren’t
bright scarlet to begin with, most effectively. The effect is minimal on objects
of other colors. Skin tones might be overly warm. Even after subsequent
color timing or correction to balance out any unwanted bias in these other
areas, the effect on reddish objects will still be apparent. Prior testing should
be done because film color sensitivities vary.
underwater Color-Correction Filters

When imaging underwater, the light you are recording is filtered by the
water it passes through. Longer wavelength reds and oranges are absorbed
until only blue is left. The actual effect is determined by numerous factors,
such as light source (sun or artificial), water quality and the water path. The
latter is the distance the light travels through the water. In natural (sun)light,
this is the depth of the subject from the surface plus the subject-to-camera
distance. For artificial lighting, it is the light-to-subject-to-camera distance.
The longer the water path, the greater the filtering effect of the water. In many
cases, certain color-correcting filters can absorb enough shorter wavelengths
to restore better color balance. For example, Aquacolor®, a patented system
that precisely compensates for the light-filtering effects of water, comes in
filter densities that match different water paths. The difference between cor-
rected and uncorrected color can be dramatic. The use of faster speed films
and lenses will facilitate the use of light-absorbing correcting filters.
differences Between Camera and lab/Post Correction

It is the job of the lab timer to fine-tune the finished color rendition of
| 167
Camera Filters
a film production. This accounts for variables in exposure, print stock and
processing. Timing can also be used to impart certain color effects for densi-
ties available in the film emulsion to work with, and is limited to the range
of variation of the optical printer. These are much more limiting than the
multitude of colorants in the real world and the number of ways in which
adjustments can be made at the camera. Filtering on the camera brings the
lab that much closer to the desired result, providing a greater latitude of
timing options. It can also center more creative control in the hands of the
cinematographer.
There will be times when counting on the lab is either the only choice, or
can produce some unusual effects. When faced with a low-light situation in
daylight using tungsten film, it may be necessary for exposure reasons to pull
the #85 filter and correct in the printing. When you do this, however, neutral
gray tones will appear slightly yellow, even when all else looks correct. This
effect can be used to artificially enhance lush green foliage colors through
the addition of yellow. It may have other uses, but you will not achieve the
same result as if you had used the 85 filter.
ll-d®
The LL-D was designed to help in the above situation. It requires no ex-
posure compensation and makes sufficient adjustments to the film to enable
the timer to match the color of a properly 85-filtered original. It is not an
all-around replacement for the #85. Use it only where needed for exposure
purposes, and for subsequently printer-timed work.
digital Post Processing options

There are many new options available that allow images to be manipulated
after capture. Software now facilitates extensive control of color, contrast,
sharpness, and other factors. There are numerous versions and platforms so
it will be of benefit to find what applies best to the particular task at hand.
One variation among these produces digital filter effects that are close
emulations of optical filter effects. These have special use in that they can
more readily blend the looks of optically and digitally filtered scenes when
combined in editing.
SPeCial aPPliCaTion FilTerS

Contrast-Viewing Filters
Balancing lighting by eye is a matter of experience. Decisions can be aided
through the use of contrast-viewing filters. These are designed to handicap
the eye, with its much greater range of apparent densities, to resemble the
range of the various types of film. Use contrast viewers to judge relative
highlight and shadow densities. There are viewers for black-and-white film,
168
|
Camera Filters
Figure 5a. WITHOUT THE SPLIT-FIELD LENS (1): The foreground here is out of focus
when the background is sharp. You can’t focus on both at the same time.
as well as various viewer densities for color film. A darker viewer is used for
slower film speeds, where you would tend to use brighter lighting. Faster
film, which can be used in dimmer settings, would require a lighter viewer.
Further details can be obtained from the manufacturers.
There are also Green, Red and Blue viewing filters used to judge lighting
effects when doing process work like greenscreen.
Figure 5b. WITHOUT THE SPLIT-FIELD LENS (2): The foreground here is sharp when the
background is out of focus. You can’t focus on both at the same time.
| 169
Camera Filters
Figure 5c. WITH THE SPLIT-FIELD LENS: The close-up lens half allows sharp focus on
the foreground while the camera lens is focused on the background. There is a soft
transition between the two areas at the edge of the split-field lens at this middle-of-the-
range lens opening.
Close-up and Split-Field diopter lenses

Close-up lenses allow for closer focusing than would otherwise be avail-
able with the unaided camera lens. Cutting such a lens in half produces the
split-field lens. This can be used to have two fields of focus, one very near, the
other very far, in one scene.
A more recent variation is to have a narrow linear sliver of lens that can
be positioned in the middle region of the frame with clear spaces on either
side. (See Extreme Close-Up Cinematography page 667.)
oTher FilTer ConSideraTionS

effect of depth-of-Field and Focal-length Changes
Standard color filters generally function without change through varia-
tions in depth-of-field and focal length. This may not be true of many of the
special-effect filter types. There are no solid rules for predicting the variation
in filter effect due to depth-of-field or focal-length changes. There are some
things we can expect, however, especially when changing focal length. Let’s
look at a fog/mist type filter that causes a light to glow or flare. Take the ex-
ample of a certain grade filter where we can see that the ratio of light diameter
to glow diameter is, say, 1:3. As we view this through a changing focal length,
we will see that the ratio remains the same, although the magnification will
vary accordingly. So the decision to use a filter of a different grade to main-
tain a certain appearance at different focal lengths will be based on wanting
170
|
Camera Filters
to change the ratio, as opposed to any otherwise corresponding relationship.

Tests are advisable for critical applications.
Multiple Filter use

When any single filter is not enough to produce the desired results, use
combinations. Choose carefully to minimize the number required. Try to get
the job done with no more than three filters. Use filters that individually add
to the final effect without canceling each other out. For example, don’t use a
polarizer, which can increase color saturation, in combination with a low-
contrast filter that reduces saturation, unless it works for some other reason
(the polarizer could also be reducing window reflections, for instance). It
is important that diffusion filters be mounted closest to the lens with any
additional filters going after. Some popular filter combinations may also be
available in one filter to simplify their use.
Secondary reflections
Lighting can cause flare problems, especially when using more than one fil-
ter. Lights in the image pose the greatest difficulties. They can reflect between
filter surfaces and cause unwanted secondary reflections. Maintaining paral-
lelism between filters, and further aligning the lights in the image with their
secondary reflections where possible, can minimize this problem. In critical
situations, it may be best to make use of a matte box with a tilting filter stage.
Tilting filters of good optical quality only a few degrees in such a unit can
divert the secondary reflections out of the lens axis, out of the image, without
introducing unwanted distortion or noticeable changes in the filter’s effect.
Custom (homemade and Field-ready) Filters

There will be times when you need an effect and don’t have time to obtain
one ready-made. Certain effects can be produced that, although different
from factory filters, can be useful in a pinch or for unusual custom situations.
Net-diffusion effects can be produced as they were originally by stretching
and affixing one or more layers of stocking material to the lens end, held
in place with a rubber band. (This is also an easier way of mounting net
effects behind the lens.) There are also numerous things you can do if you
have a clear filter (or several). Petroleum jelly can cause flare or diffusion,
or even some star-like streaks, depending on its application on a clear filter,
spread with a finger or cloth. The chief benefit here is that the effect can also
be applied only to selected portions of the scene. Breathing on a clear filter
can produce interesting but temporary fog-like results. Using cut gels can
simulate certain graduated-filter effects. When doing this, be sure to keep
the filter close to the lens, and use larger lens openings to keep the visible
edge as soft as possible.
| 171
Camera Filters
rain Spinners
When rain or boat spray send distracting water drops onto the lens, you
can mount a rain spinner, sometimes called a rain deflector, as you would a
matte box on the front of the lens. A round filter, usually clear, spins at over
3000 rpm and flings off droplets before they form. This is very effective when
filming the Tour De France in the rain.
Filtering for digital intermediate

When adding computer generated effects to filmed original, it may be best
to minimize or eliminate the use of diffusion filters for image capture. Until
digital software effects provide suitable correlated results as optical effects, it
will be important to produce a suitable base for the digital post work.
It will still be best to record all of the important detail in the form that is
closest to the desired finished result. Polarizing filters are perhaps the most
useful in that they produce effects, such as reducing reflections from water
and windows, that are unlikely to be done any other way. Contrast control
filters can render shadow details visible that also may not be digitally added
in later without a lot of effort if at all.
(For more information on filters see charts section pages 819-820 and 822-
834.)
Ira Tiffen, ASC Associate Member, has created many of the important filter ef-
fects of the last forty years. He earned a 1992 Academy Technical Achievement
Award for the development of the Ultra Contrast filter series and received a Prime
Time Emmy Award in 1998. He was elected a Fellow of the SMPTE in 2002.
172
|
Camera Filters
| 173
Camera Stabilizing Systems
T here is good reason why the medium discussed in this manual is popu-
larly known as “motion” pictures. Since the earliest days of the industry,
cinematographers have been challenged to serve their stories by finding new
and inventive ways of moving the camera. This section examines the latest
devices that allow the operator to maintain control of the image while taking
the viewer on a ride up, down or anywhere else.
Body-Worn SyStemS
Modern camera-stabilizing systems enable a camera operator to move
about freely and make dolly-smooth handheld shots without the restrictions
or the resultant image unsteadiness encountered with prior methods. These
systems transfer the weight of the camera unit to the operator’s body via a
support structure and weight distribution suit. This arrangement frees the
camera from body motion influences. It allows the camera to be moved by
the operator through an area, generally defined by the range through which
his arm can move.
Camera smoothness is controlled by the “hand-eye-brain” human servo
system that we use to carry a glass of water around a room or up and down
stairs. Viewing is accomplished through the use of a video monitor system
that displays an actual through-the-lens image, the same image one would
see when looking through a reflex viewfinder. The advantage of these
camera-stabilizing systems is that the camera now moves as if it were an
extension of the operator’s own body, controlled by his or her internal servo
system, which constantly adjusts and corrects for body motions, whether the
operator is walking or running. The camera moves and glides freely in all
directions—panning, tilting, booming—and all movements are integrated
into a single, fluid motion that makes the camera seem as if it were suspend-
ed in midair and being directed to move at will. These camera-stabilization
systems turn any vehicle into an instant camera platform.
As with remotely controlled camera systems, servo controls may be used
for control of focus, iris and zoom on the camera lens.
AdvAnCed CAmerA SyStemS, InC.

BodyCam
The BodyCam camera support system moves the camera away from the
operator’s eye by providing a monitor for all viewfinding. Once separated
that way, the camera is able to go into places and point in directions that
would not be possible if the operator’s head were positioned at the eye-
174
|
piece. The lens can be placed anywhere the arm can reach. This also allows
the operator’s unblocked peripheral vision to see upcoming dangers and
obstacles.
The BodyCam prevents unwanted motion from being transferred to the
camera. With the average video camera, lens focal lengths up to 125mm are
possible, and on 35mm film cameras, focal lengths up to 250mm are usable.
The Balance Assembly is suspended in front of the operator. It supports
the camera, a viewfinder monitor and one or two batteries. The monitor is
the green CRT-type. Batteries are brick-style and last two to three hours.
Adjustments allow precise balancing of the Balance Assembly around a
three-axis gimbal. At this gimbal is an attach point where the Balance As-
sembly connects to the Suspension Assembly.
The Suspension Assembly consists of an arm supported and pivoted be-
hind the left shoulder. The rear part of the arm is attached to a double-spring
arrangement that counters any weight carried by the other end of the arm and
provides vertical movement over a short distance. This arm isolates higher
frequency motion. The front end of the arm attaches to the Balance Assembly
with a hook. This hook is connected to a cable that travels from the front of
the arm to the pivot point of the arm and an internal resilient coil.
The Suspension Assembly is attached to the Backbrace. The Backbrace is a
framework that carries the load and transfers it to the human body.
Model L weighs 24 lbs (10.9kg) without camera or batteries and will carry
up to 25 lbs (11.3kg) of camera package (not counting batteries). Model XL
weighs 28 lbs (12.7kg) without camera or battery and will carry up to 40 lbs
(18.2kg) without battery. Both models offer options for wired or wireless
focus, zoom and iris control.
GlIdeCAm InduStrIeS, InC.

Glidecam v-20
The Glidecam V-20 professional camera stabilization system was con-
structed primarily for use with 16mm motion picture cameras and profes-
sional video cameras weighing 15–26 lbs.
The lightweight, adjustable Support Vest can be adjusted to fit a wide
range of operators. High-endurance, closed-cell EVA foam padding and
integral T-6 aluminum alloy create a vest which can hold and evenly distrib-
ute the weight of the system across the operator’s shoulders, back and hips.
For safety, quick-release, high-impact buckles allow the vest to be removed
quickly.
The adjustable, exoskeletal Dyna-Elastic Support Arm is designed to coun-
teract the combined weight of the camera and Camera Mounting Assembly
(Sled) by employing high carbon alloy springs. The arm can be boomed up
and down, as well as pivoted in and out and side to side. The spring force is
| 175
field adjustable to allow for varying camera weights. For safety, dual-spring
design is used to reduce possible spring failure damage.
The free-floating Three-Axis Gimbal, which incorporates integrally shield-
ed bearings, creates the smooth and pivotal connections between the front
end of the arm and the mounting assembly. A locking mechanism allows the
gimbal to be placed at varying positions on the Central Support Post.
The Camera Mounting Assembly (Sled) is designed with a lower Tele-
scoping Center Post which allows for vertical balance adjustment as well as
varying lens heights. The center post can be adjusted from 22" to 32". The
Camera Plate has both ¼" and 3⁄8" mounting slots to accommodate a variety
of camera bases. For remote viewing, a LCD Monitor can be attached on
the Base Platform, or either a LCD or a CRT Monitor can be attached to
the base’s adjustable monitor bracket. The base platform can also be config-
ured to use counterbalance weight disks if a monitor and/or battery is not
mounted on the base platform. The back of the base platform has threaded
mounting holes for an Anton Bauer Gold Mount battery adapter plate.
Accessories include a low-mode camera mount, sled offset adapter, ve-
hicle mount, and Vista Post 33" central support post extender.
Glidecam v-16
The V-16 is the same as the V-20 but designed to support lighter cameras
weighing 10–20 lbs.
Glidecam Gold Series

The top-of-the-line Gold Series is made up of the Gold Vest and the Gold
Arm. The Gold Vest offers no-tools adjustment with a breakaway safety sys-
tem. It has a quick pressure-release dual-buckle design with positive locking
buckles and fast-ratcheting adjuster buckles. The vest features integral black
anodized T-6 aluminum with EVA foam padding, reversible and vertically
adjustable arm mounting plate, and an industry standard arm connector.
An optional V-series arm connector is available. Arm connectors are made
of titanium.
The Gold Arm incorporates six titanium springs in order to handle a cam-
era load of 13–38 lbs. Its combined camera and sled carrying capacity ranges
from 31–56 lbs. Unique to the Gold Arm are its Hyper-Extension Hinges,
which allow the arm more freedom of movement. The vest connectors are
made of titanium.
GeorGe PAddoCk InC.

Pro-tm
The PRO-TM System’s Donkey Box II camera mounting platform has
linear slide bearings. The metal-on-metal design allows for smoother, easier
176
|
movement fore/aft and side/side. Finer-thread lead screws and captured

lead nuts enable the operator to make smaller, more accurate adjustments.
A quick-release mechanism allows the camera to be mounted/dismounted
with ease.
The PRO Gimbal is compatible with all 1.5" center posts and existing
oversized grips. A locking mechanism achieves concentric clamping about
the post, ensuring that all axes converge at the post’s center.
The PRO Battery Module II batteries provide an independent and clean
power supply to the monitor and video-related accessories while eliminating
the need for a converter when running 24V cameras. The Sled can be con-
figured to carry one to three batteries in multiple combinations to support a
wider range of camera systems.
The PRO Arm, a departure from traditional tensioning methods has re-
sulted in minimal friction and a force curve designed to complement an op-
erator’s instincts. It can be easily configured to accommodate a wide variety
of load requirements ranging from 13–75 lbs.
The Post system was designed to eliminate all external cables except the
monitor cable. The Upper Junction Box and the 17½"–26½" extendable post,
which houses one internal cable, connects to the lower Junction Box via a
quick disconnect/connect bayonet mount.
The Superpost allows the operator to achieve super hi/low-mode shots.
It extends from 51"–60" (5') and is of the same design configuration as the
PRO-TM Post.
The PRO Vest offers increased comfort due to improved load distribution.
It is designed to conform anatomically to the individual operator and allows
the operator a greater choice of hip and/or shoulder load distribution. A
revolutionary latching system permits vest tension to be relaxed between
takes without a change of settings.
The 5" Diagonal High Intensity Monitor II is a self-contained design that
incorporates all electronic components within a water-resistant housing.
The high-voltage, high-intensity screen features familiar onscreen graph-
ics and indicators such as framelines, crosshairs and low battery and level
indicators. Controls for brightness, contrast, image orientation, standard/
anamorphic and graphics are all located on the faceplate for easy adjustment.
Nickel Metal Hydride Batteries deliver 14.4V @ 3.5A hours. A Fuel Gauge
LED display on each battery provides immediate readout of battery condi-
tion. Weight is 1.9 lbs per battery.
The Pro Gyro Module is designed specifically for the PRO Battery Sled II.
The new Gyro system increases the configuration possibilities of one to three
gyros, while decreasing the time and complexity involved in transitioning to
and from the use of gyros. Keeping one or two batteries on the sled leaves
sled power in place, eliminating the need for an obtrusive umbilical cable.
| 177
The umbilical cable is replaced by a small, lightweight AC-only cable. A bat-

tery belt carries one battery and the DC/AC inverter.
mk-v
mk-v evolution
The MK-V Evolution Modular Sled system can evolve with the operator.
With the addition of optional extras, the basic MK-V Evolution rig can
become a 24V 35mm sled. The base features a modular framework, allow-
ing for customization in the field to any configuration and battery system
and the addition of various accessories. It has a 4-by battery bar and clamp
system for stability and universal battery compatibility. The modular base is
also compatible with other leading stabilization systems. All battery systems,
12V and 24V, plug into the D-box distribution box. The D-box Deluxe has
built-in digital level sensing and an onscreen battery display. MK-V’s latest,
highly configurable Nexus base is compatible with the Evolution.
The available carbon-fiber posts are two- and four-stage telescopic and
available in three different lengths. The base has optional gyro mounts. The
modular telescopic monitor works with CRT and LCD monitors. The V2
advanced, frictionless gimbal is compatible with leading sled systems, tools-
free and designed to have no play in any axis. Front-mounting standard and
advanced Xo vests are available. The Xo arm is frictionless and can support
cameras of any weight with the change of its springs and features a modular
front end.
SAChtler
Artemis Cine/hd
The Artemis Cine/HD Camera Stabilizing System’s post has a telescopic
length ranging from 16.5"–27.5" (42–70cm). A variety of modules can be
fixed to either end of the post to enable cameras, monitors and battery packs
to be mounted. Scales on both tubes allow fast and precise adjustment. The
gimbal’s self-centering mechanism ensures aligned, precise position of the
central bearing for easy system balancing.
The side-to-side module situated between post and camera gives the cam-
era a sliding range of 1.18" (30mm) left, right, forward and back and includes
a built-in self-illuminating level at the rear of the module. Two positions at
the front and two at the rear enable mounting of remote focus brackets. All
adjustments or locking mechanisms can be performed tool-free or with only
a 4mm Allen wrench. The Artemis vest, made from materials such as cor-
dura, has a full-length pivoting bridge made of reinforced aluminum alloy.
It can be vertically adjusted independently of body size. Most of the weight
is carried on the hips, reducing pressure on the operator’s back. The arm is
available with different spring sets that offer a load capacity ranging from
178
|
30–70 lbs (15–35kg). The two standard arm capacities are 44 lbs (20kg) and
57 lbs (25kg). It is fully compatible with all other stabilizer systems using the
standard arm-vest connector and a 5⁄8" arm post.
tIffen SteAdICAm
universal model III
The Steadicam system consists of a stabilizing support arm which attaches
at one end to the camera operator’s vest and at the other end to a floating
camera mounting assembly which can accept either a 16mm, 35mm or
video camera. The comfortable, adjustable, close-fitting camera operator’s
vest transfers and distributes the weight of the Steadicam system (including
camera and lens) across the operator’s shoulders, back and hips.
The stabilizer support arm is an articulated support system that parallels
the operator’s arm in any position, and almost completely counteracts the
weight of the camera systems with a carefully calibrated spring force. The
double-jointed arm maximizes maneuverability with an articulated elbow
hinge. One end of the arm attaches to either side of the vest’s front plate,
which can be quickly reversed to allow the stabilizer arm to be mounted on
the right or left side of the plate. A free-floating gimbal connects the stabi-
lizer support arm to the camera-mounting assembly.
The camera-mounting assembly consists of a central support post, around
which the individual components are free to rotate as needed. One end of
the post supports the camera mounting platform, while the other end termi-
nates in the electronics module. The adjustable video monitor is attached to a
pivoting bracket. An electronic level indicator is visible on the CRT viewing
screen. Electronically generated framelines can be adjusted to accommodate
any aspect ratio. Positions of the components may be reversed to permit “low
mode” configuration. The Steadicam unit is internally wired to accept wire-
less or cable-controlled remote servo systems for lens control. A quick-release
mechanism permits the operator to divest himself of the entire Steadicam
unit in emergency. A 12V/3.5A NiCad battery pack mounts on the electron-
ics module to supply the viewfinder system and film or video camera.
master Series
The cornerstone of the Steadicam family, this high-end production tool
incorporates a new motorized stage, an enhanced Master Series vest, an ad-
vanced 5" monitor with artificial horizon and every state-of-the-art feature
in the industry today.
System Includes: Camera Mounting Chassis (Sled), Motorized Stage,
Low-Friction Gimbal, Composite Center Post and Covers, Advanced Moni-
tor: 5" CRT w/16:9 aspect ratio switchable to 4:3, Artificial Horizon, Frame-
line Generator, On-screen Battery Display, Dovetail Accessory Bracket,
| 179
20–45 lbs camera capacity Iso-Elastic Arm, No-tools Arm Post, Master
Series Leather Vest, four CP batteries with built-in Meter Systems, 12V/24V
DC to DC Power Converter, Dynamic Spin & Docking Bracket, Long and
Short Dovetail Plates, 12V power cable, 24V power cable (open-ended),
two power cables for Panavision and Arri (24V) cameras, two power/video
cables for Sony XC-75, soft case for the Arm, soft case for the Vest, hard case
for the Sled, Owner’s Manual.
master Series elite

With many of the same features as the Master Series, the Elite is a more
basic and affordable Steadicam with a variety of upgradeable options.
System Includes: Camera Mounting Chassis (Sled), Low-Friction Gimbal,
Composite Center Post and Covers, Monitor: 4" CRT w/4:3 aspect ratio
switchable to 16:9, Frameline Generator, 20–45 lbs camera capacity Iso-Elas-
tic Arm, No-tools Arm Post, Lightweight Master Series Vest, Dynamic Spin
& Docking Bracket, two CP batteries with built-in Meter System, Long and
Short Dovetail Plates, 12V power cable, 3' Lightweight video cable, hard case
for the Sled, soft case for the Arm, soft case for the Vest, Owner’s Manual.
vector 12, vector el, vector Cn

Tiffen’s latest system for film and video is surpassed only by the Ultra se-
ries. Each model offers no-tool arm technology that accommodates camera
weights up to 45 lbs (20.5kg). The Vector Sled features carbon fiber post
and covers, a three-post system that extends from 27"–48". The slim 2.75"
wide sled has modular construction, 24V single battery system (Vector 12 is
13.2V only) and low-friction precision gimbal. Standard or compact light-
weight vest.
The CN model includes the Tiffen HD UltraBrite Version2 color 8.4"
LCD monitor that can display virtually all component high-definition and
composite NTSC/PAL/SECAM signals (optional on 12 and EL models). The
iso-elastic Vector arm has 24" of boom travel and utilizes titanium springs
that can be adjusted while supporting the camera.
ultra elite, ultra Cine

These models can configure quickly to the best advantage for each indi-
vidual shot. The key component is a +20° head that maximizes use of ex-
tended posts, preserves dynamic balance in regular and long modes, helps
with clearance reach and viewing, and enables whip pans with lens angled
up or down. It features carbon-fiber telescoping posts, monitor, battery rods
and brackets and a proprietary stiffening system.
The Ultra Cine system includes: Wireless motorized stage V.2.0, Dynamic
Balance Calculator, Super Performance Monitor 4:3/16:9 switchable, Arti-
180
|
ficial Horizon, Frameline Generator, 20–40 lb camera capacity Iso-Elastic

Arm, 24V battery, Onscreen Battery Display, No-tools Arm Post, Low-
Friction Gimbal, Composite Post and Covers, Leather Vest.
hAndheld StABIlIzAtIon SyStemS

doGGICAm SyStemS
the doggicam
The Doggicam is a self-contained, lightweight and fully balanceable hand-
held camera system. In low mode, shooting forward allows dynamic tracking
free of tracks, and shooting backwards, the operator can view the monitor
to see where he is going while leading the subject. High mode allows the
operator to move the camera in an extremely dynamic fashion. Configuring
from low mode to high mode takes 3–4 minutes.
At the heart of the Doggicam system is a lightweight, 6.5 lbs, highly modi-
fied Arri 2-C with crystal speeds from 4–60 fps, PL or Panavision mounts
and reflex viewing. The camera may be set up for Super 35 or anamorphic
formats. It is easily adapted for 16mm cameras. Handheld bracketry is avail-
able. The system is readily integrated for studio mode use by adding a riser
block, rotating eyepiece and video door.
robodog
This motorized version of the Doggicam allows the operator full-tilt con-
trol of the camera by the handle-mounted motor controller. The inherent
stability of the system is maintained by keeping the post vertical while allow-
ing the full range of camera tilt within the yoke.
doggimount
The Doggimount quickly secures the camera at any point in space using a
compact aluminum framework similar to speedrail, yet much smaller.
Bodymount
Bodymount uses a specially designed vest and the standard Doggimount
bracketry to allow easy and comfortable attachment of the camera to a per-
son. The vest can be covered with wardrobe to allow framing to the middle
of the chest.
Pogocam
The Pogocam is a small, weight-balanced handheld system that utilizes a
converted 35mm Eyemo (100' daylight loads) with a videotap and onboard
monitor. Nikon lenses are used. The compact size allows you to shoot where
most stabilization systems would be too cumbersome to go.
| 181
mItChell-BASed PlAtform-mounted
StABIlIzAtIon SyStem
CoPtervISIon
rollvision
Rollvision is a lightweight, portable, wireless 3-axis camera system with
360° unlimited pan, tilt and roll capabilities and a gyro for horizon compen-
sation. The dimensions are 14.5" x 19" x 19.5" with a weight of 16 lbs (camera
system only). Rollvision works on numerous platforms including, but not
limited to, cable systems, cranes, jib arms, camera cars, Steadicams, tracking
and rail systems and much more. Mounting options consist of a 4" square
pattern 1⁄4"–20, Mitchell Mount adapter receiver 1⁄4"–20 and two adjustable
T-nuts sized from 1⁄4"–20.
The Rollvision features electronic end stops, an hour meter with total sys-
tem running time, lock for pan and tilt, removable sphere covers, film and
video mode switch, and voltmeter.
Camera options include: Arri 435, 35-3, 2-C, 2-C SL Cine, 16SR-2 and
16SR-3, Aaton XTRprod and A-Minima plus multiple Sony and Panasonic
High-Definition cameras.
Lenses up to 6 lbs can be used, including but not limited to, Panavision
Primos from 17.5mm to 35mm, Ultra, Super and Normal Speeds; 35mm
primes; and 16mm primes. Lens mount options include Arri standard, light-
weight PL and lightweight Panavision mounts.
Several types of film magazines can be used including 400' SL Steadicam
Mag, 200' Standard and 200' Steadicam Mag.
Remotevision,also designed by Coptervision, is the wireless, 4-in-1 radio
control used to operate the Rollvision. Remotevision features a 5.6" color
monitor and pan, tilt, and roll controls as well as camera zoom and iris con-
trols. Gyro on/off and roll/record switches are included. The speed of the
pan, tilt and roll is adjustable and a reverse switch allows the head to reset to
the zero position.
The Rollvision fits in one to two ATA-approved size cases, and travels with
the Rollvision operator as luggage.
GlIdeCAm InduStrIeS, InC.

Gyro Pro
The Gyro Pro camera stabilizer units liberate you to film while moving
in any direction, even over rough ground, on rough rivers, in high seas or
turbulent air. Accommodates most cameras weighing up to 100 lbs. Sets up
in 20 minutes and features 360° pan, 60° tilt and 30° roll. 36V DC power is
required. Ships in three cases totaling 310 lbs.
182
|
lIBrA heAd, InC.

libra III
The Libra III is a three-axis, digitally stabilized camera mount. It can
be operated in either direct (digital) mode, or the operator can choose to
stabilize one, two or all three axis at the flick of a switch. Custom-designed
motors give it the precision of a well-tuned geared head that is extremely
stable. Wet and muddy conditions are well tolerated. It fits on all industry-
standard cranes and gripping equipment and either can be suspended or
mounted upright. Fail-safe is comprised of two-systems-in-one electronics.
It pans 350°, tilts 90° and rolls 90°. The Libra III comes with Preston FIZ
remote-control system, a lens witness cam, 14" and 8" Sony color monitors,
and AC and DC power supplies.
motIon PICture mArIne

hydro Gyro
A state-of-the-art digital-stabilization head that enables shot-making in the
harshest of filming environments. The Hydro Gyro mounts under your pan/
tilt head or camera. The result is a stable horizon without the pitch and roll
generated by camera boats, cars, aircraft, dollys or other moving platforms. It
weighs 32 lbs with a shipping weight around 70 lbs. It handles camera systems
up to 150 lbs and can use lenses up to 500mm. The Hydro Gyro is waterproof
up to 30'. Fine vibration can be reduced with a Hydro Gyro Shocksorber, a
lightweight, 2" high, anti-shock and vibration damping system.
mInIAture AerIAl SyStemS

CoPtervISIon
CvG-A
The CVG-A is an autonomous, unmanned, small-scale helicopter plat-
form and camera system. Possessing the same characteristics as Coptervi-
son’s line-of-sight, remote control helicopter (5' in length with a 6' blade
span and weighing 40 lbs), the CVG-A incorporates a 3-D GPS waypoint
navigation and flight-control system which is programmed through a com-
puterized Ground Control Station (GCS).
Two different camera systems are available—the standard Coptervision
three-axis, gyro-stabilized camera system that carries 35mm, 16mm and
digital video cameras and/or a new smaller, three-axis, gyro-stabilized
gimbal designed by the company called Flexvision that carries smaller video
cameras. The 35mm camera package consists of a modified Arri 2-C and
comes with either an Arri Standard mount, PL mount or Panavision mount.
Lenses include 16mm and 24mm Zeiss prime lenses, with an optional 40mm
anamorphic or Cooke S4 Wide Prime lens.
| 183
Any given flight plan can be preprogrammed, flown and repeated as

many times as is necessary thus allowing flight patterns to be memorized.
This feature allows for scalable shots that can be used in visual effects plates.
A “return to home” command can be input into the flight parameters so
that the helicopter will automatically return back to its starting position if
necessary.
Like the Coptervision line-of-sight helicopter, the CVG-A is modular in
design and has all of the same flight characteristics: up to 75 mph forwards,
45 mph backwards, 35 mph side to side, and hover. Since a digital microwave
downlink is used to bring the signal down to the ground, the CVG-A camera
operator can maintain precise framing while the helicopter is in motion. The
system runs on aviation-grade fuel and does not emit smoke from the ex-
haust while flying backwards. The modular design allows for ATA-approved
boxes so that a total of six cases travel with the crew as luggage. With the
CVG-A, there is increased range (up to 3 miles and 450') and endurance
(from 30–45 minutes).
flying-Cam
The Flying-Cam is an unmanned aerial filming helicopter. The Flying-
Cam aeronautical design, manufacturing quality, safety features and main-
tenance procedures are equivalent to those used in General Aviation. The
vehicle has all the characteristics required for safe operation close to actors.
The system is self-sufficient and combines the use of aeronautics, electron-
ics, computer vision and wireless transmission technologies that are tailor
made to fit in a 30 lbs take-off weight platform and a 6' main rotor diameter.
Camera available are: 16mm, S16mm, 35mm, S35mm and various Video
standards including HD and Broadcast live transmission.
The embedded Super 35mm 200' camera is custom made by Flying-
Cam and mounted in a 1' diameter, three-axis gyro-stabilized remote
head. The camera is the integration between an Arri 2-C and an Eyemo.
The movement has been designed to achieve the same steadiness of an
Arri 35-3 up to 30 fps and of an Arri 2-C up to 50 fps. The 200' magazine
use standard core and is part of the camera body. When short reloading
time is requested, the cameras are hot-swappable. The electric camera
motor is crystal-controlled with two-digit precision. Ramping is optional.
The minimum speed is 4 fps, the maximum 50 fps. Shutter is 160°. Cam-
era trigger is remote. Indicators and readouts—timer, sync signal and roll
out—are monitored from the ground control station. A color LCD moni-
tor and a digital 8mm tape recorder are provided for monitoring playback
on the ground. A color wireless video assist, used as parallax, gives pe-
ripheral information, allowing for anticipation in the camera remote head
operation. A frameline generator is provided with prememorized ratio.
184
|
The mattebox used is an Arri 3"x3". ND3-6-9, 85ND3-6-9, Pola and 81EF
filters are provided as standard. Aperture Plate is full and optical axis is
centered on full. Available lenses are wide-angle: Cooke, Zeiss and ana-
morphic 40mm. Aperture is remote, focus remote is optional on standard
lens and included on Anamorphic. Lens mounts are Arri Standard, PL,
BNCR, and Panavision.
The HD Camera is a 3-CCD with HD-SDI output and onboard recording
capability.
The Flying-Cam gyro-stabilized patented Remote Head includes one-of-
a-kind top-shot horizon control. Pan: 360° unlimited, 180°/sec adjustable
from ground. Roll: 360° unlimited, 60°/sec adjustable from ground. Tilt:
190º including 90º straight up and 100° down, 60°/sec adjustable from
ground. On 90° tilt down with the 16mm Zeiss lens the front glass is 1"
above ground. The maximum height above ground is 300' (100m) to re-
spect FAA safety separation with general aviation. Flying-Cam provides
the wireless Body-Pan® proprietary system: the camera operator has the
option to take control of the tail rotor giving an unlimited unobstructed,
gyro-stabilized 360° pan capability. The pilot has a transparent control and
override authority if needed.
The Flying Platform has a top speed of 75 mph (120 kph) and can com-
bine all the moves of a full-size helicopter. Take off weight of 30 lbs (15kg)
reduces down wash to a minimum. Autorotation capability is enhanced
by high rotor inertia. Tail rotor is overpowered for Body-Pan operation.
The flight is controlled by a Flying-Cam licensed Pilot using a computer-
base radiodensity. Density altitude is affected by temperature, humidity
and pressure. The radio range is more than one mile. The maximum flight
altitude is 14,000'. The practical range is the visual one. Pilot-in-relay op-
erations are optional.
SuSPended CAmerA SyStem

SPyderCAm
Spydercam is a suspended camera system that allows filmmakers to move
their lens in safe, precise, multi-axis paths through space by combining
highly specialized rigging components with computerized winch systems
and an experienced crew.
• Speeds in excess of 70 mph
• Distances in excess of 4000' long, 600' high
• Compatible with most modern remote heads (Libra, Stab-C, Wes-
cam, Flir, SpaceCam, etc.)
• Video, HD, 16mm, 35mm, VistaVision, IMAX
• Configurations for one-, two- or three-dimensional envelopes
| 185
• Seamless integration with previz systems

• Live fly or preprogram movements interchangeably on any axis. (X, Y,
Z, pan, tilt, roll, focus, zoom)
• Small studio systems to giant outdoor runs
• Highly accurate (some 3-D systems are actually capable of double-
pass motion control work)
• Modular, portable.
The system offers greater accuracy and allows closer proximity to sets,
actors and crew than a helicopter can safely negotiate and offers a much
greater speed and range of motion than can be provided by a standard
camera crane. It a incorporates a 3-D GPS waypoint navigation and flight-
control system which is programmed through a computerized Ground
Control Station (GCS).
CABleCAm
A scaleable, multi-axis camera and stunt platform for film and HD. Cable-
cam scales its rigs in interior and exterior locations. Truss, track, ropeways,
towers, and crane attachments are modular.
Types of rigs include,
Basic: One-axis point to point.
Elevator Skate: Two-axis XZ travel.
Flying V: Two-axis XZ travel.
Traveling High Line: Three-axis XYZ travel using a gantry-like rail and
rope rig.
Dual High Line: XYZ travel using parallel high lines.
Multi-V: Three-axis XYZ travel using four attachment points to construc-
tion cranes or high steel.
Mega-V: The upside down crane. XYZ flight of a 20' vertically telescoping
pedestal. Attached to the bottom of the pedestal is a horizontal 12'ex-
tension/retraction arm which pans, tilts, and rolls. The camera head of
choice is attached at one end of the arm.
Teathered Ballooncam: One-three-axis of aerial control over expansive
distances.
Speed range of systems: 0–70 mph. Length, width, and height of rigs:
100– 5,000'. Propulsion: electronic or hydraulic. Software: Kuper and
custom.
Cablecam’s Kuper or custom, controlled servo-driven camera and stunt
rigs are modeled in Maya and Soft image and exported to a previz model.
“Virtual” Cablecam is flown in the previz milieu using teach and learn
protocol.
Aerial platforms, balloons, and flying 3-D cranes support the Libra, Stab-
C and XR as well as the Cablecam LP (low profile) stabilized HD head.
186
|
Lightweight, mobile operating stations. Repeatable motion control of both

hydraulic and electric servos. Noise-free communications over long distanc-
es. Custom joystick control. Interfaces with other motion control systems
for control of steppers, servos, relays, lighting, cameras and camera heads.
Remote monitoring of servo drive functions. Mechanical and electronic
braking.
| 187
Previsualization
by Colin Green
(Incorporating material formulated and approved by the

ASC-ADG-VES Joint Technology Subcommittee on Previsualization
and adopted and maintained by the Previsualization Society)
I. IntroductIon to PrevIsualIzatIon
Previsualization, also known as “previz,” is a collaborative process that gen-
erates preliminary versions of shots or sequences, predominantly using 3-D
animation tools and a virtual environment. It enables filmmakers to visually
explore creative ideas, plan technical solutions and communicate a shared
vision for efficient production. Filmmakers can leverage the previz process to
effectively work out and convey the details of their vision to the people that
will help execute it. The process allows sets, stunts, visual effects, sequences of
action, story flow and other elements to be designed, developed, understood,
refined and planned in advance using the visual power of 3-D modeling and
animation and with a high degree of production-specific accuracy.
The prehistory of previsualization prior to the availability of computer-
based imaging includes many years of creative solutions using a variety
of physical tools. Popular techniques included mocking up shots and key
elements with materials such as foam core and prefilming the “action” with
lipstick video cameras. Digital previsualization first developed in the early
1990s in tandem with major milestones and enhancements in computer
animation software and processing speed. Professionals around the industry
quickly embraced these enhancements, with the majority utilizing the new
3-D animation capabilities towards creating stunning, never-before-seen
synthetic creatures and imagery in visual effects. Concurrently, a smaller
group was applying the same advancements to improve fundamental film
design, planning and production techniques to create what is now called
previz.
The earliest innovations in digital previz were made by pioneers at stu-
dios such as Sony Pictures Imageworks, Lucasfilm/ILM and The Trumbull
Company. These innovators focused on large, visual effects-driven projects,
paving the way for a wider use of previz with an expanding pool of previz-
specific artists. Today previz is used on projects of all genres where a high
degree of control over creative execution or advanced problem solving is
sought. Previz continues to map to the needs of the market as the demands
188
|
Previsualization
of shooting and shot completion become more challenging, time-critical

and complex.
In 2009, the Previsualization Society, a permanent and ongoing previsu-
alization organization inspired by groundwork from the interdisciplinary
ASC-ADG-VES Joint Technology Subcommittee On Previsualization, was
formed. The Previsualization Society provides an aggregate of resources on
previz and offers membership to all disciplines that interface with the previz
process. Materials maintained by the society and collected on its website,
www.previssociety.com, include a library of previz examples, tips for using
previz, best practice guidelines, professional forums and further links.
II. defInItIons and tyPes

The term “previz” has seen its fair share of differing meanings and us-
ages over the years. The description in the first paragraph of section I above
provides an accepted formal definition. A number of distinctive subtypes of
previz have also been defined:
Pitchvis illustrates the potential of a project before it has been fully funded
or greenlit. As part of development, these sequences are conceptual, to be
refined or replaced during preproduction.
Pitchvis is often used with a more conceptual intention. Because it is
typically created before a production staff is even assembled, pitchvis departs
from the main definition of previz in that it is not specifically relevant to
physical production or reflecting input and collaboration with key crew.
Technical Previz incorporates and generates accurate camera, lighting, de-
sign and scene layout information to help define production requirements.
This often takes the form of dimensional diagrams that illustrate how partic-
ular shots can be accomplished, using real-world terms and measurements.
Technical previz may include virtual stage planning in which correct rep-
resentations of actual shooting conditions can be defined, manipulated and
studied to yield greater production efficiency and a more informed crew.
This can include virtual representations of stages, blue- or greenscreens,
camera cranes, motion control equipment and other gear. Technical previz
can also provide motion control move exports, where data files containing
the exact coordinates of the camera for each frame of the shot are exported
and uploaded into motion control camera systems to precisely replicate
the previsualized motions by the camera capturing the photography. The
cinematographer should be consulted during this stage in order that the
previz sequence has a harmonious flow with the project’s visual objectives
as a whole.
On-set previz creates real-time (or near real-time) visualizations on loca-
tion to help the director, cinematographer, visual effects supervisor and crew
| 189
Previsualization
quickly evaluate captured imagery. This includes the use of techniques that
can synchronize and composite live photography with 2-D or 3-D virtual
elements for immediate visual feedback.
Rather than delivering a specific end previz product or finished sequence,
the purpose of on-set previz is to leverage previz assets to provide a quick
mechanism for visualizing and evaluating any number of “what if?” sce-
narios. Decisions on trying out a specific setup can be made effectively, for
example, before heavy rigging or equipment is moved. On-set previz can
additionally serve a purely creative function as instant “postviz,” with the
video tap image being integrated into the previz sequence to immediately
see how the shot footage works.
Postviz combines digital elements and production photography to validate
footage selection, provide placeholder shots for editorial, and refine effects
design. Edits incorporating postviz sequences are often shown to test audi-
ences for feedback, and to producers and visual effects vendors for planning
and budgeting.
Postviz is typically driven by the editorial process and is especially essen-
tial when there are key plot points, CG characters, etc. that are dependent
on being realized through visual effects. Postviz, typically a combination of
previz combined with principal photography, bridges the gap to provide im-
mediate visual placeholders for storytelling purposes until final visual effects
elements arrive. Postviz often also provides helpful information for visual
effects completion and integration.
III. the PrevIsualIzatIon Process steP By steP

The typical previz workflow can be described by the following phases:
asset assembly
Working from a script, treatment, story outline or storyboards, key scene
elements comprising the main visual subject matter of the sequence are
constructed accurately in 3-D in the computer. The elements can include
characters, sets, vehicles, creatures, props—anything that appears in the
sequence being previsualized. The assets are ideally sourced through the
Art Department (e.g., as concept art, from storyboards or from physical
or digital models) and have already been approved by key creatives on the
project. If specific assets, such as locations, are still being determined, best
approximations are made.
sequence animation
The assets are laid out in space and time using 3-D software to create an
animated telling of the story’s action. The resulting previz sequence is re-
190
|
Previsualization
viewed by key creative personnel and is further tweaked until accurate and
approved.
virtual shoot
Virtual cameras are placed in the scene. Previz artists ensure that these
cameras are calibrated to specific physical camera and lens characteristics,
including camera format and aspect ratio. This step allows the animated
scene to be “seen” as it would from actual shooting cameras. These views can
be based on storyboard imagery or requests from the director or cinematog-
rapher. It is relatively easy to generate new camera angles of the same scene
or produce additional coverage without much extra effort.
sequence cutting
As previz shots/sequences are completed, they can be cut together to re-
flect a more cinematic flow of action, and are more useful as the entirety of
the film is evaluated and planned. Changes required for previz sequences
often are identified in the editing process.
study and review

The completed previz scene, with animation and virtual cameras, can be
studied to consider many useful production aspects: configuration of the
sets, speed of the action, frame composition, stunt planning, camera motion,
rigging and equipment requirements and so forth.
The relevance of the previz is contingent upon including as many actual
production specifications and limitations as possible. The previz process
should inherently incorporate information from key collaborators, such as
the cinematographer and art director, who influence the process and provide
input to ensure the desired results can actually be achieved. In cases where
information changes or new details come into play later in the process, the
previz team may re-tool the sequence to fit within the new specs.
crew communication and outreach

A key part of the previz process is distributing the previz sequences, once
they have been approved by the director, to multiple production depart-
ments. The previz can be “communicated” in a number of forms—diagrams,
documents, overview movies, stage layouts, visual effects breakdowns, etc.
This information can be broken down by each department to ascertain the
achievability of the previsualized shots.
additional Problem solving

During production, the previz team may take on a “rapid response” prob-
lem-solving role by working to address evolving planning and coordination
| 191
Previsualization
needs that would benefit from on-set previz. This team can continue to be
effective after the shoot is complete in a postviz capacity, using previz assets
to make temp shots that accelerate the editing process and streamline the
production of final visual effects.
Iv. some Benchmarks for successful PrevIz

Taking into consideration the idea that previz by nature is a constantly
evolving collaborative process, it is difficult to describe a particular previz
sequence as specifically “finished” and successful. But there are some basic
measures that are all equally important in measuring the success of previz
for any given project.
1. The previz represents an approved creative road map. The director can
watch a previz sequence and is confident that it expresses the film’s in-
tended story and creative vision.
2. The previz represents an approved technical road map. The previz is
based on a foundation of the most accurate and current representations
of sets, locations, production conditions and incorporates input from
across many production departments.
3. The previz is distributed across the production and viewed as achievable.
Previz sequences have been evaluated and provided as needed, and ev-
eryone on the production that the director has chosen to help realize the
project feels it can be done as planned in both the practical and budgetary
sense.
4. The previz is completed early enough to be effective. Previz sequences are
done in time to affect the actual production of the shots.
v. PrevIz and the cInematograPher

Input from the cinematographer, both from a technical and an aesthetic
perspective, is a valuable part of making the previz process beneficial to a
production. As early as possible, making the previz team aware of the shoot-
ing specs for the project, including the expected camera format, lens package
and shooting aspect ratio, will help ensure that these critical parameters are
incorporated into the previz work.
Previz serves a number of purposes for filmmaking, but it cannot simu-
late all aspects of motion picture photography. Previz in its normal form
offers approximations of lighting conditions only in a general sense, not as
the basis for determining lighting choices. Producers should note that, even
though the cinematographer may be consulted for technical information
during the previz process, previz sequences are not intended to represent
the cinematographer’s visual approach to the lighting, color and general
aesthetic of the film.
192
|
Previsualization
| 193
3-D Stereoscopic
Cinematography
by Robert C. Hummel III
Updated from the April 2008 issue of American Cinematographer magazine
T he recent resurgence in stereoscopic motion pictures, more often called

3-D, is completely due to the application of digital technologies in both
exhibition and production. As a result, they are being taken more seriously
than ever before.
In exhibition, the advantage is digital projection that allows a single pro-
jector to project alternating left-eye/right-eye images at a fast enough frame
rate that allows the images to be integrated by our brains into a continuous
24 fps 3-D motion picture. Also, because of digital imaging, it is allowing
filmmakers to explore higher frame rates in 3-D. The Hobbit was photo-
graphed at 48 fps, and James Cameron has announced his Avatar sequel will
be presented at 60 fps.
In production, the smaller form factor of many digital cameras allow for a
more compact 3-D camera rig that was previously not possible. 35mm beam
splitter rigs were virtually impossible for one person to lift, yet today’s HD
rigs are easily hand held by one person.
This chapter is intended as a primer on the basic constructs of this com-
plex medium, and should give you a solid foundation from which to pursue
further research on the subject. In order to save some ink, we will cave to
convention and henceforth refer to stereoscopic cinematography as “3-D.”
Before we begin the discussion, let’s first define some terms.
Left Eye: Refers to photographed images that are intended to be seen only by
a person’s left eye. This term can refer to the lens or camera that is cap-
turing the left-eye images, or the left-eye images projected in a theater.
Right Eye: Photographed images that are intended to be seen only by a per-
son’s right eye. In 3-D Cinematography, often called the master eye, or
dominant eye.
Stereoscopic: Refers to the dual imagery obtained when viewed from two
vantage points slightly offset horizontally from one another. Quite sim-
ply, it is what we observe when viewed with our left and right eyes, and
gives a sense of dimensionality to objects closer than 19' to 23' (6 to 7
meters). Also called “binocular vision.”
Monocular Depth Perception: Refers to depth perception not requiring
dual image cues, or the depth perception that comes into play with
194
|
3-D Cinematography
objects farther than 19'

to 23' away.
Convergence: What hap-
pens with the human
visual system as two im-
ages seen with the left
and right eyes become
overlaid so they be-
come one image. When
looking at an object at Figure 1. When we look at something on the horizon,
infinity, your eyes are our eyes are focused at infinity and look straight
ahead.
looking straight ahead
(Figure 1).
Convergence hap-
pens when stereoscopic
depth perception comes
into play, i.e., when ob-
jects you are targeting/
focusing on are closer
than 13'-16' (Figure 2).
When focused at infin- Figure 2. Eyes converging on a close object.
ity, objects close to you
appear as two trans-
parent images; as you
converge on those close
objects, they become
one solid image, and
objects in the back-
ground become double
images. Convergence
in 3-D cinematography Figure 3. Interocular is the distance between the left
is when the two taking and right eyes. In 3C Cinematography, the distance be-
tween the LEFT and RIGHT taking lenses is called the
lenses are aimed to con- interaxial, and is the variable distance between the
verge on a single point left and right taking lenses, adjusted for each scene.
in space.
Screen Plane: The position in a theater where the projection surface is
located; a vertical plane coincident with the screen that helps define
where objects appear in front of, behind, or on the screen.
Plane of Convergence: The vertical plane where your eyes are directed to
converge on a 3-D object. If an object appears to be floating in front of
the movie screen, the plane of convergence is where that object appears to
be. The same would apply to objects appearing to be “behind” the screen.
| 195
3-D Cinematography
Proscenium Arch: For our purposes, this refers to the edge of the screen
when an object becomes occluded.
Interocular (also called Interaxial): The distance between your eyes
(Figure 3) is properly referred to as interocular. In 3-D cinematogra-
phy, the distance between the taking lenses is properly called interaxial;
however, more recently, you will find filmmakers incorrectly referring
to the distance between taking lenses as the interocular. In 3-D cinema-
tography, if done properly, the interaxial distance between the taking
lenses needs to be calculated on a shot by shot basis. The interaxial
distance between the taking lenses must take into account an average
of the viewing conditions in which the motion picture will be screened.
For large screen presentations, the distance is often much less than the
distance between an average set of human eyes. Within reason, the in-
teraxial can be altered to exaggerate or minimize the 3-D effect.
The 3-D cinematographer must weigh several factors when determining
the appropriate interaxial for a shot. They are: focal length of taking lenses,
average screen size for how the movie will be projected, continuity with the
next shot in the final edit, and whether it will be necessary to have a dynamic
interaxial that will change during the shot.
Because the interaxial distances are crafted for a specific theatrical presen-
tation, a 3-D motion picture doesn’t easily drop into a smaller home viewing
environment. A movie usually will require adaptation and modification of
the interaxial distances in order to work effectively in a small home theater
display screen environment.
The facts presented in this chapter are indisputable, but once you become
enmeshed in the world of 3-D, you will encounter many differing opinions
on the appropriate ways to photograph and project a 3-D image. For ex-
ample, when you’re originating images for large-format 3-D presentations
(IMAX, Iwerks, etc.), some people will direct you to photograph images in
ways that differ from the methods used for 1.85:1 or 2.40:1 presentations.
Part of this is due to the requirements for creating stereoscopic illusions in
a large-screen (rather than small-screen) environment, but approaches also
derive from personal preferences.
Many think stereoscopic cinematography involves merely adding an ad-
ditional camera to mimic the left-eye/right-eye way we see the world, and
everything else about the image-making process remains the same. If that
were the case, this chapter wouldn’t be necessary.
First off, “3-D” movies are not actually three-dimensional. 3-D movies
hinge on visual cues to your brain that trigger depth stimuli, which in turn
create an illusion resembling our 3-D depth perception. In a theatrical envi-
ronment, this is achieved by simultaneously projecting images that represent,
respectively, the left-eye and right-eye points of view. Through the use of
196
|
3-D Cinematography
glasses worn by the audience, th e left eye sees only the left-eye images, and
the right eye sees only the right-eye images. (At the end of this chapter is a
list of the types of 3-D glasses and projection techniques that currently exist.)
Most people believe depth perception is only created by the use of our
eyes. This is only partially correct. As human beings, our left-eye/right-eye
stereoscopic depth perception ends somewhere between 19' to 23' (6 to 7
meters). Beyond that, where stereoscopic depth perception ends, monocular
depth perception kicks in.
Monocular depth perception is an acquired knowledge you gain gradually
as a child. For example, when an object gets larger, you soon learn it is get-
ting closer, and when you lean left to right, objects closer to you move side to
side more quickly than distant objects. Monocular depth perception is what
allows you catch a ball, for example.
3-D movies create visual depth cues based on where left-eye/right-eye
images are placed on the screen. When you want an object to appear that it
exists coincident with the screen plane, both left- and right-eye images are
projected on the same location on the screen. When photographing such a
scene, the cinematographer determines the apparent distance of the screen
plane to the audience by the width of the field of view, as dictated by the focal
length of the chosen lenses. For example, a wide landscape vista might create
a screen-plane distance that appears to be 40' from the audience, whereas
a tight close-up might make the screen appear to be 2' from the audience.
Screen Plane
Audience
Figure 4. Eyes converging on an “on screen” object. As seen from above, as if viewing from
the theater’s ceiling, looking down on the audience and the screen plane.
| 197
3-D Cinematography
Apparent location
of object to audience
Left Eye Image on screen Right Eye Image on screen
Screen Plane
Audience
Figure 5. How a behind screen object is created.
Figure 4 illustrates when an object is at the screen plane and where the au-
dience’s eyes converge while viewing that object. (Figure 4 also effectively
represents where your eyes converge and focus when watching a standard
2-D movie without special glasses).
If we want an object to appear behind the screen, the image is photo-
graphed with the lenses converged behind the screen plane. On set, the
screen plane is an invisible plane that you establish to control where objects
will be perceived by the viewer of the 3-D motion picture. In the theater, of
course, the screen plane is a very real, physical object.
When the object you want to appear behind the screen is projected, it
appears similar to what is presented in Figure 5.
In Figure 5, the right-eye and left-eye images are kept separated by the
special glasses worn by the audience; in other words, the left eye sees only
the left-eye image and the right eye sees only the right-eye image. If you were
to remove your glasses, you would see both images simultaneously, as in the
Figure 6 example of a behind-screen object viewed without special glasses.
Next, we want an object to appear in front of the screen plane so that from
the audience’s perspective, the object appears to be coming into the theater
198
|
3-D Cinematography
and closer to the viewer’s face.

This is achieved on set by ad-
justing the convergence of the
left- and right-camera lenses
so they are converging in front
of the in theater screen plane.
When projected, the images
are viewed by the audience as
This technique can be used
Figure 6. A projected 3-D image viewed without
to make audience members special glasses.
perceive that an object is
very, very close to their faces. This creates a very effective 3-D illusion, but
experience has shown that extreme examples of this effect should be used
sparingly, if at all. Remember that while the viewer will be converging that
object mere inches from his face, he will still be focusing on the screen plane
many meters away. As a result, this type of 3-D “gag” (when properly done)
always gets gasps from an audience yet, because of that disparity of focus,
never quite matches reality.
This example illustrates an important difference between 3-D movies and
what you experience in real life. In life, when an object is half a meter from
Right Eye Image on Screen Left Eye Image on Screen
Screen Plane
Apparent location
of object to audience
Audience
Figure 7. How an object appearing in front of the screen is created.
| 199
3-D Cinematography
Left Eye Image on Screen Right Eye Image on Screen
Screen Plane
Emily
Audience
Claire
Figure 8. On-screen objects are seen in the same location by all audience members.
your face, your eyes converge and focus at half a meter from your face. In
a 3-D movie environment, you can choose an angle of view and scale that,
from your perspective, makes an object appear to be half a meter from your
face even as your eyes are focused on the screen plane, which may be any-
where from 5 to 30 meters away from you.
This doesn’t mean the 3-D approach is “wrong;” it’s just an example of
why 3-D depth cues in a 3-D movie often seem to be exaggerated—why 3-D
movies seem to be more 3-D than reality.
When an object appears on the screen plane, every member of the audi-
ence sees the object at the same location on the screen because the left- and
right-eye images appear precisely laid on top of each other (and thus appear
as one image). Basically, the image appears the same as it would during a
regular “2-D” movie projection (Figure 8).
Take a look at Figure 9, however, and see how things change when an
object is placed behind the screen plane. As you can see, a person’s specific
location in the theater will affect his perception of where that behind-screen
object is located. Also, how close that person is to the screen will affect how
far behind the screen that object appears to be; the closer one’s seat is to the
screen, the shorter the distance between the screen and the object “behind”
it appears to be.
Again, it is not “wrong” that this happens. Figure 9 simply clarifies the
point that stereoscopic cinematography is not 3-D. Were it truly 3-D, every
audience member would see these behind-screen objects in the same loca-
200
|
3-D Cinematography
Apparent location
Apparent location of object to Claire
of object to Emily
Right Eye Image on Screen

Left Eye Image on Screen
Screen Plane
Emily
Audience
Claire
Figure 9. Audience position affects both lateral and depth convergence of behind-screen
objects.
tion. When planning shots for a 3-D motion picture, the filmmaker should
be conscious of how a dramatic moment might be received by viewers seated
in various locations. Audience position will also affect the perceived location
of off screen objects as well.
My next points concern the proscenium arch and “off-screen” objects. As
mentioned earlier, the edges of the screen image (top, bottom, left and right)
are collectively referred to as the proscenium arch. This is a nod towards
live theater, where the term applies to that area of the stage in front of the
curtain. In 3-D, the term is used when referring to objects that appear to be
in front of the screen plane.
In short, the edges of the screen are relevant to objects appearing in front
of the screen plane. Such an object can have very strong stereoscopic con-
vergence cues that will make it appear to be “floating” very close to a viewer’s
face. A good example of this phenomenon occurs in the film Muppet*vision
3-D, during a scene in which the characters Waldo and Kermit the Frog ap-
pear to extend into the audience while clinging to the end of a ladder. Many
recent 3-D motion pictures avail themselves of this 3-D principle, from
Beowulf, when a spear is thrust toward Beowulf ’s face after he arrives on a
beach, to dogs lunging towards the camera in Hugo.
| 201
3-D Cinematography
However, if that floating object moves so close to the edge of the screen
that it is occluded by that edge, your brain will quickly employ its knowledge
of monocular depth cues, and your perception that the object is floating in
front of the screen will diminish to the point of inconsequence. Your brain
has learned that when one object is occluded by another, the occluded object
must be farther away. In spite of all the stereoscopic depth cues, your brain
knows that if that object is occluded by the edge of the screen, then it must
be at, or behind, the screen plane. This scenario will be very noticeable to
viewers as their brains attempt to sort out these contradictory depth cues.
Monocular depth perception overrules stereo depth cues because we are
hard-wired to protect ourselves from danger. Because most danger (such as
an approaching lion, bear or saber-toothed tiger) starts from outside our
stereoscopic depth zone, it’s easy to understand how the brain defaults to
the depth cues that govern most of our life. The 3-D axiom to remember is
that off-screen objects should never touch the edge of the screen, because
if they do, the illusion will be disrupted. The illusion is most effective with
objects that can float or be thrust toward the audience. You will also no-
tice that when you experience these illusions, filmmakers are keeping the
off-screen objects closer to the center of the screen in order to avoid the
proscenium arch.
As with many axioms, however, there is sometimes an exception. There
is a scenario in which an occluded object can still appear as though it is
coming off the screen. Imagine a medium shot of a man who walks from
behind the screen plane toward the screen plane, and then continues toward
the audience until he is in front of the screen. Surprisingly, this shot will still
work with the character apparently coming off the screen, even though the
lower half of his body is cut off by the bottom of the projected image. The
requirement for it to work, contrary to our earlier axiom, is that the viewer
must have other audience members in front of him, with the bottom of the
screen occluded by people’s heads. When the bottom of an object is occluded
by people very close to you, your brain will still believe the object is getting
closer. However, even a clear view of the bottom of the screen will result
in a fairly good effect of the man coming off of the screen; because we’re
programmed to look straight ahead, and often don’t see, or focus on the
lower half of a person coming towards us, obscuration of the lower half of a
person usually won’t entirely ruin the off screen effect.
One must also be aware of the constraints on editing in 3-D. This concept
is relatively simple to grasp but is often disregarded to the detriment of a
3-D presentation. When editing for 3-D, it is important to consider the con-
vergence extremes that the audience will experience in order to realize the
stereoscopic illusion. For example, if the audience is viewing action that oc-
curs behind the screen plane, it is inadvisable to then cut directly to an object
202
|
3-D Cinematography
in front of the screen. The average viewer will have difficulty converging the
suddenly “close” object, to the point where he might see double images for
several moments.
Experienced viewers of 3-D films won’t have this problem, and this can
lead to mistakes if you happen to be part of the creative team involved in a
3-D production. If you work extensively in post for 3-D movies, you become
more and more adept at quickly converging disparate objects. However,
your audience won’t have the advantage of exercising their eyes as much as
someone working on a 3-D film. If this disparity isn’t taken into account,
the resultant movie can cause problems for the audience. The filmmakers
will have no trouble watching it, but the average viewer will find it difficult
to converge 3-D images that cut between extreme positions in front of and
behind the screen plane.
These extremes of convergence can mitigated to a great degree in the
digital intermediate or by the post handling of the 3-D images. A technique
of “handing off ” the A side of cut to the B side by quickly easing the con-
vergence between the two cuts can be very effective at making the cut much
easier for the audience to watch.
Some 3-D films attempt to guide the viewer to converge objects in front of
the screen. They do this by slowly bringing an object closer to the audience,
allowing viewers to track the object as it comes farther and farther off the
screen. The makers of the theme-park attraction Captain EO accomplished
this with a shot of a floating asteroid that comes off the screen at the begin-
ning of the film. In Muppet*vision 3-D, the effect is created with the simple
gag of a “3-D” logo positioned at the end of a broomstick that is pushed into
the audience’s face; the effect is repeated at the end of the film with the shot of
Kermit perched at the end of a fire truck’s ladder. In Terminator2®: 3-D, Robert
Patrick’s molten head slowly comes out off the screen towards the audience.
The glaSSeS
Creating the illusion of true stereoscopic imaging requires that the left
eye and right eye see unique left/right imagery. This is achieved with various
kinds of glasses that perform the task for the viewer. It should be noted that
all 3-D glasses—and projection systems—have an impact on the brightness
of the projected image, generally a light loss on the order of 2–3 stops.
The first public use of polarized glasses for a 3-D motion-picture presenta-
tion took place at the 1939 New York World’s Fair. Called linear polarizers,
these glasses work by orienting the left- and right-eye polarizing material at
right angles to each other. Though this is not actually visible to your eye, the
effect is graphically illustrated in Figure 11.
In this system, the left- and right-eye projected images are each projected
through linear polarizers that match the orientation of the glasses. This meth-
| 203
3-D Cinematography
od is effective at separating
the left and right image ma-
terial; however, if the viewer
tilts his head at an angle,
there will be “leakage” of the
image information, allowing
the viewer to see both im-
ages simultaneously, which,
of course, is not ideal. Polar-
izing glasses, and all glasses
that do not require any elec-
tronics in order to function, Figure 10. Linear Polarized Glasses
are called “passive glasses.”
Polarized 3-D presentations
always require projection
onto a silver screen in order
to maintain polarization of
the projected image.
Glasses that employ
circular polarization, with
a casual observation, look
just like their linear cousins
in Figure 10, but perform
quite differently. A simple
explanation is that circular Figure 11. Linear Polarizing Glasses, graphically
illustrated.
polarization can be oriented
in clockwise and counter-
clockwise directions. The
left- and right-eye projected
images are polarized in op-
posing circular-polarization
orientation, as are the
respective lenses on the
glasses. The circular-polar-
ization effect is graphically
A principle advantage of
circular polarization is that
the integrity of the left- and Figure 12. Circular Polarization Glasses, graphically
illustrated.
right-eye image informa-
tion is always maintained, no matter how much the viewer tilts their head.
Circular polarization is the technique employed with the Real D “Z Screen”
204
|
3-D Cinematography
technique, used in conjunc-

tion with single-projector
digital-cinema projection.
A variant on the passive-
glasses front is Dolby 3-D
technology (Figure 13).
Developed in conjunction
with Infitec GmbH, this sys-
tem can use existing screen Figure 13. DOLBY 3-D Glasses.
installations, as it doesn’t
require a silver screen. By
shifting about half a dozen
different RGB “primaries” of
the projected image, the pro-
cess is able to create separa-
tion of the left and right eye
images. Dolby’s glasses then
filter these spectrums of light Figure 14. Anaglyph 3-D Glasses.
so each eye only sees its re-
spective left and right image
information. Dolby’s color-
spectrum technique could
technically be described as
anaglyph; yet it is actually a
much more complex system
that doesn’t cause any of the
color distortion usually as- Figure 15. Liquid Crystal or Active 3-D Glasses.
sociated with anaglyph.
Another type of passive
glasses is the anaglyph
system (Figure 14). The
separation of left- and right-
eye imagery is achieved by
representing the left-eye
material with the cyan
spectrum of light and the Figure 16. Pulfrich 3-D Glasses.
right-eye material with the
red spectrum. When the glasses are worn, those colors are directed to your
left and right eyes, respectively. This system is often mistakenly thought to
be how all 3-D films were viewed in the 1950s; actually, this method was not
used very often, and almost all films in the 1950s utilized polarized projec-
tion and glasses. Over the past few years, it has been revived for some 3-D
| 205
3-D Cinematography
presentations, such as Spy Kids 3-D. This technique works on a standard

white screen and does not require a silver screen.
Active glasses, also called “shuttered glasses,” employ an electronically
triggered liquid-crystal shutter in each lens (Figure 14). This method alter-
nately synchronizes the left- and right-eye projected images with the respec-
tive liquid-crystal shutters contained in the glasses. An advantage of active
glasses is that they do not require a silver screen. The glasses are electronic,
require battery power to function, and usually operate in conjunction with a
transmitted infrared signal that synchronizes the glasses with the projected
left- and right-eye images.
Pulfrich glasses (Figure 16) are definitely the poor man’s choice for 3-D,
because they do not even produce true stereoscopic 3-D. The “Pulfrich Ef-
fect” was first documented by Carl Pulfrich in 1922. This passive eyewear
has one clear lens and one lens that is significantly darker. It operates on
the principle that it takes longer for your brain to process a dark image than
a bright image. The resultant delay creates a faux 3-D effect when objects
move laterally across the field of view. Objects “appear” 2-D until they move
laterally across the frame, which causes an optical illusion that the brain
interprets as a 3-D depth cue. This technique has usually been limited to
broadcast-TV applications such as the short 3-D sequences in a 1990 Rolling
Stones concert broadcast on Fox, and Hanna-Barbera’s Yo Yogi! 3-D.
Is there more to 3-D production than the ground we’ve covered? You bet.
But this chapter should gird you with a foundational understanding of the
medium, allow you to converse in the vocabulary of 3-D, and enable you
to begin to make the medium your own. I also hope you now appreciate
that the complexities of 3-D stereoscopic cinematography should not be
underestimated.
206
|
3-D Cinematography
| 207
Day-for-Night, Infrared and
Ultraviolet Cinematography
by Dr. Rod Ryan
Day-for-NIght
T he speed of modern color films makes it possible to shoot night-for-

night scenes. However, there are night scenes that are impractical to
illuminate artificially and actually film at night. Shooting such scenes day-
for-night eliminates the additional problems and expense of night shooting
and can deliver excellent pictorial results.
Techniques for filming day-for-night scenes in color or black-and-white
vary greatly because of the many factors involved. Cinematographers
naturally differ in their interpretation of what constitutes a night effect. The
overall effect must be one of darkness. Processing laboratories differ in their
negative preferences, although most prefer sufficient density on the original
negative since it is always possible to “print down” for a darker effect, but im-
possible to obtain a rich, full-bodied print from a thin, shadowless original
negative (if black shadows are desired, the scene must print at center scale
or higher).
Choice of filters and degree of underexposure will vary according to sky
conditions; color and contrast of subject and background; the strength,
quality and direction of sunlight; and the particular effect desired. Very
generally speaking, the most convincing day-for-night shots, in either color
or black-and-white, are made in strong sunlight, under blue skies and with
low-angle back-cross lighting.
Direct backlighting results in a “rim-light” effect which, although pleas-
ing in a long shot, lacks the necessary three-dimensional, half-illuminated
facial effects required in medium and close shots. Front lighting will flatten
and destroy all shadows. Side and front-cross lighting is permissible but
not as effective as back-cross illumination. Because production does not
always permit shooting when conditions are exactly right, and since day-
for-night shots must sometimes be made all day long, often the choice of
sun angle must be compromised. Under these conditions, avoid front light-
ing as much as possible and stay with any sun angle that results in partial
illumination, preferably with shadows toward the camera.
Skies give the most trouble, since they will invariably read too high
and are difficult to balance against foreground action. Graduated neutral-
density filters, which cover the sky area only, and Pola Screens, which will
208
|
Day-for-Night, Infrared and Ultraviolet Cinematography
darken the sky with the sun at certain angles, are both useful for either color
or black-and-white films because they do not affect color values and can be
used in combination with other effect filters.
Neutral-density filters will tone down a “hot” sky even if it is bald white.
A partial or graduated neutral-density filter covering only the sky will
therefore be very useful for bringing the sky into exposure balance with the
foreground. Care must be taken, however, that action does not cross the
demarcation line between the filter material and the clear glass area. Pola
Screens are most useful when the sun is directly overhead at right angles to
the camera.
A Pola Screen should not be employed if the camera must be panned
through a wide arc, since the polarization will vary and the sky tone will
change in density as the camera revolves. Typical underexposure is 1½ to
2½ stops, rarely more. Brilliant sunlight will require greater underexpo-
sure, gray days less. The underexposure can be handled in several ways.
One is by ignoring the filter exposure increase required, if it is close to
the amount of underexposure desired. For instance, the filter being em-
ployed may require two stops increase in exposure for a normal effect.
The increase is ignored and the diaphragm set for the exposure without
the filter, thus delivering the necessary underexposure for the night effect.
Or, a neutral density of the desired strength is employed and its exposure
increase ignored.
Proceed as follows: insert the effect filter or combination of filters for the
desired effect, and allow for their exposure increase as in normal filming.
Add the desired neutral (a .30 for one stop, .50 for 1½ stops or a .60 for
two stops). Ignoring the neutral filter’s exposure increase will automatically
underexpose the negative the necessary amount. This is a quick and effec-
tive method in fast production shooting where night effects are suddenly
required and little or no time is available for computations.
If the sky is not sufficiently blue to filter properly, and if it is impossible to
use a graduated neutral-density filter, try to avoid the sky as much as pos-
sible by shooting against buildings or foliage, or choose a high angle and
shoot downward.
The contrast between the players and the background is very important
because a definite separation is desirable. Dark clothing, for instance, will
merge with a dark background, and the player will be lost. It is better to
leave a dark background dark and players in lighter, although not necessarily
white, clothing than to have a light background and players in dark clothing.
The latter combination will result in a silhouette, rather than a night effect.
This is the reason that back-cross lighting is preferable: so the background
is not illuminated and the players have a definite separation through edge
lighting, which also imparts shimmering highlights.
| 209
Black-and-White film
The illusion of night in black-and-white cinematography is obtained by
combining contrast filtering with underexposure. Since the sky is light by day
and dark by night, it is the principal area of the scene requiring correction. Any
of the yellow-orange or red filters may be used. A very popular combination is
the light red Wratten 23A plus the green 56. This combination does everything
the red filters accomplish—plus it darkens flesh tones, which are rendered too
light by the red filters alone. When combining filters, remember that red filters
add contrast but green filters flatten, so if a greater flattening effect is desired
add a heavier green filter. Because flesh tones are not important in long shots,
such shots are sometimes filmed with heavier red filters, and only the medium
and close shots are made with the combination red-green filters. Care must
be taken, however, that clothing and background colors do not photograph
differently when filters are switched in the same sequence. If in doubt, shoot
tests before production filming begins. Remember that only a blue sky can be
filtered down. No amount of color filtering will darken a bald white sky. Use
graduated neutral densities, or avoid the sky under these adverse conditions.
The 23A-56 combination is usually employed with a filter factor of 6, rather
than the 20 normally required (5 for the 23A and 4 for the 56, which multiplied
equals 20). The factor of 6 automatically underexposes this filter combination
approximately 1½ stops and achieves the desired effect without further com-
putation. If a red filter is used alone, bear in mind that it will lighten faces, and
use a darker makeup (approximately two shades) on close shots.
reversal Color film

Typical blue night effects can be obtained with reversal color films bal-
anced for exposure with tungsten light by removing the Wratten 85 filter
and under exposing 1⅓ stops. If the bluish effect is too great, an ultraviolet-
absorbing filter can be used to filter out the excess ultraviolet. Flesh tones
in closeups can be adjusted by using gold reflectors or 3200°K fill lights to
light actors faces. Care must be taken that the actors are not overlit or that
such lights appear as ambient light with the sun acting as a moonlight key.
Negative Color film

A cameraperson shooting day-for-night with negative color film should
check with his or her processing laboratory before the production begins.
Laboratories have a far greater range of color correction available than the
cinematographer has at his disposal during the original photography. They
may add or subtract any color, or combination of colors, provided the origi-
nal negative has sufficient exposure. Once the 85 filter is removed, however,
it is often impossible to restore normal color balance to the film.
If the 85 filter is removed, it should be replaced with an ultraviolet filter,
210
|
which will prevent overexposure of the blue sensitive layer and keep the
negative within printing range. Warmer effects may be obtained by sub-
stituting a light yellow filter for the 85. A Pola Screen may also be used to
darken a blue sky and provide the required underexposure (by ignoring its
filter factor). It will have no effect on a bald sky, but it will act as a neutral-
density filter and provide the needed underexposure. Remember that ap-
proximately ⅔ stop exposure is gained by removing the 85 filter. This must
be included in exposure calculations.
INfrareD CINematography
Because cinematography by infrared light has had limited pictorial use,
this will be a brief review. For more information, refer to Kodak publica-
tions number N-17 “Kodak Infrared Films” and M-28 “Applied Infrared
Photography.” Infrared for photographic purposes is defined as that part of
the spectrum, approximately 700 to 900 nanometers (nm), which is beyond
the visible red but not as far as would be sensed by humans as heat.
All infrared films are sensitive to heat and should be kept refrigerated
before exposure and during any holding time before processing. While no
longer listed as a regular catalogue item, Eastman Kodak still manufactures
a black-and-white infrared-sensitive film, Kodak High Speed Infrared Film
2481, and modified color-sensitive film, Kodak Ektachrome Infrared Film
2236. Both of these films are on Estar base. Before deciding to use either
film in a production, the manufacturer should be contacted regarding its
availability, minimum order quantities and delay in delivery.
Black-and-White films
For pictorial purposes, the greatest use of infrared-sensitive film for
motion-picture photography has been for “day for night” effects. Foliage
and grass reflect infrared and record as white on black-and-white film.
Painted materials that visually match in color but do not have a high in-
frared reflectance will appear dark. Skies are rendered almost black, clouds
and snow are white, shadows are dark but often show considerable detail.
Faces require special makeup and clothing can only be judged by testing.
A suggested EI for testing prior to production is daylight EI 50, tungsten
EI 125 with a Wratten 25, 29, 70 or 89 filter, or daylight EI 25, tungsten EI
64 with 87 or 88A (visually opaque) filter. Infrared light comes to a focus
farther from the lens than does visual light. (Speak to your lens supplier for
correct focus compensation for infrared photography.)
Color
No human can see infrared; color film can only record and interpret it.
Kodak Ektachrome Infrared Film 2236 was originally devised for cam-
| 211
ouflage detection. Its three image layers are sensitized to green, red and
infrared instead of blue, green and red. Later applications were found in
medicine, ecology, plant pathology, hydrology, geology and archeology. Its
only pictorial use has been to produce weird color effects.
In use, all blue light is filtered out with a Wratten 12 filter; visible green re-
cords as blue, visible red as green, and infrared as red. The blue, being filtered
out, is black on the reversal color film. Because visible yellow light is used as
well as infrared, focus is normal, and the use of a light meter is normal for
this part of the spectrum. What happens to the infrared reflected light is not
measurable by conventional methods, so testing is advisable. A suggested
EI for testing prior to production is daylight EI 100 with a Wratten 12 filter.
UltravIolet CINematography
There are two distinctly different techniques for cinematography using
ultraviolet radiation, and since they are often confused with each other,
both will be described.
In the first technique, called reflected-ultraviolet photography, the expo-
sure is made by invisible ultraviolet radiation reflected from an object. This
method is similar to conventional photography, in which you photograph
light reflected from the subject. To take pictures by reflected ultraviolet,
most conventional films can be used, but the camera lens must be covered
with a filter, such as the Wratten 18A, that transmits the invisible ultraviolet
and allows no visible light to reach the film. This is true ultraviolet photog-
raphy; it is used principally to show details otherwise invisible in scientific
and technical photography. Reflected-ultraviolet photography has almost
no application for motion-picture purposes; if you have questions about
reflected-ultraviolet photography, information is given in the book “Ultra-
violet and Fluorescence Photography,” available from Eastman Kodak.
The second technique is known as fluorescence, or black-light, pho-
tography. In motion-picture photography, it is used principally for visual
effects. Certain objects, when subjected to invisible ultraviolet light, will
give off visible radiation called fluorescence, which can be photographed
with conventional film. Some objects fluoresce particularly well and are
described as being fluorescent. They can be obtained in various forms such
as inks, paints, crayons, papers, cloth and some rocks. Some plastic items,
bright-colored articles of clothing and cosmetics are also typical objects
that may fluoresce. For objects that don’t fluoresce, fluorescent paints (oil
or water base), chalks or crayons can be added. These materials are sold
by art-supply stores, craft shops, department stores and hardware stores.
(Many of these items can also be obtained from Wildfire Inc., 10853 Venice
Blvd., Los Angeles, CA, 90034, which manufactures them specially for the
motion-picture industry.)
212
|
Fluorescence may range from violet to red, depending on the material

and the film used. In addition to the fluorescence, the object reflects ultra-
violet light, which is stronger photographically. Most film has considerable
sensitivity to ultraviolet, which would overexpose and wash out the image
from the weaker visible fluorescence. Therefore, to photograph only the
fluorescence, you must use a filter over the camera lens (such as the Wrat-
ten 2B, 2E or 3, or equivalent) to absorb the ultraviolet.
The wavelengths of ultraviolet light range from about 10 to 400nm. Of
the generally useful range of ultraviolet radiation, the most common is
the long-wavelength 320 to 400nm range. Less common is the short to
medium-wavelength range of 200 to 320nm. In fluorescence photography
you can use long-, medium-, or short-wave radiation to excite the visible
fluorescence depending on the material. Some materials will fluoresce in
one type of ultraviolet radiation and not in another.
Certain precautions are necessary when you use ultraviolet radiation.
You must use a source of short- or medium-wave ultraviolet with cau-
tion because its rays cause sunburn and severe, painful injuries to eyes
not protected by ultraviolet-absorbing goggles. Read the manufacturer’s
instructions before using ultraviolet lamps.
Eye protection is generally not necessary when you use long-wave ultra-
violet because this radiation is considered harmless. However, it’s best not
to look directly at the radiation source for any length of time, because the
fluids in your eyes will fluoresce and cause some discomfort. Wearing glass
eyeglasses will minimize the discomfort from long-wave sources.
There are many sources of ultraviolet radiation, but not all of them are
suitable for fluorescence photography. The best ultraviolet sources for the
fluorescence technique are mercury-vapor lamps or ultraviolet fluorescent
tubes. If an object fluoresces under a continuous ultraviolet source, you
can see the fluorescence while you’re photographing it.
Since the brightness of the fluorescence is relatively low, the ultraviolet
source must be positioned as close as practical to the subject. The objective
is to produce the maximum fluorescence while providing even illumina-
tion over the area to be photographed.
Fluorescent tubes designed especially to emit long-wave ultraviolet
are often called black-light tubes because they look black or dark blue
before they’re lighted. The glass of the tubes contains filter material that
is opaque to most visible light but freely transmits long-wavelength ul-
traviolet. These tubes, identified by the letters BLB, are sold by electrical
supply stores, hardware stores and department stores. They are available
in lengths up to 4' and can be used in standard fluorescent fixtures to il-
luminate large areas. Aluminum-foil reflectors are available to reflect and
control the light.
| 213
Mercury-vapor lamps are particularly suitable for illuminating small

areas with high ultraviolet brightness. When these lamps are designed
for ultraviolet work, they usually include special filters which transmit
ultraviolet and absorb most of the visible light. Mercury-vapor ultraviolet
lamps are available in two types, long-wave and short-wave. Some lamps
include both wavelengths in the same unit so that they can be used either
separately or together. If you use a light source that does not have a built-in
ultraviolet filter, you must put such a filter over the light source. The filter
for the radiation source is called the “exciter filter.”
You can use a Kodak Wratten Ultraviolet Filter No. 18A or Corning
Glass No. 5840 (Filter No. CS7-60) or No. 9863 (Filter No. CS7-54) for this
purpose. Kodak Filter No. 18A is available in 2- and 3-inch glass squares
from photo dealers. The dealer may have to order the filter for you. The
Corning Glass is available in larger sizes from Corning Glass Works, Opti-
cal Photo Products Department, Corning, NY, 14830. The filter you use
must be large enough to completely cover the front of the lamp. The scene
is photographed on a dark set with only the ultraviolet source illuminating
the subject. In order for the film to record only the fluorescence, use a
Kodak Wratten gelatin filter, No. 2A or 2B, or an equivalent filter, over
the camera lens to absorb the ultraviolet. When used for this purpose, the
filters are called “barrier filters.” Since the fluorescence image is visible,
no focusing corrections are necessary. Focus the camera the same as for a
conventional subject.
Determining exposure
Many exposure meters are not sensitive enough to determine exposure
for fluorescence. An extremely sensitive exposure meter should indicate
proper exposure of objects that fluoresce brightly under intense ultraviolet,
if you make the meter reading with a No. 2A or 2B filter over the meter
cell. If your exposure meter is not sensitive enough to respond to the rela-
tive brightness of fluorescence, the most practical method of determining
exposure is to make exposure tests using the same type of film, filters and
setup you plan to use for your fluorescence photography.
films
While either black-and-white or color camera films can be used for
fluorescence photography, color film produces the most dramatic results.
The daylight-balanced films will accentuate the reds and yellows, while
the tungsten balanced films will accentuate the blues. Since fluorescence
produces a relatively low light level for photography, a high-speed film
such as Eastman Vision 3 500T (5219) or Eastman Vision 250D (5207) is
recommended.
214
|
SpeCIal CoNSIDeratIoNS
Some lenses and filters will also fluoresce under ultraviolet radiation.
Hold the lens or filter close to the ultraviolet lamp to look for fluorescence.
Fluorescence of the lens or filter will cause a general veiling or fog in your
pictures. In severe cases, the fog completely obscures the image. If a lens or
filter fluoresces, you can still use it for fluorescence photography if you put
the recommended ultraviolet-absorbing filter over the camera lens or the
filter that fluoresces. It also helps to position the ultraviolet lamp or use a
matte box to prevent the ultraviolet radiation from striking the lens or filter.
| 215
Aerial Cinematography
by Jon Kranhouse
A rudimentary understanding of basic aviation is essential for safe film-

ing. When airborne, things can go wrong very quickly. Within respec-
tive aircraft flight envelopes, altitude and speed provide an extra margin of
safety; “low and slow” leaves little margin for error. The immutable laws of
physics are unforgiving and must not be ignored, despite egos, budgets and
schedules. Aerial filming can be exceptionally safe, or turn deadly with poor
judgment.
Air-to-ground and air-to-air filming require the cinematographer to con-
sider the proper type of helicopter or fixed-wing aircraft to use as a camera
platform, as well as the appropriate camera mount. Factors include: match-
ing air speeds of camera aircraft and story aircraft, total weight of people/
equipment, top speed, low speed, range and time aloft, terrain to be flown
over, working altitude, prevailing winds, precipitation, temperature and, of
course, the budget.
While helicopters offer unobstructed views and outstanding maneuver-
ability at low speed and/or very low altitude, they are much more expensive
to operate per hour than most fixed-wing aircraft. Adequate results might
be achieved from a small Cessna, if aircraft architecture doesn’t block the
required view. Many fixed-wing aircraft have greater range and altitude ca-
pability than helicopters.
For maximum stability and safety, select a camera aircraft that will, during
actual filming, be working near the middle of its flight envelope. There have
been times when a Lear jet had to drop its landing gear and flaps during
filming at altitude just to match speed with a slower story aircraft. In another
example that I recall, a helicopter couldn’t keep up with a race car; a fixed-
wing aircraft should have been selected instead.
Sometimes the wrong aircraft is selected because a helicopter pilot doesn’t
want to lose a job to a fixed-wing pilot, or vice versa. At other times, the
producer, director and cameraman were not envisioning the same shots;
there’s no substitute for good prep. Many variants of each aircraft type exist;
if one variant is suitable for your needs, don’t assume they all are. Blimps and
hot-air balloons offer the most vibration-free mounting platforms, but they
are the least maneuverable.
On-board cinematography is often accomplished with a handheld camera,
onboard Steadicam, or by various methods of mounting the camera to the
airframe (under guidelines of FAA #337 certificate; see FAA rules below).
216
|
Skydiving cinematography is a rare specialty of very experienced skydiv-

ers, usually working with helmet-mounted Bell & Howell 35mm Eyemo
cameras or variants. Skydivers can be photographed from the ground with
very long lenses, as well as from helicopters using various stabilization
mounts.
Miniaturized radio-controlled helicopters bearing ultra-lightweight film
cameras and video transmitters have yielded promising results. Depending
on the shots desired, these might be the right tool for the job. Using these
machines above crowds is illegal, their blades are as lethal as those of a full-
sized helicopter.
AesthetiCs
Time of day: Of course, early morning and late afternoon light is desirable
to reveal textures on the ground, and wind is usually calmest at these
times. Consideration must be made for keeping aircraft shadow or
reflection out of the shot. Midday heat can cause wind gusts, affecting
stability. It is common for pilot and camera crew to be aboard a warmed-
up helicopter, parked and “turning” on the ground, awaiting the earliest
flyable light—it is not safe to fly low-to-ground in pitch-black night.
Night filming is perfectly safe at altitude—above the power lines.
Metering: When the sky is clear, an incident light meter is quick and easy;
the light hitting you is the same as that hitting your subject. When shafts
of sunlight poke through stormy clouds, a spot meter is most useful.
Lens hoods for spot meters prevent flare from biasing readings.
Filming speeds: To create shot dynamics, shooting at less than 24 fps is
common. When viewed from the air, most inanimate air-to-ground
subjects (e.g., city traffic, ocean waves and large ships) appear to move
slowly; they can safely be shot at 18–22 fps and still appear quite natural.
Exterior gyro mounts allow for exceptional stability at very slow fram-
ing rates (e.g., 2 fps), but usable pans and tilts must be pre-programmed.
A door mount or handheld may be more appropriate if a less steady
POV is called for. Conversely, in turbulent air with a door mount, slight
overcranking (30–36 fps) can smooth results.
Choice of film stocks/T-stops: Geography and sun conditions may require
changing the f-stop in-shot. It is not unusual to pull from T4 to T22
in one shot; the trick is hiding the change with shot choreography and
a smooth remote aperture motor. Door mounts allow the changing
of filters and magazines in mid-air, allowing for quick adaptation to
changing conditions. Nose and exterior gyro mounts, however, require
a smart compromise to be considered before take-off during sunrise/
sunset periods of rapidly changing light intensity. You don’t want to
have to land to change filters or stocks just when the light is magic.
| 217
Exterior gyro mounts require focus to be a combination of estimations

from the videotap finder, coupled with the estimations of a great assis-
tant. But it doesn’t hurt to have a T16 when doing air-to-air at 250mm.
Low-speed daylight-balanced stocks with UV filtration are excellent for
aerials if you have the light. When details in dark-green trees are impor-
tant, some prefer tungsten-balanced film with No. 85 filter, or with only
partial correction (e.g., Coral No. 1).
Creating dynamic action: Finding foreground elements to shoot through
(e.g., trees, rock formations) can increase shot “energy.” For air-to-air,
plan intersecting vectors of the aircraft. If air-to-ground, utilize the
horizon roll capability of a nose mount or the pirouette ability of a door
mount. Exterior gyro mounts can combine zooms with helicopter mo-
tion to create tremendous approaches and reveals (peek-a-boo from be-
hind hills/buildings), or can be undercranked for super-fast approaches.
Most gyro-based systems don’t tolerate panning as fast as fixed nose
mounts, helicopter turns should not be too extreme. For example, long
focal lengths with slower orbits around the subject can create a rap-
idly moving background behind a subject on a mountaintop. Because
helicopters can’t stop on a dime, some gyro-system approach shots are
performed in reverse, rapidly climbing shots.
AviAtion BAsiCs
FAA Regulations for Camera Mounts
Federal Aviation Administration rules require that when any object is
bolted or strapped anywhere to any airframe—Cessna or 747—the instal-
lation must be certified airworthy by a recognized FAA inspector. The in-
spector issues a temporary-use type 337 certificate specific to that aircraft.
If you use the same mounting bracket on another aircraft of identical type,
a new 337 must be obtained. Once granted, the 337 is added to a particular
aircraft’s manual, so future reinstallations don’t require FAA recertification.
Time must be allowed in the production schedule for this FAA field inspec-
tion, if required; the inspector is typically busy and backlogged.
Mounting hardware that has earned an STC (Standard Type Certificate)
means that the equipment vendor has persevered through a costly process.
STC installations do not require field inspection by an FAA examiner and
are certified flight-ready by a more readily available A&P (airframe and
power-plant mechanic).
Pilots
A truly qualified pilot is critical for both the safety and success of the pro-
duction; it is obviously essential for the pilot to have many hours of “time-in-
type” of similar aircraft. When filming in the United States, a pilot should be
218
|
operating under his or her own (or company’s) FAA Motion Picture Manual.
This allows a pilot some latitude within the FAA guidelines, which restrict
the activities of all other aircraft. A high level of expertise and judgment
must be demonstrated before the FAA grants such manuals. Of course, many
flying situations still require advance approval from the regional FAA office,
as well as other local authorities. Preproduction consultation with pilots is
strongly recommended.
Remotely operated gyrostabilized camera systems are often called upon
for very close work with long lenses; precise camera positioning is absolutely
critical. Few pilots have the skills necessary to hold a helicopter in a steady-
enough hover for a shot with a tight frame size. While the gyro systems isolate
the camera from helicopter vibration, an unstable hover causes XYZ spatial
excursions, resulting in constant relocations of the film-plane—an undesir-
able parallax shift. The footage may appear as if the camera is panning or
tilting; actually, the helicopter is wobbling about in the hover position.
A local pilot and helicopter may be the only option when on very remote
locations. Some regional helicopter companies may not allow other pilots
to fly their helicopter. These local pilots must understand that some filming
maneuvers might require an exceptional degree of mechanical reliability from
their aircraft; when hovering below 300 feet in calm air, a safe autorotate is
impossible. Spend the minimum amount of time hovering low, and don’t press
your luck with unnecessary takes. If you must work with a helicopter pilot who
has no film experience, try to choose one with many hours of long-line work
(i.e., heavy cargo suspended by cable, or water-bucket fire suppression), as this
type of flying requires both a high level of aviation skill and judgment.
Protocol, Planning and nonessential “Payload”

For those new to aerial filming, getting swept up in the excitement of
something new, combined with the typical rush of production, can distract
one from safety. Always think before moving around aircraft—more than
one crew member has died walking into spinning tail rotors. Never walk
toward or around the rear of a helicopter!
Air-to-air shots must be choreographed so that one pilot can maintain vi-
sual contact with the other aircraft; the pilot with “visual” can switch off dur-
ing filming, but never depend on a video monitor to determine safe spacing.
Ground locations that involve close helicopter work may require a wet-
down to eliminate flying dust that might damage jet engines. Helicopter
pilots should not be expected to land or fly low in unfamiliar terrain with no
prior knowledge of electricity or telephone wires.
When a camera mount is installed aboard a helicopter under an STC or
337, that aircraft is put in a restricted category. Nonessential personnel are
not allowed on board, according to FAA rules. Extra people cause the pilot to
| 219
have less lift and agility. Unexpected wind gusts may require extra torque (or
temperature demands of the turbine), which is not available when close to
maximum payload. An extra person also means that less fuel can be carried.
Communication: Radios and Pilot Jargon

Handheld ground-to-air aircraft frequency radios are available and are
most preferred. Tyler and Advanced Camera Systems (formerly Continental
Camera) offer intercom boxes for helicopters, which allow communication
to and from the helicopter using standard motion-picture walkie-talkie fre-
quencies. For one-way communication with on-camera talent, an induction
loop can be worn under wardrobe, which relays signals to an “earwig,” a tiny
wireless induction pick-up earphone (available from audio suppliers).
While flying in helicopters or fixed-wing aircraft, don’t distract the pilot
from his primary job. Theoretical conversations belong on the ground and
should be kept to a minimum when airborne. In metropolitan areas, the
pilot needs to listen for radio calls from air-traffic controllers and look for
“Sunday fliers” who are “off ” radar control.
Make things easy for your pilot; be specific and concise so he can an-
ticipate his flight path. Pilots require a clear description of where the lens
is pointing and what you’re seeing. During a preflight briefing, learn each
other’s language. Usually when filming aboard a fixed-wing aircraft bigger
than a small Cessna, the camera operator is positioned somewhere in the
back, unable to communicate visually with the pilot and often relying upon
headset communication. If the cinematographer cannot see out the front of
the camera aircraft, the pilot might have to steer the plane off-course by 90°
so the cinematographer can see what lies ahead.
Lining up a particular background behind a story aircraft can be tricky,
even for experienced pilots. Getting story aircraft to enter an empty frame
requires formation flying from very capable pilots. Military-trained pilots
with high formation time are adept at such positioning.
Fuel trucks
Helicopter agility suffers greatly as payload increases; having a fuel
truck on location allows fuel weight to be kept to a safe minimum. Not
all airports have jet fuel, which is required for the jet-turbine helicopters
listed below. Not all fuel trucks are permitted to leave airports and travel
on public roads.
Weather Basics
Aircraft performance suffers when air becomes less dense, which happens
when heat and/or altitude is increased. Helicopters are most stable when
moving forward faster than 20 mph air speed (translational lift). Therefore,
220
|
steady breezes can be beneficial, allowing the helicopter to hover relative

to the ground while actually flying forward in the airstream. Conversely, a
strong, steady wind can make it impossible for a helicopter to maintain a
hover in a crosswind or downwind situation. The types of helicopters used
for filmming are rarely intrumented to fly in heavy fog or clouds; they operate
under Visual Flight Rules (VFR).
Choice of sides for Mounting

All things being equal, it is best to mount the camera on the pilot’s side of
the aircraft, giving him or her the best visibility of the subject.
Max speed
When an aircraft has any kind of exterior modification for camera mounts
(i.e., door removed, nose/belly mounts or gyro mounts), a maximum allow-
able air speed known as VNE or “Velocity Not to Exceed” is determined
through FAA certification. Be sure this VNE speed will match your subject.
Air speed should not be confused with ground speed; prevailing wind con-
ditions may help or hinder filming logistics.
safety
See the related Industry-Wide Labor-Management Safety Bulletins at:
http://www.csatf.org/bulletintro.shtml.
GlossARy oF FliGht PhysiCs

Awareness of these concepts, most of which are related to helicopters, will
expedite communication and increase safety.
AGL: Above Ground Level altitude, expressed in feet or meters.
Flight Envelope: Published by the aircraft manufacturer, this term refers
to the conditions in which a given aircraft can be safely operated. This
takes into account air density (altitude, humidity and temperature)
and allowable torque and temperature output of the engine/transmis-
sion/turbine.
Ground Effect: Condition of improved lift when flying near the ground.
With helicopters, this is roughly within ½ rotor diameter, becoming
more pronounced as the ground is approached.
Losing Tail-Rotor Authority: When helicopters attempt to hover or crab
in a crosswind, especially when heavily loaded in thin air (close to the
edge of their flight envelope), it sometimes happens that not enough
power can be diverted to the tail rotor to maintain the sideways posi-
tion relative to the wind direction. When the rudder pedal is maxed
out, a wind gust can spin the helicopter on its mast like a weather
vane. This causes an abrupt loss of altitude. If shot choreography re-
| 221
quires such a position, have plenty of altitude above ground level for
recovery.
Rudder Pedals: Cause an aircraft to yaw; if using a nose mount, the same as
panning. Airplanes have a vertical rudder, though single-rotor helicop-
ters use a tail rotor to control yaw. Helicopters change the pitch of the
tail rotor blades to alter the force of the tail rotor, which counteracts the
torque from the main rotor.
Settling with Power: When helicopters hover out of ground effect, but still
near the ground in still air, prop wash can bounce off the ground and
wrap back around the top of the helicopter as a cyclonic down-draft.
A pilot can be fooled into applying more horsepower, which only in-
creases the down-draft intensity. All pilots know of this phenomenon,
but those who work infrequently in low-AGL situations may not be on
guard. Try to change shot choreography to continuously move out of
your own “bad” air.
Yoke/Control Stick: Makes the aircraft pitch and roll (stick forward = nose
down, stick right = roll right). When using nose-mounts, coordinate
with the pilot so aircraft pitch and camera tilts are harmonious.
CoMMon heliCoPteRs
Aerospatiale A-star/twin-star (As350 and As355 twin)
Pro: Very powerful engine(s) for superior passenger and fuel-range capacity.
Extremely smooth flying with three-blade system. Accepts door and ex-
terior gyro-stabilized mounts (gyro mounts require helicopter to have aft
“hard points”). Tyler Nose Mount can fit if the aircraft is equipped with
custom bracket by the helicopter owner. High-altitude ability is excellent.
Con: Costlier than Jet Ranger. Does not hold critical position well in
ground-effect hover.
Aerospatiale llama
Pro: Powerful engine mated to lightweight fuselage specifically designed
for great lifting ability and high-altitude work. Accepts door and some
gyro-stabilized mounts. Holds position quite well in ground-effect
hover. Set the helicopter height record in 1972 at over 40,000'.
Con: Expensive; very few available.
Bell Jet Ranger (206)

Pro: Agile and relatively economical. Accepts door and nose mounts. Most
variants will accept gyrostabilized mounts (must have external “hard
points”); excellent close-hovering stability.
Con: Earlier models have low-power engines. Poor high-altitude perfor-
mance; reduced range/payload for exterior gyro-mount work.
222
|
Bell long Ranger (206l)

Pro: Excellent for both door and nose mounts. Very smooth and powerful;
high-altitude performance is good.
Con: More expensive than Jet Ranger. Reduced inventory of external gyro-
mount brackets from various vendors.
Bell huey Uh-1 (civilian version known as 204/205)

Pro: Accepts door mounts. Very powerful; high-altitude performance is
good. Utilized more as a story ship than as a functional camera ship;
camera often onboard with shots involving talent.
Con: Very bouncy, expensive. Requires permanent modification of airframe
to fit external gyro mounts.
hughes 500 series (A through e + notAR models)

Pro: Smoothest-flying (five rotor blades) and most agile of all helicopters;
relatively economical. High-altitude ability OK.
Con: Extremely small rear compartment, impractical for door/side
mounts; better for handheld. Custom-built (337 installation rules)
bungee mount, “Outside Mount,” or Spacecam bracket requires a per-
manent modification to the airframe. Earlier models have low-power
engines.
CoMMon Fixed-WinG AiRCRAFt

But first another word about pilots:
The costs of operating a jet-turbine helicopter weeds out most “Sunday
flyers.” By comparison, many amateur fixed-wing pilots want to expand
their experience, and some professional fixed-wing charter pilots may
have low “time-in-type” of the aircraft most appropriate for your filming
needs. Practically all fixed-wing pilots will be eager for a film job, but most
underestimate how complicated piloting tasks can become while position-
ing for the camera. It’s your life—choose wisely.
Cessna (many variants)

Pro: High-wing good for air-to-ground situations. Economical and ubiqui-
tous. Some types may be flown with window/door ajar or removed with
337 waiver. Stability dramatically increased for short duration when
flown in side-slip.
Con: Limited space for rigging mounting devices.
twin (e.g. Piper navaho, Cessna 414)

Pro: Higher speed, greater range. Some types may be flown with rear win-
dow/door ajar or removed with 337.
| 223
Con: Limited field of view to avoid aircraft structure for both air-to-air or
and air-to-ground.
Customized Propeller aircraft

(e.g., B-24, B-25, Cessna 414, Pilatus Porter, Pitts)
Each offers unique capabilities: hard mounts on an aerobatic plane (Pitts);
wide range of low to high speeds with harness to lower operator (protected
by wind shroud) for wide field of view aft using door-mount type arm
(Porter); unobstructed view through gun-port fuselage openings at speeds
higher than the helicopter domain (B-24, 25). Nonpressurized aircraft be-
come quite cold at altitude, which may create condensation problems for
optics (see Arctic Cinematography page 239).
Customized Jet systems (learjet, l-39)

Pro: High-altitude and/or high-speed work. First choice for air-to-air of jet
aircraft.
Con: Expensive.
tyPes oF CAMeRA MoUnts

A handheld camera may be adequate for wide perspectives with minimal
pans/tilts for small cameras, but when more sophisticated shots are desired,
there are three basic types of mounts.
helicopter nose/Belly Mounts

Abilities: Shoot straight forward or straight back, with tilting ability. Video-
assist viewing. Very dynamic shots when used with a wide-angle lens,
low to the ground, in narrow canyons or city streets and/or when
the helicopter is making steeply banked turns. These mounts can tilt
straight down or rearward, a position used when ferrying to protect the
lens from bug strikes. Designed for Jet Rangers and Long Rangers, if
equipped with standard or high skids.
Limitations: These systems do not pan; the helicopter must be yawed (steer
left/right). No vibration isolation; smooth air and wider focal lengths
required. Takes 400' of film at a time; landing to reload required. Bug
strikes and moisture droplets cannot be cleaned in-flight, and if small
might not be seen through the video viewfinder while in flight. These
systems cannot be installed on helicopters with low skid gear. As the
frontal area of the camera package exposed to wind-pressure increases,
vibration increases; screw-on filters are recommended over matte boxes.
Lens servo motors and the like should be kept to an absolute minimum.
Vibration restricts use of focal lengths to no longer than 50mm in the
best of conditions.
224
|
helicopter door/side Mounts

(can be used from larger fixed-wing aircraft and camera cars)
Comprised of a platform attached to the aircraft floor with a vertical mast,
which supports the camera arm. The rear of the camera arm extends behind
the operator and forms a counterweight. Some arm systems may be attached
to a Mitchell tripod plate for camera car use. Some mounts isolate the opera-
tor from vibration atop a sprung seat.
Abilities: Shoot fast pirouettes at radical down angle, either air-to-ground or
air-to-air. Allows pilot to fly in a fairly normal manner, presenting the
side of the aircraft toward the action. Takes 400' of film at a time; cam-
era can be reloaded in the air. For optimal results, the camera should
be allowed to float on the isolation mast, guided with slight fingertip
pressure on hand grips, which contain servo focus/zoom controls. In
general, the greater the mass of the vibration-isolated section of the
mount, the better the image stability. Wind deflector fitted to the avail-
able aft doorhinge helps reduce wind pressure on front of lens. By gang-
ing multiple Kenyon gyros on the camera arm and counterweight, using
a 275mm lens is possible, if minimal panning and tilting is required.
Limitations: Fairly smooth air required. Slight overcranking may be re-
quired to steady the image. Shots for background plates may call too
much attention to image chatter. Pan/tilt range is usually limited by the
interior architecture of the aircraft cabin interfering with the swinging
counterweight.
exterior Gyrostabilized Mounts (a.k.a. Ball Mounts)

Offering ultimate image stability, a remotely operated camera housed
inside a spherical wind shroud is bolted to the outside of the airframe with
special brackets. All controls remain inside the helicopter; optical viewfinder
is replaced with videotap. Most mounts attach to most Jet Rangers, A-Star/
TwinStars and Llamas. These systems are also used on camera cars, cranes
and boats. Some systems allow removal of wind-shroud sphere for use of
wide-angle optics for ground applications.
Rate vs. Mass: The first gyro-stabilized aerial systems used heavy spinning
wheels to generate angular momentum (mass gyros). Miniature solid
state rate gyros and direct-drive motors, used in current military and
video systems, have spawned a new genereation of on-the-ground
stabilizing heads for film cameras (e.g., Flight Head, Libra, Stab-C,
WescamXR); some of which are superior for longer focal lengths. Com-
pared to mass-gyro systems offer superior operator response and allow
far more aggressive helicopter maneuvering; gear wheels are optional.
Abilities: Complete isolation of the camera from all vibration; independent
from pitch, roll and yaw movements of the vehicle. Frame rates below
| 225
4 fps yield perfectly stable images. Focal lengths in 35mm format up to

500mm can be used. As the helicopter yaws (steers left and right), the
mount acts like a compass, keeping the camera pointed in the direction
last panned. Shoot straight back, forward and down within one shot.
Regardless of how bumpy the air, these systems deliver perfectly smooth
images on days that would ground a door mount. Bugs and moisture
droplets can be wiped off in flight in some configurations. Ideal for shoot-
ing background plates. All current gyro systems use a fourth vertical
reference gyro to maintain a level horizon, regardless of helicopter bank
angle. Some systems allow the camera to Dutch angle during helicopter
turns if desired. All systems can be remotely operated at the end of very
long tethers (e.g., 500 feet), some by radio link. All current systems allow
camera speeds to be varied in-shot with automatic iris compensation.
Magazines are 1,000' and require a full load for proper weight and bal-
ance. All current systems offer BNCR or Panavision mount cameras with
choice of zooms or primes; 10:1 zoom is standard configuration.
Limitations: Mass-Gyro systems do not pan/tilt as fast as remote heads, and
exhibit minor lag and over-shoot. They utilize a similar two-axis joy-
stick control for pan/tilt, which takes time for operator skills to develop;
no gear-wheel input available for on-the-ground application. All gyro
systems require landing to reload and change filters; an experienced
technician is required. Critical balance of these systems restricts their
choice of lenses; lens changes between zooms and primes requires con-
siderable time.
AlteRnAte stABilizeRs in the AiR

Bungee or surgical tubing Mounts
For filming inside-to-outside of a story aircraft, this is often the best solu-
tion. The camera is hung by a web of rubber shock cords that are secured to
the airframe, so that the camera weight pulls the cords to the middle of their
stretch range, thereby allowing for maximum excursions of the camera in
any direction.
Limitations: Multiday custom design and fabrication; 337 inspection re-
quired.
steadicam
Using a short center post between camera and sled and attached to an
an appropriate vehicle mount (not worn), good results can be achieved as
long as the camera is kept sheltered from the wind. Best for onboard a story
aircraft, rather than as a substitute for a door mount.
Limitations: Wind pressure striking the camera causes gross instability. 337
inspections required.
226
|
Rigid/hard Mounts (fixed-wing or helicopter)

Sometimes rigidly mounting the camera to the inside or outside of the
aircraft achieves the best results. While the image of the background outside
the aircraft may vibrate, the image of the airframe and passengers remains
steady relative to the camera (think car-mount, but engineered for the FAA).
Limitations: Multiday custom design and fabrication; 337 inspection
required.
The Arriflex stabilizer (limited availability) is a small periscope box of
two front-surface mirrors, one gyrostabilized. The Arri unit has somewhat
limited pan/tilt ability due to gyro lag and is for small-diameter, longer focal-
length primes (75–180mm in 35mm format).
Dynalens stabilizers, built in 1962, are liquid-filled prisms that are the
predecessor to the “steadi-shot” found on modern home-video cameras. Two
glass discs are coupled together with a bellows; sandwiched between the discs
is a layer of high-index refractive fluid. Gyro motion sensors detect devia-
tions in pitch and yaw, sending signals to servo motors which counter-pivot
the appropriate “pan” or “tilt” disc. This counter-pivot creates a partial prism
shape. The Dynalens can correct rapid vibration excursions up to 6° in either
pitch (tilt) or yaw (pan); however, camera pan/tilt speeds must be reduced.
Only six units remain but have been refurbished; the maximum diameter
available is 6". Available from Steve Peterson in California: (714) 579-7857.
The Kenyon stabilizer is a small package containing two gyros to resist
motion in two axes (typically pitch and roll). It mounts to the bottom of the
device to be stabilized. Works best with ultralightweight handheld cameras
with a videotap finder/mini-monitor replacing the optical viewfinder. Note
that the gyro is useless if attachment to camera is not extremely rigid (bolted
on). (800) 253-4681; http://www.ken-lab.com
CAMeRA MoUnts And

dediCAted GyRo systeMs
Advanced Camera systems (formerly Continental Camera)
(Door and nose mounts; other ground based gear)
(818) 989-5222; http://advancedcamera.com
Door mounts for most common helicopters for video/16mm/35mm and
specialty large formats. GyroMaster, Master and Magnum mounts handle
cameras up to 46 lbs. GyroMaster uses up to four Kenyon gyros bolted to
Master arm and counterweight.
Two belly mounts are available for Bell 206 A/B and L helicopters with
standard or high skids. Megamount can be used on a crane arm as a three-
axis head, or on the nose of a helicopter as two axis (no pan; tilt 225° and roll
360°). Megamount can handle large cameras up to 80 lbs and has the unique
ability among nose mounts to look forward and rearward without having
| 227
to land to change camera mounting. Megamount can also record 40-second

moves for auto replay, and can be controlled by wheels or joystick. The Belly
Mount accommodates 35mm cameras, no pan, 100° tilt. Can be mounted
with camera looking fore or aft. Also Body-Cam stabilizer and various re-
mote heads.
Gyron systems (dedicated Rate-Gyro system for Film and video)

(626) 584-8722; www.gyron.com
Fiber optic rate-gyro coupled to active bearings allow for very rapid and
precise pan, tilt and roll input during the most extreme helicopter maneu-
vering. Fast stabilized horizon roll possible (45° for film/90° for video); gear
wheels available. Video system can be mounted on customized twin-eingine
Cessna 337. Waterproofed components inside ball allow for windowless
operation in most conditions; flies without shroud for extreme wide-angle
film lenses.
Camera: Standard Arri 435, 1000' mags, lenses up to Panavision 11:1
Primo. Video systems use 36:1 lens (non-doubled).
Gyrosphere (dedicated Mass-Gyro system)

(818) 787-9733; www.gyrosphere.com
Two Gyrosphere systems were built in the mid-1980s using earlier Wes-
cam parts; the extensive upgrade and redesign work represented many
“firsts.” Vertical-reference gyros were added to automate level horizon;
the speed-aperture computer was intergrated; improved stabilization and
camera steering enabled faster and more accurate pans/tilts with less lag;
improved ergonomics were made possible with hanheld joysticks; prime
lens capability. Mixed analog and digital electronics. The vertical slit curved
plexiglass window negates the use of a polarizing filter.
Camera: Modified Mitchell Mk-2 (3–36 fps) with underslung XR-35
magazine. Also available with Empireflex VistaVision camera from ILM
(2–48 fps) or Vistacam from BCS (2–48 fps).
Pictorvision (Gyro-stabilized digital & Film Aerial and

Ground systems for Cinematographers)
(818) 785-9282; pictorvision.com
The original gyro-stabilized aerial systems were begun in the 1960s as
Wescam systems—and Pictorvision is the original rental group from that
Wescam era. Now Pictorvision offers all options of aerial systems—from the
original Wescams, Cineflexes, the XR-Series for ground applications, and
the most current and advanced technology—the Eclipse system. Revolu-
tionizing a new generation of aerial camera systems, the Eclipse offers the
most advanced steering, stability, and pointing capabilities available on the
228
|
market. By partnering with some of the world’s most creative and experi-
enced DPs, Pictorvision’s engineering team integrated user functionality
with industry defining technology to produce the Eclipse—ushering in a
NEW dawn of aerial cinematography.
Cameras: The open-architecture of the Eclipse & XR-Series allows for an
unlimited choice in digital or film camera and lens combinations, includ-
ing Arri 435, Alexa-Series, F65, Genesis, IMAX MSM, iWerks Compact,
Phantom 65, Red One, RED Epic. The Wescam can support many of the
current digital camera systems, or our own Mitchell Mk-2 (1–60 fps) all with
a variety of cinema lens choices.
spacecam (dedicated Mass-Gyro system)

(818) 889-6060; www.spacecam.com
Unique gyro system using heavier gyro wheels spinning at greater rpms.
Patented powered cardin joint allows more responsive and faster pans/tilts.
Digital electronics allow many abilities (e.g., Dutching in sync with heli-
copter turns). The lens looks through a windowless port. Action-arm nose
bracket for modified Hughes 500 helicopter allows steep banking; nose and
tail position brackets for some other helicopters.
Camera: Customlightweight Mitchell Mk 2 (1–120 fps); VistaVision
(1–48 fps); 65mm 5-perf
| 229
Underwater Cinematography
by Pete Romano, ASC
U nderwater photography presents the cinematographer with a wide va-

riety of situations not encountered in above water photography. Water
is a dense medium and has certain light-absorption properties that cannot
be changed. When working in open water, weather, current and visibility can
change daily or sometimes even hourly and you have to be prepared for just
about anything. Tank and pool filming present you with water temperature is-
sues for talent comfort and visibility issues that might not work with the scenes
to be shot. Locations might vary from a tropical reef to a frozen ice pack, from
a backyard pool to a cement tank 200' in diameter and 60' deep, and water vis-
ibility could vary from 100' to less than a few feet. It is crucial to understand
how to work with and around these factors. If you are an experienced diver
and really understand the mechanics and optics of your underwater camera,
you will be able to deal with most underwater filming assignments.
Buoyancy control is a very important consideration when shooting un-
derwater. At certain times you will want to be neutrally buoyant so you can
swim the camera like a Steadicam® and crane combined, or you might be
stationed on a set of parallels or a ladder in a tank, without fins and be heav-
ily weighted for stability and controlling the camera.
As the underwater specialist, your skills have to go beyond just dropping
in the water and getting the shots. A lot of your assignments will have you
as part of a first unit where your knowledge and experience will be called
upon to explain the in’s and out’s of working underwater. Shots might be
presented to you that might not be possible underwater or might be a safety
concern. Your diplomatic skills will be called upon to explain the problems
but you should have shot options to present to still be able to tell the story.
Do not forget that you are there as a tool to record the director’s vision while
working closely with the film’s director of photography.
When working with talent, who for the most part are in wardrobe without
mask and fins and also holding their breath, your job is to get the shot as
safely and quickly as possible. Your ability to move in to help make the shot
happen instead of waiting for it to happen, could be the difference between
getting or not getting the shot,
HD UnDerwater
HD technology is constantly evolving. Although you can instantly see
what you’ve recorded, at this time, HD is not as forgiving as film. The need
230
|
to more accurately determine your underwater exposure will be essential.

Relying on determining a working ASA/ISO setting for the HD camera
you are using will not take into account the differences in color space and
contrast between the various cameras available, and how those cameras will
react to the lighting conditions underwater. You will most likely need a good
HD engineer as part of your support crew. There are two camps on the best
way to color correct for underwater HD. One camp white balances under-
water, sometimes after every depth change. The problem with this scenario
is when you tilt up your image goes orange because you have compensated
for the lack of warm tones in the water. The other camp goes with the preset
auto white balance and corrects the color in post. My advice is to keep it as
simple for the operator/DP as possible. Most HD auto settings will deliver an
adequately balanced image for color and exposure, and will save the under-
water operator the hassle of trying to determine the proper settings while in
an already abstracted environment.
Most HD cameras are sensitive to shifts in temperature and humidity. This
can lead to problems with tape transport or back focus shifts. The sudden
temperature change that can occur from the surface of the water to a few feet
below the surface may in some instances be enough to impair a camera’s nor-
mal function. Take adequate precautions to ensure that the camera is handled
in a dry environment during tape or disc changes, and if filming in an area of
cold depths, allow sufficient time while surfacing for the camera to gradually
approach surface temperature in order to avoid moisture condensation.
How LigHt is CHangeD by water

The intensity and quality of light available to the underwater cinematog-
rapher depends on many factors. Light rays from the sun, sky or motion
picture lamps that strike the sea’s surface or the surface of a pool or tank
produce a number of phenomena that should be understood, so the unde-
sired effects can be avoided or compensated for.
Reflection: Light is reflected back into the air at an angle equal to the angle
of incidence. Light penetration is dependent upon the angle and in-
tensity of the light source and water’s surface conditions. Heavy seas
and wave action will reflect much more light than a calm sea or pool
and the waves will also affect the light intensity. Sunlight penetration
in open water situations is best in the latitudes of the United States be-
tween 9 a.m. and 4 p.m. in the summer and 10 a.m. and 2 p.m. in the
winter. This does not mean you cannot shoot film underwater before or
after these times. It just means sunlight penetration is best during these
times. With the latitudes of today’s high speed films and HD’s ability
to capture images in low light, underwater cinematography can be ac-
complished successfully in low light situations.
| 231
Refraction: Light travels in a straight line, but light that penetrates the water
undergoes a change in direction—it is refracted. For example, a pencil
placed in a half-full glass of water appears to be bent at the place where
it enters the water. Refraction is the source of most of the problems that
are encountered in underwater photography.
Color Absorption: As light travels through water, the water acts as a filter
reducing the spectrum of light that penetrates it. The long wave lengths
of light—reds, oranges and other warm colors—are the first to be
absorbed and the visible light is dominated by the blue-green (cyan)
effects just a few feet below the surface. All water acts as a continuous
filter and the water depth added to the distance of the subject to the
camera equals the total water path that the light has to travel.
Scattering: Caused by suspended minerals, plankton, dirt, etc. The absorp-
tion and scattering of light rays reduces the intensity of light and the
saturation of colors.
eqUipment
The cameraperson and assistant should be completely familiar with the
underwater camera equipment and have the ability to perform simple field
repairs. In general, the camera housing should have provisions to adjust its
weight for salt or fresh water and be slightly negative in buoyancy when sub-
merged. Features to look for are: simplicity in design; easy-to-locate, bright
reflex viewing; the ability to accept a variety of lenses; interchangeable flat
and dome ports; external focus, iris and camera run controls; video assist
to the surface; an underwater monitor; remote camera run capabilities; and
the ability to mount the housing to a tripod. Servicing the housing for lens
changes, fresh batteries and film reloads should be quick and easy so as not
to eat up production time, and a spare parts kit for repairs is mandatory.
Even though a water alarm might be incorporated into the housing’s de-
sign, the assistant should have the ability to check the watertight integrity
of the equipment before it goes into the water. Depending on the design of
the underwater housing, you can check the sealing of the housing by either
pulling a vacuum or adding slight positive pressure (approximately 2 to 3
pounds). Check with the underwater housing manufacturer as to whether a
vacuum or positive pressure should be used.
Care of eqUipment
At the finish of every shooting day, always rinse off the camera housing,
lights and accessories with fresh water, especially when shooting in salt
water. A daily clean and check of all exposed O-rings should be performed,
as well as checking the external controls for camera run, iris and focus for
ease of operation. In full sun, always cover the housing with a solar blan-
232
|
ket or at least a white towel to avoid overheating the camera, batteries and
especially the film. Do not overgrease any O-ring. The purpose of silicone
on an O-ring is to keep the rubber supple, allowing it to lay comfortably in
the groove. Too much silicone grease does nothing for the seal and actually
attracts grit and debris, which can cause a leak.
In case of salt water exposure to any camera system, immediately dis-
connect battery power. Break down the entire camera system and rinse
with fresh water. After the fresh water rinse, pour alcohol over the parts to
displace the fresh water and allow the unit to dry. If there was fresh water
exposure, go directly to the alcohol step and allow the unit to dry completely.
These emergency steps might not get your camera system back on line, but
will at least prevent much of the damage that would be caused to the camera
if it were sent in for repairs while wet.
As a camera assistant caring for an underwater housing, one of the most
important things to remember is that it’s your responsibility to maintain the
watertight integrity of the housing. No matter where the equipment comes
from, no matter how recently it has been serviced or how thoroughly it was
prepped, there are countless accidents, mistakes and oversights that can
cause a potential leak. Every time a housing is opened for a magazine or
lens change, careful attention must be paid while resealing the housing. No
housing is 100 percent leakproof if safeguards are not applied.
In addition to the camera assistant’s normal complement of tools, an
underwater service kit will cover most of the problems that might arise. This
kit should include a bottle of alcohol, a tube of silicone lube and spray, sili-
cone adhesive, dental picks, Q-tips, a Scotch Brite pad, electrical tape, WD-
40, soldering iron and solder, cotton towels, a multi-meter, spare O-rings,
self-adhesive Velcro and an Allen wrench set.
fLat or Dome Lens ports

One of the most important and often misunderstood features of an under-
water camera housing is the lens port. Flat ports were all that were available for
underwater photography from its beginning in 1893 until 1931, when hemi-
spherical dome ports were first used to correct for the refractive properties of
water. Flat and dome ports both have their place in underwater photography
and it is important to know the theory and practice of each. For underwater
photography with wide lenses, the dome port is the best choice, but if the shot
starts or ends above water, or if you need to shoot a close-up with a long lens,
the flat port should be used. Glass is the best material for underwater ports. Be
careful of plastic ports. They scratch easily and will cause flares on the image.
flat port
The flat port is unable to correct for the distortion produced by the dif-
| 233
ferences between the indexes of light refraction in air and water. Using a flat
port introduces a number of aberrations when used underwater. They are:
Refraction: This is the bending of light waves as they pass through different
mediums of optical density (the air inside the camera housing and the
water outside the lens port). Light is refracted 25 percent, causing the lens
to undergo the same magnification you would see through a facemask.
The focal length of your lens also increases by approximately 25 percent.
Radial Distortion: Because flat ports do not distort light rays equally, they
have a progressive radial distortion that becomes more obvious as wider
lenses are used. The effect is a progressive blur, that increases with large
apertures on wide lenses. Light rays passing through the center of the
port are not affected because their direction of travel is at right angles to
the water-air interface of the port.
Chromatic Aberration: White light, when refracted, is separated into the
color spectrum. The component colors of white light do not travel at
the same speed, and light rays passing from water to glass to air will
be unequally bent. When light separates into its component colors, the
different colors slightly overlap, causing a loss of sharpness and color
saturation, which is more noticeable with wider lenses.
Dome port
The dome port is a concentric lens that acts as an additional optical ele-
ment to the camera lens. The dome port significantly reduces the problems
of refraction, radial distortion and axial and chromatic aberrations when the
curvature of the dome’s inside radius center is placed as close as possible to
the nodal point of the lens. When a dome port is used, all the rays of light
pass through unrefracted, which allows the “in-air” lens to retain its angle
of view. Optically a “virtual image” is created inches in front of the lens.
To photograph a subject underwater with a dome port you must focus the
lens on the “virtual image”, not the subject itself. The dome port makes the
footage marks on the lens totally inaccurate for underwater focus. Therefore
lenses should be calibrated underwater. The dome port offers no special
optics above water and functions as a clear window.
UnDerwater Lens CaLibration

To guarantee a sharp image when using either the flat or dome ports, it is
best to calibrate the lenses underwater by placing a focus chart in a swimming
pool or tank. Even on location, the hotel pool will offer a lot more control
than an ocean or lake. Set the camera housing on a tripod and hang a focus
chart on a C-stand. If possible, calibrate your lenses at night or in an indoor
pool or tank. Crosslight the focus chart with two 1200 watt HMI’S, 2' in front
at a 45-degree angle to the chart. Starting at 2', tape measure the distance
234
|
from the underwater housing’s film plane to the focus chart. Eye focus the
lens and mark the housing’s white focus knob data ring with a pencil. Slide
the camera back to continue the same process at 3', 4', 5', 6', 8', 10', 12', and
14'. This should be done for all lenses. Once a lens has been calibrated, you
must establish reference marks between the lens and data ring so that you
can accurately sync up for underwater focus when lenses are changed during
the shoot. After marking the data rings underwater with pencil, go over the
calibration marks with a fine point permanent marker on the surface.
Dome port
To calibrate a lens with a dome port, use this basic formula to determine
the starting point for underwater focus: Simply multiply the inside radius of
the dome by four. That number will be the approximate distance in inches
from the film plane that the lens should be set on as a starting point for
underwater eye focus calibration. The most commonly used dome radius is
4". Multiply the 4" dome radius times 4. That gives you a measurement of 16"
at which to set your lens in the housing to begin calibration for underwater
photography. If a lens cannot focus close enough to take advantage of the
dome port, use a plus diopter to shift the lens focus back. Ultimately, your
lens should be able to focus down to at least 10" to be able to follow focus
from 1' to underwater infinity. When using most anamorphic lenses with
the dome port, you will have to add a +3 diopter to the lens to shift close
focus back in order to focus on the aerial image.
flat port
The refractive effect of the 25 percent magnification produces an apparent
shift of the subject towards the camera, to ¾ its true distance. As a general
rule, for flat port underwater photography, if you measured your camera to
subject distance at 4', you would set your lens at ¾ the distance, or around 3'.
For critical focus, especially on longer lenses and when shooting in low light,
underwater eye focus calibration is recommended. Shooting through a win-
dow or port of a tank or pool is the same as using a flat port. If you do shoot
through a tank window and want to minimize distortion, the camera lens axis
must be kept at 90° to the window’s surface. Camera moves will be limited to
dollying and booming to keep the lens perpendicular to the window’s surface.
Panning and tilting should not be done unless a distortive effect is desired.
Lens seLeCtion
Wider lenses are usually the choice in underwater photography because
they eliminate much of the water column between the camera and subject
to produce the clearest, sharpest image. This is especially important in low
visibility situations. Longer lenses, which are being used more and more
| 235
underwater, usually will not focus close enough to take advantage of the
dome port’s “virtual image” (you need to focus down to 12") and a diopter
will have to be added to the lens to shift the focus back, or you can switch to
a flat port and allow for the 25 percent magnification. Most Zeiss and Cooke
lenses between 10mm and 35mm can close focus on the dome ports “virtual
image” without using diopters. When using Panavision spherical lenses, the
close focus series of lenses allow the use of the dome port without using
diopters. The commonly used anamorphic lenses for dome port cinematog-
raphy are the Panavision “C” Series 30mm, 35mm and 40mm focal lengths.
Because these anamorphic lenses only close focus to approximately 30", you
will have to use a +3 diopter to shift minimum focus back to 12" to 14" to
focus on the “virtual image” created by the dome port. When using longer
lenses for spherical or anamorphic close-ups with the flat port, I set the lens
at minimum focus and move the camera back and forth on the subject or a
slate to find critical focus instead of racking the focus knob.
fiLters
Filtering for film underwater, as in above water applications, is used to
lower light levels, correct color balance and improve image contrast. Aside
from the standard 85 filter used to correct tungsten film for daylight ex-
posure, the use of color correction filters can be a very subjective issue.
Water’s natural ability to remove reds, oranges and other warm colors can be
overcome by using underwater color correction filters. These filters alter the
spectrum of light reaching the camera to reproduce accurate skin tones and
the true colors of the sea. Because all water is a continuous filter, the deeper
you go beneath the surface the more colors are filtered out. Also, the distance
between the camera and subject must be added to the depth of the water
to determine the correct underwater filter distance. UR/PRO™ and Tiffen
AQUACOLOR® filters both provide color correction for underwater filming,
and both manufacturers offer detailed information on the use of their filters.
Polarizing filters can improve contrast where backscatter from artificial light
is a problem but sunlight illumination underwater is not sufficiently polar-
ized and makes polarizing filters ineffective.
exposUre meters
Both incident and reflected light meters can be used for reading your ex-
posure underwater. In ambient light situations, the reflected meter is usually
used because it measures the light reflected from the subject through the
water column.
The most commonly used underwater light meters are the Sekonic L-164
Marine Meter and the Ikelite digital exposure meter. The Sekonic L-164, de-
signed in 1969 specifically for underwater photography, was discontinued in
236
|
1993 but then reintroduced in 1998 because of demand. It is self-contained,

needs no special housing, and is pretty straightforward in operation. The
Sekonic L-164 has a reflected 30-degree acceptance angle and gives very reli-
able underwater exposures. The Ikelite digital is also a self-contained under-
water meter and can measure incident or reflected light. It has a 10-degree
acceptance angle in the reflected light mode. Both the Sekonic L-164 and the
Ikelite digital meters can be used above water or down to 200' underwater.
There are also commercially available underwater cases for the Spectra IV,
Minolta IV, Minolta 1° Spot and the Minolta Color Temperature.
UnDerwater CommUniCations
Communication between the director/AD and cameraperson and crew
is critical. We’ve certainly come a long way since the days of banging two
pieces of pipe together, so depending on your budget and shot list, there are
different degrees of technology available.
A simple but effective means of communications between the camera-
person and the above water director/AD is a one-way underwater P.A.
speaker system. Used in conjunction with video assist from the underwater
camera, cues for roll, cut and talent performance can be instigated by the
director on the surface. The cameraperson can also answer questions via the
video assist with hand signals or a slate presented to the camera lens or by
nodding the underwater camera for “yes” or “no”.
For more complex set-ups where the cinematographer and the director
need to talk directly to each other, or when the director works underwater
with the crew, two-way communication is very efficient. While wearing full-
face masks equipped with wireless communication systems, the underwater
cinematographer can converse with the director and also talk to the under-
water and surface crews via underwater and above water speakers.
LigHting
Specifics on lighting underwater, as in above water, will vary from cin-
ematographer to cinematographer, and the following information is merely
a guide. Experimenting and developing your own technique should be your
goal. Today’s underwater cinematographer has a wide variety of HMI, in-
candescent, fluorescent and ultraviolet lights and accessories from which to
choose. Underwater lights designed for motion picture work should have
the ability to control the light output by means of diffusion glass, filters,
scrims, snoots and barndoors. Always check the condition of lampheads,
connectors, plugs and cables before use. Become familiar with the lamp’s
lighting system and be able to perform simple repairs and maintenance.
Unlike air, even in the best water, the clarity is significantly reduced in dis-
tances as little as 10' to 20'. Artificial lighting is often needed to adjust light
| 237
levels for exposure and vision, as well as to modify the color balance and
image contrast of the subject. By incorporating supplemental lighting, the
light’s water path, from lamp head to subject to camera, can be kept relatively
short, and the selective color absorption properties of water will be much
less apparent than if the light had to originate at the water’s surface.
HMIs light has a higher color temperature, approximately 5600 K (longer
light wavelengths), and thus penetrates further, providing more illumination
over a wider area.
The reflection of artificial light off suspended particles in the water is
known as “back scattering”. The effect is much like driving your car with
your headlights on in heavy fog or a snowstorm. In addition to back scatter-
ing there is also side scattering and in some instances even front scattering.
Light scattering can be greatly reduced if you separate the light source from
the camera and by using multiple moderately intense lamps rather than
using a single high intensity lamp. It is advisable to keep the light at a 45-de-
gree to 90-degree angle to the lens axis of the camera. Generating a sharp
light beam to help control the light and further reduce backscatter is done
with the aid of reflectors, barndoors or snoots.
In addition to the basic effect of light intensity reduction in water due to
absorption, the matter is further complicated by the fact that absorption is a
function of color. Red light is absorbed approximately five times faster than
blue-green light in water. This is why long distance underwater photographs
are simply a blue tint without much color.
Fill lighting underwater on a moving subject is best accomplished by
having the lamp handheld by the underwater gaffer. With the lamp being
handheld, the underwater gaffer can maintain a constant distance between
the lamp and the moving subject, keeping the exposure ratio constant. In a
more controlled situation where the subject is confined to an area of a set,
light can be bounced off a white griffolyn stretched on a frame or through a
silk on a frame.
bLUesCreen anD greensCreen

There are different schools of thought on which color screen is best to use
but both bluescreen and greenscreen have been used successfully underwater
for many years. Backlighting a bluescreen underwater has been done suc-
cessfully, but when the screen material is exposed to chlorine, it sometimes
picks up density and requires more and more light to maintain its exposure
reading. Front lighting fabric screens underwater has been very successful,
and depending on the size of the screen, the logistics of the location and the
budget, there are a variety of underwater lights to choose from. Underwater
lamps also can be augmented with above water lamps. It is best to mount
the screen on a movable frame, so it can be pulled out of the water and
238
|
hosed off after every day of shooting. Chlorine bleaches the screen quickly,
and rotating the screen 180° every day helps minimizes the color difference.
From the bottom to the top, the frame should tilt back approximately 10°.
Extending the screen above the surface of the water, even if only a few inches
is very important. By extending the screen above the water, it allows the
under surface of the water to act as a mirror and reflect the screens color on
the surface. This extends the screens coverage and allows the cameraperson
to tilt up to follow the shot while using the waters surface as an extension of
the screen. It is also important to stop or at least reduce the agitation of the
water to prevent a visible ripple effect on the screen.
LigHting safety
From a safety standpoint, when using AC power in or near water or other
potentially wet locations, it is essential (and in most cases, mandatory) to use
a Class “A” UL approved GFCI (Ground Fault Circuit Interrupter) for actor
and film crew protection.
The purpose of a GFCI is to interrupt current flow to a circuit or load
when excessive fault current (or shock current) flows to ground or along
any other unintentional current return path. The standard level of current
needed to trip a “people” protection Class “A” GFCI is 5mA.
Class “A” GFCIs are designed for “people” protection. Other GFCIs are
designed for various levels of “equipment” and “distribution” protection. In
general, if you can’t readily see the Class “A” marking on a GFCI, the device
is probably not designed for “people” protection. To make sure that the
GFCI being used is a Class “A” device, Section 38 of UL 943 (the standard for
GFCIs) requires that the “Class” of the GFCI is clearly marked on the device
in letters that are clearly visible to the user.
Today, Class “A” GFCIs are readily available for loads up to 100 Amps,
single- and three-phase, for both 120v and 208/240v fixed voltage applications.
Certain special GFCIs can also operate on variable voltage power supplies
(behind dimmers). If the device’s label does not clearly state the working
voltage range of the unit, check with the device’s manufacturer before using
the unit on dimmers (or other variable voltage applications), since conven-
tional GFCIs may not operate correctly below their rated voltage.
Specialty GFCI manufacturers produce advanced GFCI devices that offer
other important safety features, such as monitoring predictive maintenance,
power supply phase and ground correctness.
Choose your GFCIs carefully, because if misapplied, these important
safety devices may unknowingly fail to function and render your installation
with a false sense of security.
| 239
Arctic and Tropical
Cinematography
M ost of the difficulties encountered when using motion picture equip-

ment in the Arctic are caused by extreme cold and very low relative
humidity. Average temperatures may vary from 45°F (7°C) to -45°F (-43°C),
and temperatures as low as -80°F (-62°C) have been recorded. (Such low
temperatures may also be encountered at very high altitudes.)
The lubricating oils usually used in photographic equipment in more tem-
perate climates will congeal in an arctic environment so that moving parts
of cameras or other equipment will not operate. Leather and rubber also be-
come brittle at these temperatures. With motion picture films, loss of mois-
ture from the film emulsion when the original packing material is opened
may result in film emulsion shrinkage and brittleness, and subsequent film
curl in the camera gate. Such difficulties are not minimized by using films
with a polyester base unless these films (or those with a triacetate base) have
a gelatin coating on the support to compensate for emulsion shrinkage. It is
the effect of the very low relative humidity (less than 5%) and its emulsion-
drying characteristics that produces film curl. (Small heaters are sometimes
used in cameras to prevent film brittleness when working under conditions
of extreme cold, but under certain conditions this practice could actually
increase the chance of emulsion shrinkage by further reducing the relative
humidity in the film chamber.)
The film speed is also lowered by extreme cold and may be about one
lens opening slower at -50°F (-46°C) to -70°F (-57°C) than at 60°F (16°C).
Film becomes progressively more brittle as the temperature drops below
0°F (-18°C), but there is no marked change at any one temperature. Even at
subzero temperatures, film emulsion that retains its proper moisture con-
tent in the original package (equivalent to equilibrium at 40 to 60% relative
humidity) is more flexible than film that has been allowed to become too
dry. Film can also be bent with the emulsion side in with less chance of
breaking than if bent with the emulsion side out. Whether the film emul-
sion cracks or the film support breaks at very low temperatures depends on
1) how soon the film is exposed after removal from the original package;
2) the care taken in handling the film; and, 3) on the type and condition of
the camera in which it is used.
Temperatures generally encountered in the arctic will not cause polyester-
base films to break.
240
|
Arctic and Tropical Cinematography
PrePArATion of equiPmenT
While the difficulties of photography under arctic conditions can be se-
vere, they are by no means insurmountable. Careful advance preparation
will pay rich dividends in the form of easier and more reliable equipment
operation and better pictorial results. The first step in preparing for filming
in the arctic, high mountain regions, or in unheated aircraft at high altitudes
is to select the most suitable equipment with due regard for the work to be
done and the results desired.
Each kind of camera has its adherents, and no one type seems to be out-
standingly superior to the others. However, considering the working condi-
tions, good judgment dictates that the camera or cameras selected should
be compact, lightweight, easy to use, dependable, adaptable, and portable.
In choosing a 16mm motion picture camera, many arctic explorers prefer
the ease and convenience of magazine loading, threading roll film can be
very difficult under conditions of extreme cold. Certain camera models
are advantageous for low-temperature use because large-radius bends in
the film path and low film accelerations help prevent broken film. For best
protection of the film emulsion at extremely low temperatures, film-travel
rollers should have a diameter no smaller than ½". (13mm). Electric power,
if available from a reliable source such as a generator or vehicular power
system, is more dependable than spring-driven or battery power. However,
under field conditions, a spring-driven motor may prove more reliable than
an electric motor drive that depends on portable or storage batteries, which
can fail when subjected to extremely low temperatures.
Cameras should be winterized for satisfactory service under frigid
conditions. Some camera manufacturers provide a winterizing service
for cameras that are to be used at low temperatures over a long period of
time. Winterizing is a highly specialized operation, best entrusted to the
manufacturer or a competent independent camera service representative.
Essentially, the procedure calls for dismantling the camera and removing
the original lubricants. The shutter, lens diaphragm, film transport mecha-
nism, and other moving parts are then relubricated with materials that will
not thicken when the camera is exposed to extreme cold. In some cases,
powdered graphite is still used for this purpose. However, so-called “broad-
range” lubricants (such as Teflon and silicone) are becoming increasingly
popular, not only because of their effectiveness at low temperatures, but
also because they can be left in the camera permanently. In fact, such
lubricants are being used in manufacture. Hence, a camera that has been
lubricated with a broad-range lubricant, either in manufacture or as part of
a winterizing operation, need not be dewinterized and relubricated when
it is returned to use under normal conditions. When cameras are stripped
down for winterizing, weakened or damaged parts may be discovered and
| 241
should be replaced to avoid possible failure under the extra stress of severe
arctic temperatures.
It is also sometimes necessary to machine parts to allow greater clearance
between components. This is because aluminum and certain alloys have
greater coefficients of thermal contraction and expansion than steel. Since
small levers and knobs on cameras are difficult to operate when the photog-
rapher is wearing thick gloves, extensions can sometimes be added to levers,
and small knobs can be replaced with larger ones.
It may be helpful to run even recently winterized motion picture cameras
for a period of three or four hours to break them in thoroughly. A piece of
film three or four feet long can be spliced end to end (to form a continuous
loop), threaded into the camera, and allowed to run during the breaking-in.
In cameras intended for use with film magazines, the loop should be formed
in a dummy magazine. After the breaking-in period, the camera should be
checked for speed and general behavior. It should be noted that, although
magazine-type motion picture cameras can be winterized, the magazines
themselves are not winterized and may jam under conditions of extreme
cold. If film magazines are used, each day’s working reserve carried into the
field should be kept as warm as possible under the cinematographer’s parka.
Another possibility is to carry the film supply in an insulated thermal bag,
along with one or two small hand warmers.
Before your location shoot, a test run should be made in a refrigerator or
freezer capable of reaching temperatures as low as -30°F (-34°C) or -40°F
(-40°C). Even “winterized” cameras can fail in use because some detail was
overlooked in preparation, so this final test run is quite important. The film
and camera should be cooled for at least 24 hours prior to the test. This long
period of precooling is often overlooked, and the test becomes invalid.
Motion picture cameras should be given as much protection from icy winds
as possible during use. When battery-driven motors are used on cameras, the
motors and batteries should be kept as warm as possible. A flat black finish
on the cameras has some advantage in the arctic because it absorbs heat when
the sun is shining. Covers made from black felt material or fur and fitted with
eyelets or other suitable fasteners protect the camera from frigid winds and
help to retain its initial warmth for a time. Snaps and slide fasteners are not
recommended for use in subzero temperatures. Small magazine-type motion
picture cameras can be hung inside the coat to obtain some warmth from the
body; you may even need to wrap a chemical heating pad around the camera.
Inspect the camera’s lens each time it is removed from the clothing to take a
picture. The amount of “body static” generated under cold, dry conditions
can cause the lens to attract lint from the clothing.
Tripods should also be conditioned properly for use in the arctic. When
lubrication is required, there are oils available for use at temperatures down
242
|
to -70°F. Tripod heads for motion picture equipment should be winterized

if they include gyros, motors, or other revolving parts. As noted previously,
extreme cold causes leather and rubber to become brittle. A wax leather
dressing of good quality should be rubbed into leather carrying cases and
leather-covered cameras to prevent the absorption of moisture. Rubber
should be eliminated wherever possible.
Silk or lightweight cotton gloves under heavy woolen mittens are recom-
mended. Gloves or mittens made from unborn lambskin are excellent for
arctic weather. Silk gloves will keep the hands warmer and will afford con-
siderable protection when the outside mittens are removed for loading the
camera, adjusting the lens, etc.
equiPmenT And filming TeChnique

In the arctic or on mountain-climbing expeditions, as the altitude and
the subsequent cold increase, breathing becomes difficult, and working
normally involves a great effort. Reactions are slow. Therefore, everything
pertaining to the use of the camera should be made as simple as possible.
Exposure estimates may be poor when the faculties are dulled, so exposure
and other data should be printed on a card and fastened to the camera or its
cover in plain view.
Certain general cold-weather recommendations are in order for any cam-
era, still or motion picture. Breathing on a lens or any other part of the cam-
era to remove snow or other material will cause condensation that freezes
instantly and is very difficult to remove.
An important factor to keep in mind is the ever-present danger of frostbite,
a particular threat when hands or face come in direct contact with the metal
of the camera body. Cameras that are used at eye level and must be brought
close to the face for proper viewing and focusing should have their exposed
metal areas covered with heavy electrical tape, plastic foam, or some other
insulating material. Under no circumstances should the photographer touch
the camera or other metal equipment with ungloved hands, because the skin
will freeze to the cold metal almost instantly. A painful loss of skin almost
always results.
A thoroughly chilled camera cannot be used in a warm room until its tem-
perature equals the surrounding warmer temperature. Conversely, a warm
camera cannot be taken out into a blizzard because the blowing and drifting
snow will melt upon striking the warm camera, and soon the instrument
will be covered with ice. Loading film, even during a driving snowstorm, can
be accomplished with the use of a large, dark plastic bag, big enough to fit
over the head and shoulders.
A deep lens hood is very desirable for filming in the snow. It will help keep
the lens dry even during a fairly severe storm.
| 243
film
Great care must be used in handling film in subzero weather. The edges of
cold, brittle film are extremely sharp, and unless caution is exercised, they
can cut the fingers severely.
It is important that film be loaded and exposed promptly after removal
from the original packing, not left in the camera for long periods of time. If
motion picture film is allowed to stand in the camera for a day or so, the film
may dry out and break where the loop was formed when the camera is again
started. The film is adequately protected against moisture loss as long as the
original packaging is intact. When loading the camera, make sure the film and
the camera are at the same temperature—if possible, load the camera indoors.
Static markings are caused by an electrostatic discharge, and they ap-
pear on the developed film emulsion as marks resembling lightning, tree
branches, or fuzzy spots. When static difficulties occur they can usually be
traced to the use of film that has a very low moisture content.
Static markings are not likely to occur if the film is loaded and exposed
within a short time after the original package is opened. In general, field
photography under arctic conditions involves subjects of extremely low
brightness scale and very high levels of illumination. Exposures should be
held to a minimum and overexposure should be avoided.
STorAge
If a cold camera is taken indoors where it is warm and humid, condensa-
tion may form on the lens, film, and camera parts. If the camera is then
taken back outdoors before the condensed moisture evaporates, it will freeze
and interfere with operation; the condensate can also cause metal parts to
rust. One way to solve this problem is to leave the camera, when not in use,
in a room at about 32°F (0°C).
If a camera is left in its case outdoors, the case should be made reason-
ably airtight. In the arctic, blown snow becomes as fine as dust or silt and
can enter the smallest slit or crevice. If allowed to enter the camera around
the shutter or other moving parts, the snow will affect the operation of the
equipment. The speed and timing of motors should be checked frequently.
Batteries should be checked every day and recharged at a base every night,
if possible.
T. R. Stobart, who filmed the first conquest of Mt. Everest, preferred to seal
the camera in an airtight polyethylene or rubber bag and then take the cam-
era into the warmth of indoors. Any condensation takes place outside the
bag, not inside, and the camera remains both dry and warm. This method
has the advantage of keeping the camera from becoming “saturated in cold”
for long periods of time. There is no problem in taking warm equipment
back out into the cold, provided the snow isn’t blowing.
244
|
TroPiCAl CinemATogrAPhy
Heat and humidity are two basic sources of potential difficulty when using
or storing photographic goods in tropical climates. Heat alone is not the
worst factor, though it may necessitate special equipment care and process-
ing techniques and may shorten the life of incorrectly stored light-sensitive
materials. High humidity is by far the greater problem because it can cause
serious trouble at temperatures only slightly above normal, and these trou-
bles are greatly increased by high temperatures.
Associated with these conditions are several biological factors—the
warmth and dampness levels encountered in the tropics are conducive to
the profuse growth of fungus and bacteria and encourage the activities of
insects. Many photographic and other related products are “food” for these
organisms—gelatin in films, filters, leather, adhesives and so on. Even if fun-
gus, bacteria or insects cannot attack materials directly, they can develop an
environment that can. Fungus can also either directly or indirectly induce
corrosion in metals, attack textiles and leather, change the color of dyes,
attack glass, and cause a great variety of other forms of deterioration. The
probability of damage is greater with frequent handling and transportation,
especially under the difficulties met in hunting and scientific expeditions
and in military operations. Exposure to harm is greater when equipment is
used out of doors, on the ground or in makeshift facilities.
Atmospheric condition, with respect to moisture content, is usually
described in terms of “relative humidity.” This is the ratio, expressed as a
percentage, between the quantity of water vapor actually present in the air
and the maximum quantity that the air could hold at that temperature. Thus,
if a given sample of air contains only half as much water as it would at satura-
tion, its relative humidity is 50%.
When the temperature rises, a given space can accommodate more water
vapor, and hence the relative humidity decreases, and vice versa. When air
(or an object) is cooled sufficiently, a saturation point (100% relative hu-
midity) is reached, and below this temperature drops of water or “dew” are
deposited. In any locality, the temperature is much lower at high altitudes, so
that dew is likely to form on objects following their arrival by air transport,
especially when high relative humidity is present at ground level. In tropical
climates, this “dew point” is often only a few degrees below the actual tem-
perature during the day and is reached when the temperature drops at night.
The amount of moisture absorbed by films and by nonmetallic parts of
equipment is determined by the relative humidity of the atmosphere. There-
fore, the moisture absorption of photographic or other equipment can be
reduced by lowering the relative humidity, either by removing some of the
moisture with a desiccating agent or by raising the temperature of the atmo-
sphere where the equipment is stored.
| 245
Extremes of relative humidity are a serious threat to all photographic

materials, even at moderate temperatures. At high temperatures, the effects
of high humidity are greatly accelerated, particularly if the relative humidity
remains above 60%. Extremely low relative humidity, on the other hand, is
not quite so serious, but if it falls below 15% for a considerable time, as is
common in desert regions, an electric humidifier should be installed and set
to maintain a relative humidity of 40% to 50% in the storage area.
Storage of Photographic materials

Sensitized photographic materials are perishable products when stored
under extreme conditions of high temperatures and high relative humid-
ity. Proper storage is therefore important at all times. Fortunately, adequate
protection of sensitized materials can be accomplished at relatively low cost
and without extreme methods. Lightweight portable refrigerators or other
cooling units are available from expedition outfitters and other similar
equipment suppliers. Desiccants are available in bulk or kit form for reduc-
ing the moisture content of the atmosphere where film is to be stored. Fur-
ther, portable electric dehumidifiers are also available to reduce the relative
humidity in larger quarters, such as work rooms, to aid in the comfort of the
occupants. And finally, the film packaging reduces the possibility of damage
when the material is stored under recommended conditions. Usually, there
will be little or no adverse effect to the film if it is stored and handled as
described below.
Black-and-white films can be stored at normal room temperatures in an
air-conditioned room. Color films should always be stored in a refrigerator
at 55°F (13°C) or lower. To avoid moisture condensation on the chilled sur-
faces of the material, take film cans out of the cartons and allow 35mm rolls
to warm up for three hours for a 20°F to five hours for a 75°F temperature
rise above storage temperature. 16mm rolls take about one-third less time.
When the original packaging seal has been broken, films should be ex-
posed and processed as soon as possible. Since the air in a refrigerator is
moist, partially used packages should be returned to the refrigerator in a
sealed container containing a desiccant to absorb the moisture within the
container.
In general, do not keep more film than necessary in stock, particularly
when good storage conditions are not available. Photographic materials can
also be affected by the chemical activity of fumes and gases. Consequently,
do not store films in newly painted rooms or cabinets. All films should be
processed as soon as possible after exposure. If you are unable to do this for
some reason, enclose the films in an airtight jar or can together with a desic-
cant and place them in a refrigerator. Exposed films can be kept for several
days in this way.
246
|
Preparation and Protection of equipment

To save time and avoid damage, cameras and other equipment should be
readied well in advance of departure. It is well worthwhile to have the equip-
ment thoroughly overhauled and cleaned, preferably by the original manu-
facturer, who should be advised as to the type of climate in which it will be
used. Cases, packing material, and moisture-absorbing material (desiccant)
should be obtained for the equipment and supplies. Protection during trans-
portation and storage is readily obtained by the use of hermetically sealed
cans or metal-foil bags or other water/vapor-proof containers and a suitable
desiccating agent. If the containers have been properly sealed and contain an
adequate quantity of desiccant, they will protect the contents practically in-
definitely. There is, however, one reservation and caution: if precision instru-
ments that require lubrication with certain types of light oils are subjected
to high temperatures while in such packing, the oils may evaporate, leaving
a gummy residue on the instrument bearings. This situation may prevent
proper equipment functioning until the equipment can be cleaned and re-
lubricated properly.
The protection of equipment that is in active use requires a somewhat
different approach. The relative humidity can be lowered in an equipment-
storage cabinet that is not used for film storage by burning electric light
bulbs or operating an electric resistance heating unit continuously in the
lower part of the cabinet. The number of lamps should be adjusted to keep
the temperature about 10 degrees above the average prevailing temperature.
Air spaces and small holes should be provided at the top and bottom of the
cabinet and through the shelves to allow a slow change of air to carry off
moisture introduced by the cameras and equipment. The positions of the
holes should be staggered on the different shelves in order to produce a more
thorough change of air. Since high relative humidity favors the growth of
fungus on lenses, filters and other surfaces, storage in such a cabinet will
help reduce fungus growth and may prevent it entirely.
Electric dehumidifiers are now appearing in stores in many of the larger
cities in tropical regions. With these units, whole rooms and their contents
can be dehumidified, provided they can be closed to outside air penetration.
In dehumidified rooms, the humidity will not increase rapidly during short
power failures, as it would in heated closets or cabinets. In a small, tightly
sealed room, an average unit in operation for 12 hours out of 24 can keep the
relative humidity below 60%. This should be checked about once a month
with a relative humidity meter or sling psychrometer. When it is not practi-
cal to use a hot cabinet or electric dehumidifier, equipment should be stored
in an airtight case containing plenty of desiccant. Two cans of silica gel the
size of shoe-polish cans will do a very good job of drying equipment in a
sealed 10-gallon paint can (one with a gasket and a “pound shut” lid).
| 247
A half-pound bag of silica gel works well in a gasketed 55-gallon “open

top” drum that can be sealed with a cover. However, where shipment and
handling are involved, or where the containers are to be opened briefly a
few times, double or even triple the quantity of gel will provide a reserve of
protection. Properly dehydrated containers will momentarily feel cool to an
inserted hand due to rapid evaporation of the normal skin moisture. The
sensation is brief but can be easily detected if one is looking for it. Its absence
means the silica gel needs replacement or regeneration.
If none of these methods is practical, and the equipment must be left in
an atmosphere of high relative humidity, the equipment should be opened
and exposed to the sun at frequent intervals in order to drive out moisture.
The exposures, however, should be kept short in order to avoid overheating.
Cameras loaded with film should not be exposed to the sun any more than
necessary.
Cameras should always be protected from excessive heat because many
of the lenses used on cameras are composed of several elements of glass
cemented together. Because some cements melt at 140°F (60°C) and begin
to soften at 120°F (49°C), it is obvious that the lens elements might become
separated or air bubbles might form if the lens were heated to such tempera-
tures. Cameras should not be handled roughly or subjected to sudden jar-
ring when used at high temperatures, because any slight shock might change
the position of the lens components.
mAinTenAnCe of equiPmenT
One of the best protective measures that can be supplied in the tropics is to
thoroughly clean every piece of photographic equipment at frequent intervals
and expose it to air and sun whenever practical. This is particularly important
for retarding the corrosion of metal surfaces and the growth of fungus or
mold on lens surfaces and on leather coverings. Lens-cleaning fluids and
papers now on the market are recommended for cleaning lenses. During the
tropical dry season, or in any desert areas, dust should be removed from the
lens surfaces with a sable or camel-hair brush before the lens tissue is used, to
avoid scratches. Lens cleaning tissues containing silicones should not be used
for coated lenses. They leave an oily film that changes the color characteristics
of the coating and reduces its antireflection properties. This film is almost
impossible to remove. Leather coverings and cases can best be kept clean by
wiping them often and thoroughly with a clean, dry cloth. Frequent cleaning
and polishing will minimize corrosion on exposed metal parts.
BlACk-And-WhiTe film
The exposure of black-and-white film in tropical areas is strongly influ-
enced by the illumination in the subject shadow areas. The moisture and dust
248
|
content of the atmosphere are important because shadows are illuminated

only by light scattered by particles suspended in the air, except where supple-
mentary lighting or reflectors are used. Thus where the atmosphere is very
dry and clear, objects that do not receive the direct light of the sun appear,
both to the eye and to the camera lens, to be in deeper than normal shadow.
In regions like the southwestern United States or central Mexico, for example,
the brightness range of average outdoor subjects is much greater than it is in
less clear climates. In photographing people, this effect and the high posi-
tion of the sun combine to put the eyes in deep shadow and even sometimes
give the effect of backlighting. Therefore, it is best to avoid taking pictures,
particularly close-ups of people, when the sun is overhead; if you must take
close-ups of people, use reflectors or booster lights to soften the shadows.
Exposure meters should always be used with a reasonable amount of
judgment and experience, and this is especially true in locations with such
unusual atmospheric and lighting conditions. In the jungle areas of South
and Central America, the local farmers often clear and burn large quantities
of trees and brush during the dry season. The smoke, composed of solid par-
ticles, hangs in the lower atmosphere and is not easily penetrated even with
filters. Also, at the height of the wet season in many localities, the water haze
becomes almost as impenetrable as a heavy cloud. Distance photography is
best done a few weeks after the close of the wet season and before burning
begins, or a few weeks after the first rains of the wet season have settled the
smoke particles and before the onset of the wet-season haze.
If extensive photographic work in the tropics is planned, the development
of a few test exposures may prevent major failures. It is usually sufficient to
determine a basic exposure that can then be modified to suit other films or
conditions. Allowance should also be made for different types of subjects.
Beach scenes, for example, generally require about one stop less exposure
than an average subject.
Color film
In general, the exposure of color films should follow the same basic
recommendations given for temperate zone exposure, with due regard to
lighting and scene classification. There are, however, some differences in
the lighting conditions and scene characteristics in the tropics that justify
special considerations.
1. During the rainy season, a light haze is generally present in the atmo-
sphere. When this haze is present, the disk of the sun is clearly discernible
and fairly distinct shadows are cast. Under these conditions, the exposure
should be increased by about one-half stop over that required for bright
sunlight.
2. Frequently the brightness of beach and marine scenes is appreciably great-
| 249
er than encountered in temperate zones. With such scenes the camera

exposure should be decreased one full stop from that required for average
subjects. It should be remembered that the term “average subject” as used
in exposure tables applies to a subject or scene in which light, medium
and dark areas are roughly equal in proportion. It should not be taken
to mean “usual” for a particular location or area. For instance, the usual
desert scene is a “light subject” rather than “average subject,” and should
be exposed as such.
3. When the sun is high overhead, heavy shadows are cast across vertical sur-
faces, very much like those occurring in side-lighted subjects. Therefore,
the exposure should be increased one-half stop more than normal, just as
is recommended for side-lighted scenes. For close-ups having important
shadow areas, a full stop increase in exposure is needed.
4. Many objects in the tropics, not only painted buildings and light-colored
fabrics, but even the leaves of many plants and trees, have a high reflec-
tance for direct lighting. Consequently, with front top- or backlighting
they should be considered average subjects.
5. Very often the colors of nearby objects will be affected by the green light
reflected from nearby bright green foliage. Similarly, in courtyards or nar-
row streets, the side that is in the shade gets much of its illumination from
the opposite sunlit wall, which may be strongly colored. There is little that
can be done to correct for this situation, but it should be recognized as a
possible cause of poor results in color pictures.
250
|
| 251
Filming Television
and Computer Displays
by Bill Hogan, Sprocket Digital and
Steve Irwin, Playback Technologies, Inc.
W hen filming television screens, video monitors or computer displays,

there are two principal obstacles to achieving consistent and clear
images on film. The two problems are the difference in frame rates between
the film camera and the TV/computer display, and the color temperature
difference between the film stock and the display technology.
Solutions for successfully photographing various display technologies,
including televisions, computer monitors, LCD monitors, plasma screens
and DLP projectors, vary depending on many factors. Are you shooting a
feature film or episodic project where you must protect and deliver a 24-
frame negative finish? Or are you shooting a commercial or documentary
where you can run the film camera at nonstandard sync-sound speeds to
match the refresh rate of the display?
We will first consider photographing display technologies based on
standard CRT (cathode ray tube) displays, or more simply, glass picture
tubes. This includes nearly all consumer televisions, broadcast and indus-
trial video monitors and a majority of computer displays and consumer
projection televisions. Flat-panel displays and projection technologies
such as LCD, plasma, and DLP will be discussed later, along with color
temperature.
Frame raTes anD exposure Time

North America and many other countries of the world have a television
standard that displays 30 frames per second. Computer monitors can have
refresh rates ranging from 50 to over 100 frames per second. The difference
in frame rates is the predominant difficulty in photographing televisions
and computers as part of a scene.
The basic concept to keep in mind is that of film camera exposure time
based on the combined elements of film camera speed and shutter angle. To
photograph clean, artifact-free images, the exposure time must be equal to
or related by an even sub-multiple of the refresh rate of the display you are
shooting. For example, standard 30 frame NTSC video (the actual frame
rate is 29.97, but we will round off for ease of explanation) is composed of
two separate images called fields. Each field is displayed by the video moni-
252
|
Filming Television and Computer Displays
tor every 1⁄60 of a second. With the film camera running at 30 frames per
second with a 180-degree shutter, the exposure time is 1⁄60 of a second. This
is exactly the right relationship we are looking for. Even though we have now
matched the film camera’s exposure time to our monitor, this still requires
what is generically referred to as a “sync box.” We must align the pulldown
or shutter closed timing of the film camera to occur exactly during video’s
equivalent of pulldown, known as vertical blanking. This is the time interval,
in video, which allows the scanning beam to jump up from the bottom of
the screen to the top and start painting or scanning the next field. The sync
box is fed a reference signal that is in exact speed and time as the video we
are photographing. A phase control on the sync box allows us to adjust the
shutter bar out of the image and precisely match the film camera’s pulldown
to the video’s vertical blanking.
This basic example and explanation holds true whether you are shooting
30-, 25- or 24-frame video. Before we review some newer display technolo-
gies, let’s discuss a few more ground rules for shooting “picture tube” type
displays or projection systems.
over- anD unDerCranking

Because it takes an interlaced video signal a full field to display a complete
image on the monitor, it is not possible to photograph the monitor at any
faster frame rates. Remember the concept of exposure time. For example, if
we are running the film camera at 60-frames per second with a 180-degree
shutter while photographing 30-frame video, our exposure time is only 1⁄120
of a second. The video monitor will only display of its image during this
time. The film camera will only “see” of the image on one frame of film and
the other half of the image on the next frame of film, creating a very drastic
strobe effect. Undercranking on a monitor is possible by keeping one simple
rule in mind. Shoot at even submultiples of the monitor’s frame rate. For
example, if shooting 24-frame video, you can shoot at 12, 6, or 3 frames a
second and have a perfect image on the monitor. If shooting at 30 frames per
second, you can shoot at 15 or 7.5 fps.
CompuTer moniTors
The same rules just discussed apply to standard computer monitors based
on glass picture tubes. The images are still scanned top to bottom, and the
frame rates must still be matched between the display and film camera.
When photographing a single computer monitor “insert style,” and where
you can shoot at nonstandard frame rates, a manual speed control and some
way to measure the computer monitor’s refresh rate are required. There are
two well-known film industry optical frequency meters that allow you to mea-
sure a monitor’s exact frame rate and then set that rate into the speed control.
| 253
Computer monitors are generally noninterlaced, but this has little impact
on shooting them. Since some computer monitors can display very high
refresh rates, it is possible to photograph them at high speeds. For example,
a computer monitor with a 96 Hz refresh rate can be photographed at 48
frames a second. The exposure time of 1⁄96 of a second matches the amount
of time it takes the monitor to display one complete image.
Flat-panel displays
With display technology constantly evolving, there is now and will con-
tinue to be a large variety of new and constantly changing image display
devices. Each one poses unique and technically challenging problems for
successful photography.
LCD monitors and projectors

This is often referred to as the “easy one.” A majority of LCD computer
monitors and projectors can be photographed without regard to matching
frame rates. The light sources and image forming technology known as
“active matrix” creates a picture that is continuous and always on, as far as
the film camera is concerned. Laptop computers fall into this category as
well. There are some low-end LCD technologies, still in use today, known as
“passive” or “dual scan.” These are not part of the “active matrix” type and
may photograph with some breathing or flicker in the image. While viewing
angles are getting very good in high-end LCDs, you must be thoughtful of
the viewing angles of any LCD laptops or monitors you are shooting. Being
only 10 or 15 degrees off the monitor’s sweet spot can cause the image to
wash-out, darken, or shift colors. It is possible to over or undercrank to
some degree on most LCD display technologies. Unfortunately, the only
way to find the limits are to run a battery of tests on the exact model you are
going to use.
plasma displays
Often misunderstood, this display technology is a unique mix of both
CRT and LCD science. The images are created by ionizing a gas which
strikes a phosphor. The images are not always scanned like a traditional
CRT display, but addressed pixel by pixel or row by row without regard for
the input signal’s refresh rate. This can create very annoying and different-
looking artifacts from those seen when shooting standard monitors. Some
manufacturers and other companies have taken steps to modify plasma
panels that allow them to be used in film productions and photographed at
30, 25 and 24 frames per second. When presented with the task of shoot-
ing plasma screens, try to include them in a test day to work out syncing,
exposure and color-temperature issues.
254
|
DLp DispLays anD projeCTors

This fairly recent technology used in a wide variety of video and data pro-
jectors employs Digital Micromirror Devices (DMDs) as the fundamental
image-creating component. Thousands of tiny mirrors acting as individual
pixel or light-switching devices are fabricated using familiar semiconductor
technology. DLP projectors are capable of extremely high lumen outputs
and can deal with very high-resolution video and computer signals. Again,
as in plasma displays, the way individual pixels are addressed and the rate
at which they are turned on and off does not always relate exactly to the
input signal’s frame rate. Often, video and computer inputs are formatted or
scan-converted onto part or all of the micro mirrors, depending on the input
signal’s resolution. When these displays are photographed, various artifacts
such as image flicker and density breathing can occur. It is usually best to
match frame rates between the DLP projector’s input signal and your desired
film camera speed. It is strongly recommended to also shoot your own tests
when dealing with new or different projectors.
viDeo waLLs anD LeD DispLays

Traditional video walls are based on either CRT monitors or CRT/LCD
projection engines. These monitors or “projection cubes” are stacked into
the desired size and shape and fed from a controller or processor that splits
up the input signals across the individual monitors. When faced with a video
wall, find out the underlying display technology and use the guidelines dis-
cussed earlier for them. Recently introduced as a large venue display tech-
nology is the LED video screen/wall. High brightness light-emitting diodes
are used to create individual pixels. These displays are made to be viewed at
long distances and are bright enough to be used outdoors in full sunlight.
LED displays are now being used in sports arenas, at rock concerts, and at
other special event venues, and can pose very unique problems to filming.
With several manufacturers in the marketplace all designing their own con-
trol electronics, special research and film tests will be required to determine
if they can be used practically.
CoLor TemperaTure
Once the frame rate and syncing issues have been worked out, the final
important aspect to be dealt with is color temperature.
Regardless of the display technology you are photographing, most
CRT, LCD, plasma and DLP devices are intrinsically daylight or near-
daylight in color temperature. Most consumer televisions will be at or near
6000-8000°Kelvin. Broadcast and industrial video monitors should be close
to 6500°K. Computer CRT and LCD monitors vary greatly between 5000°K
to 9000°K; some are even as high as 12,000°K.
| 255
Since a majority of these devices will be photographed under tungsten

lighting, some method of color correction/compensation will be required
for proper grayscale reproduction. If an uncorrected display is photographed
under tungsten lighting, the image will take on a very strong blue tone. You
can use standard color temperature meters to determine the display’s nor-
mal color temperature by measuring a known white area. This measurement
can be used as a starting point to help in adjusting the display’s internal color
balance controls. A majority of computer monitors and projectors are now
equipped with menu functions and presets to change the color temperature
of the image. However, these internal controls will not always allow the dis-
play to reach 3200°K. A possible alternate solution is to gel the monitor or
filter the projector, but these methods are not always practical.
ConCLusion
A majority of productions will hire a specialist to be responsible for play-
back and/or projection. They will usually precorrect the playback material or
display device to correct for tungsten photography. The playback specialist
will also assist you in setting exposure and work with the 1st AC concerning
film camera sync equipment setup and operation.
There is a large amount of misinformation circulating in the industry, and
it is very difficult to keep up on all the display technologies as they evolve.
Often, manufacturers will update or introduce new products several times a
year, and what worked for you six months ago may not work the same way
now. It cannot be emphasized enough to test any new or different display
device when considering its use in your production.
Editor’s note: For photography requiring synchronous sound recording that in-
cludes a TV screen, video monitor or certain computer displays, 24-frame play-
back is necessary. Refer to the American Cinematographer Video Manual for
further information.
256
|
|257
Digital Postproduction
for Feature Films
by Glenn Kennel and Sarah Priestnall
D igital technology is used throughout the filmmaking process, from pre-

visualization through origination, dailies, editing, conforming, color
grading, to the generation of film, digital cinema and home video masters
for distribution. The finishing processes of conforming, color grading and
generating the various distribution masters are often grouped under the
term “Digital Intermediate”, which is illustrated in Figure 1.
In postproduction, the Digital Intermediate process offers expanded cre-
ative freedom and faster turnaround. Digital conforming allows the movie to
be quickly assembled for market previews, and quickly recut to improve the
audience response. Digital color grading is much more sophisticated than
the traditional color timing tools of the film laboratory, offering localized
control of color, contrast and detail through multiple windows and layers
that effectively allow the cinematographer to “dodge and burn” the pictures.
The color grading tools that have been available for finishing TV programs
and commercials for many years are now available for finishing feature films.
Inputs Digital Intermediate Outputs

Original Reference
Negative Release Prints
Projector
Lab
Processing Lab Printing
3D Color LUT
Dailies
Telecine Film IN
Conform Color Archive

Editing with Grading Digital Source Laser Recorder
Opticals Master
Scan O-neg Dust Centralized Render Digital Cinema

Selects Busting Storage Color Distribution
Transform Master
Ingest from Titles and Visual Color HD Letterbox

Digital Camera Credits Effects Grading Master
SD Letterbox
Master
SD Pan/Scan
Master
Figure 1. Digital Intermediate Process.

258
|
Digital Postproduction for Feature Film
It is important to note here that digital technology is not a panacea and

is not a substitute for good lighting and skilled cinematography. Although
substantial creative control is provided by the selective and localized tools
in today’s color correctors, you have to start with a sharp and well-exposed
image or the final results will be compromised.
Digital Intermediate also supports better picture quality. Digitally graded
pictures can be more precisely tuned and more consistent than traditional
film prints. Blowups from Super-35 to 2.39 anamorphic format, or from
Super-16 to 1.85 format, are much cleaner and sharper than optical blowups.
And, increasingly, multiple printing negatives are being created so that every
print can be a first generation “show” print, eliminating the degradation of
the traditional IP/IN analog duplication process.
In addition to making the film printing negative for traditional cinema
distribution, the Digital Intermediate process produces a digital cinema dis-
tribution master, and the various home video masters. This consolidates the
creatively supervised color correction steps, saving time and money. Digital
Intermediate also enables the use of lower cost 3-perf 35mm and Super-16
film formats, as well as the mixing and matching various film and digital
origination formats. All of which can represent substantial production cost
savings over traditional 35mm film.
BrieF Digital History

Digital technology has changed the way that feature films are made,
replacing traditional processes with new ones that offer expanded creative
freedom, faster results, and lower costs. Hollywood has been slower to adopt
digital imaging technology than most other imaging industries. One of the
key hurdles has been the overwhelming size of a digital movie—a movie
scanned, stored and processed in 2K resolution is around 2 TB in its finished
form, and could easily be 50 TB or more before it is edited. And these num-
bers are four times larger if you are working in 4K.
Digital technology began to make inroads into the moviemaking process
about 20 years ago. In the mid-1980s, Avid introduced the Media Composer,
a nonlinear editing system that revolutionized the way movies were edited.
Avid wasn’t the first to offer a nonlinear editing system, but was the first to
make it simple to use and relatively inexpensive. In a few short years, the
industry converted from cutting and splicing film on a Moviola to nonlinear
editing on computer-based systems.
Following the pioneering work at ILM, Kodak introduced the Cineon
Digital Film System in the early 1990s to provide an efficient means of
converting film to digital and back to film, with the intention of offering a
computer-based environment for the postproduction of visual effects and
for digital restoration. Kodak established Cinesite, a service bureau that
| 259
offered film scanning and film recording services to enable the industry.
Within a few short years, visual effects converted entirely from traditional
optical printers and manual rotoscoping processes to digital compositing,
3-D animation and paint tools.
Kodak’s original goal for Cineon was a complete “electronic intermedi-
ate system” for digitally assembling movies. In the early 1990s, it was not
cost effective to digitize and assemble whole movies this way, but in choos-
ing computer-based platforms over purpose-built (traditional television)
architectures, these costs came down steadily, driven by the much bigger
investments that other industries were making in computing, networking
and storage technologies.
By the end of the 1990s, Kodak had discontinued the Cineon product line,
but Apple, Arri, Autodesk, Filmlight, Imagica, Quantel, Thomson and other
companies were offering products that provided the necessary tools for the
digital postproduction of movies. Kodak chose to focus its energies on the
demonstration and promotion of a new process dubbed “Digital Intermedi-
ate” that offered to revolutionize the traditional lab process of answer print-
ing and release printing. The digital intermediate process provided digital
conforming and color grading tools that opened up new creative opportu-
nities, while improving the quality of the release prints. In 2000 Cinesite
pioneered the digital intermediate process on the Coen Brothers’ O Brother,
Where Art Thou utilizing the Thomson Spirit Datacine and a Pandora color
corrector, working in a digital, tapeless workflow at 2K resolution.
Since that early start, the supporting technology has evolved substan-
tially with software-based color correctors offered from Autodesk, Filmlight,
daVinci and others, and faster film scanners and recorders available. Within
five years, the industry had embraced the digital intermediate process, with
a majority of major Hollywood films, and many independent films, now
finished digitally.
Digital Dailies
With the move to offline editing in the 1990s, and the increasing scrutiny
of the production process by studio executives, video dailies were widely
embraced. However, while the low-resolution, compressed video formats
may have been good enough for editorial or continuity checking, they were
not sufficient for judging focus, exposure and the subtleties of lighting. For
this reason, most cinematographers demanded traditional film dailies al-
though this was sometimes overruled for cost savings.
Now that HD infrastructure is widely available in postproduction, and
the costs of the transfer, playback and projection displays have come down
to reasonable levels, HD dailies are commonly used for the critical dailies
review by the cinematographer and the director. Certainly, 1080/24p HDTV
260
|
transfers with a Spirit Datacine represent a huge step up in resolution, mak-

ing it possible to judge focus and detail in dailies.
However, high quality dailies require more than just HD resolution. Com-
pression technology is generally used to reduce the cost of transport and
playback. It is critical that the encoding bit rate be high enough to eliminate
compression artifacts, so that the quality of the images can be judged. For
1080/24p this means a bit rate of at least 35 Mbps for MPEG2 or H.264
encoded content, and while this bit rate is supported by most servers, and
HD media like Blu-ray disks, it is not artifact free.
High quality dailies also require a color-calibrated telecine transfer. Many
facilities offer best-light or timed video transfers with the colorist making
the picture look good on a traditional CRT monitor. While this produces
good pictures for editorial or executive use, it does not produce a picture
that looks like the final film product, and can create false impressions. Fur-
thermore, this method provides no feedback on exposure level. A better ap-
proach that is now being embraced by some facilities is an extended range,
best-light scan that protects the full range of the original negative, while
displaying the picture through a look-up-table (LUT) that emulates print
film. The picture can be “printed” up or down by applying offsets calibrated
to match traditional printer lights. This provides the exposure feedback that
a cinematographer needs, along with the capability of “printing” the picture
up or down in the dailies playback to explore the range of the negative. Edi-
torial and executive copies are created with the LUT burned in so that the
pictures that they see include the print film characteristic.
For dailies review, it is important that a color-calibrated digital projector be
used in a dark theatre or location trailer. Small HD monitors or consumer flat-
panel displays are not acceptable for accurate color reproduction or visualiza-
tion of the scale of the movie. Recently, several 1080p projectors have been
introduced for the home theatre market that offer good contrast and stable
calibration at a reasonable (under $10,000) price point. However, calibration
and stability is critically important, so it is best to rent these from a facility that
can provide experienced technicians for installation and calibration.
Communicating the creative intent is another challenge. Previsualization
or “look management” products are available from Gamma and Density, Iri-
das, Filmlight and others that allow the cinematographer to capture a digital
still and apply color correction on a laptop computer. Alternatively, some
cinematographers use Adobe Photoshop. These systems can create a DPX
or JPEG file to be sent to the dailies colorist as a proxy for guiding the color
correction, along with notes or verbal instructions. However, this method
takes an assistant to capture and catalog the digital pictures on the set, as
well as time for the cinematographer to grade the pictures, so these tools see
limited use.
| 261
Furthermore, the pictures still require interpretation and visual matching.

Recently, the ASC Technology Committee has defined a format for inter-
changing simple color metadata, via a Color Decision List (CDL). Many of
the color corrector manufacturers have implemented the ASC CDL, provid-
ing a means of translating the CDL into their internal color correction pa-
rameters. Facilities such as Laser Pacific (aIM Dailies), Technicolor (Digital
Printer Lights) and E-Film (Colorstream) are early adopters of the ASC CDL
into their internal workflows.
The ASC CDL information can be inserted into the Avid Log Exchange
(ALE) files that are generated in telecine or video dailies, so that it can be
carried through the editorial process and into the conformed DI. Here,
the color correction decisions made on-set or in dailies can be used as the
starting point for the final DI grading. This has been implemented in a few
facilities for their internal workflows, but effective execution requires careful
and continuous calibration of all telecines, scanners and reference displays.
Data Dailies
With modern scanners offering high speed scanning capabilities, and with
the cost of storage coming down, it is now possible to forego traditional tele-
cines and videotape workflows and scan dailies footage directly to data. This
approach has been pioneered by E-Film in their CinemaScan process. Dailies
are scanned to the SAN as 2K DPX files, where they are graded, synched and
logged with software tools. The scanned files are archived to LTO4 tape and
stored in a massive tape library, where the select shots can be pulled based on
the editor’s EDL, and conformed for the DI finishing step later.
scanning
Once the film has been edited and is “locked”, the editor provides an Edit
Decision List (EDL) to the scanning facility. There is no need to cut the
negative. The EDL is translated to negative pull lists with keycode numbers
marking the in and out points for scanning. The scanners provide auto-
mated scanning based on frame counts or keycodes, so it is actually easier
to work with uncut camera or lab rolls, rather than compiling the select
shots on a prespliced roll. And the negative is cleaner without this extra
handling step.
Several manufacturers provide digital film scanners that are used for
digital intermediate work. The Thomson Spirit Datacine, and its successor
the Spirit4K scanner which provides real-time 2K scans, is widely used. The
ARRIscan and Filmlight Northlight pin-registered scanners are also popular,
but run at a somewhat slower rate. Since the Spirit4K is edge-guided, picture
stability depends on clean film edges and tight transport velocity control.
Steadiness is seldom a problem, but some people prefer pin-registration,
262
|
which has always been used for visual effects shots. All of these scanners are
calibrated to output logarithmic printing density in 10-bit DPX files.
Framing is a critical issue since camera apertures and ground glasses vary.
It is extremely important to shoot a framing chart with each production cam-
era and to provide that chart to the scanning facility for set up and framing of
the scanner. In addition to defining the scan width, it is important to define
the target aspect ratio. With the increasing use of Super-35 (full aperture)
cameras, the target aspect ratio of 1.85 or 2.39 must be specified, as well as
whether it is desired to scan the full height to use for the 1.33 home video ver-
sion. Table A summarizes the typical camera aperture dimensions and scan
resolutions for popular motion picture production formats including Acad-
emy, Cinemascope, Super-35 and Super-1.85, 3-perf 35mm and Super-16. All
of these are transferred to 35mm inter-negatives for release printing.
TABLE A:
Image Sizes, Scanned Dimensions and
Aspect Ratios for common film formats
Camera Aperture 2K Scanned Camera Projection
Film Format Dimensions Dimensions Aspect Aspect
(H x V) (H x V) Ratio Ratio
Academy 22.0 x 16.0 mm 2048 x 1332 1.37:1 1.85:1
Cinemascope 22.0 x 18.59 mm 2048 x 1556 1.18:1 2.40:1
Super-35 24.92 x 18.67 mm 2048 x 1536 1.33:1 2.40:1
Super-1.85 24.92 x 18.67 mm 2048 x 1536 1.33:1 1.85:1
3-perf 35mm 24.92 x 13.87 mm 2048 x 1144 1.79:1 1.85:1
Super-16 12.52 x 7.41 mm 2048 x 1234 1.66:1 1.85:1
2K or 4K
One of the biggest debates within the technical community has been the
question of what resolution to use for digital cinema mastering and distribu-
tion, which in itself is a continuation of a longstanding debate about the
resolution of a frame of motion picture negative film. The studios are evenly
divided between two opinions—2K (2048 pixels wide) is good enough and
more cost effective, or 4K (4096 pixels wide) is better and will raise the bar.
This debate raged for nearly a year, before DCI arrived at the grand com-
promise of supporting both 2K and 4K digital cinema distribution masters,
using the hierarchical structure of JPEG2000 compression to provide a com-
patible delivery vehicle.
Today, most Digital Intermediate masters are produced at 2K resolution
although more and more are being produced at 4K as costs come down.
| 263
Working in 4K requires four times the storage and four times the render-
ing time as 2K. The creative color grading process for 4K can be done in
essentially the same time as 2K, because proxies are used to support interac-
tive adjustment and display. The color correction can be rendered to the 4K
images once the reel has been approved. Several of the software-based color
correctors support this architecture, including products from Autodesk,
Filmlight, Digital Vision, and DaVinci.
So what do you get for working in 4K? Does it produce pictures that are
twice as good? If the original film format is 35mm and the release format is
35mm (the only viable film distribution format except for 70mm IMAX),
then the answer is no. The difference between 2K and 4K is much more subtle,
and very dependent on the quality of the camera lenses. A seminal paper by
Brad Hunt of Kodak in the March 1991 SMPTE Journal describes the basis
for Kodak’s development of a 4K High Resolution Electronic Intermediate
System in 1992 (which became Cineon), using a system MTF analysis to
illustrate the effect of sampling resolution on the resulting images.1 This is
reproduced in Figure 2.
REL response Camera neg

1.2 1000S/W
2000S/W
1.0 3000S/W
4000S/W
0.8 5000S/W
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60
Frequency (cycles/mm)
Figure 2. Modulation Transfer Function (MTF) a a function of sampling structure for Cineon
digital file systems.
Kodak modeled the system modulation transfer function (MTF) as a

function of sampling resolution in order to define the resolution require-
ments for its film scanners and recorders. Figure 2 shows the system MTF
1. B. Hunt, et al., “High Resolution Electronic Intermediate System for Motion Pic-
ture Film”, SMPTE Journal, 100:156, March 1991.
264
|
responses for a sampling series of 1000 to 5000 samples per picture width
on a 35mm Academy aperture. For reference, the MTF of a camera nega-
tive, which includes the negative film and camera lens, is also shown. MTF
curves representing an analog photochemical system do not give you a
single number to judge the resolution of film. If one takes a middle fre-
quency of 25 cycles per mm as a basis, you can see that the system MTF
response increases from about 15% at 1K to 40% at 2K and about 50% at 3K
and 53% at 4K. So the system MTF improves significantly as the sampling
resolution is increased from 1K to 2K, with diminishing returns beyond
that. Kodak decided to design its Cineon system with a base resolution
of 4K samples per picture width. With a nod towards practical and cost-
effective postproduction work, however, Kodak also provided a direct 2K
mode. Even today, more than fifteen years later, nearly all visual effects are
done at 2K resolution.
Kodak’s analysis included the input camera negative film and the output
internegative element. It modeled the aperture response of the CCD film
scanner and the Gaussian spot of the laser film recorder. It did not include
the print film or the film projector, both of which further reduce the system
MTF. Original negative films have improved substantially in the last fifteen
years, now exhibiting higher speed and lower grain. However, the MTF has
not improved much, and typical camera lenses are the same, and today’s
CCD film scanners and laser film recorders have similar MTF characteris-
tics as the original Cineon conversion devices.
Even with an optically band-limited system like a camera negative (filtered
by the camera lens and the MTF response of the film), it is still desirable to
oversample the picture to avoid aliasing. In fact, scanning at 4K, then filter-
ing and downsampling the image to 2K, produces sharper pictures with less
aliasing than a direct 2K scanning process. So, even if the final digital master
is produced at 2K resolution, there are good reasons to scan it at 4K and
“down-rez” it for 2K manipulation.
Some studios are willing to pay a little extra for a 4K digital intermedi-
ate to deliver 4K digital cinema packages to those screens that are equipped
with 4K digital cinema projectors. Although the initial installations of 5,000
digital cinema screens were all 2K systems, Sony introduced their 4K SXRD
projectors in 2006, and Texas Instruments announced their support for 4K
in 2009. There may also be some value in archiving a 4K digital master to
protect for future release options.
Data ManageMent
With most digital intermediate processes, the scanned data is centrally
stored in a Storage Area Network (SAN), where it can be accessed by vari-
ous workstations performing dirt fix, conforming and color correction op-
| 265
erations. Because of the bandwidth requirements of interactive adjustments

and real-time playback, hardware accelerators and high performance disk
systems and output cards are incorporated in many workstations.
In addition to the scanned shots, DPX files from VFX vendors are also
loaded to the networked storage. OpenEXR is gaining support as an inter-
change file format for visual effects, particularly computer generated images
(CGI). These files need to be converted into 10-bit printing density for sub-
sequent conforming with the live action film scans.
Dirt Fix
All film images contain some dust or handling marks that have to be
cleaned up. This process is called dirt fixing or “dustbusting.” Basically, the
process is to examine each frame for white marks caused by dirt, cinch marks
or scratches and to remove these artifacts with a combination of automated
and manual tools. Common tools include MTI’s DRS and PF Clean by The
Pixel Farm. The amount of dirt fixing varies with film format and handling.
Typically, 16mm film requires more work than 35mm due to the higher
magnification. Lab dirt and scratches are twice as big with 16mm format.
Some scanners are now equipped with an infrared sensor that can detect
dirt and either remove the dirt or provide a defect matte for later removal
by software. Kodak licenses a technology called Digital ICE for this purpose,
and it is now available on ARRI and Northlight scanners.
conForMing
The next step is to conform or assemble the shots to the EDL, and render
the optical transitions (fades, dissolves etc.) to match the offline edit. As nec-
essary, the data editor also generates subtitles and inserts visual effects shots,
main titles and end credits.
color graDing
The color of the assembled shots is graded in a mastering suite equipped
with a calibrated digital projector. DLP Cinema™ projectors from Barco,
Christie or NEC are widely used as the reference display for mastering. Most
facilities use a 3-D color display LUT to emulate a film print, implemented
either in the color corrector or an external LUT box. This LUT is tuned for
the characteristics of the specific print stock and lab that will be used for
release printing.
Color grading is more than just adjusting or correcting the color from
scene to scene to provide consistency and continuity. It also helps to impart
the emotional context of the story, and complements the lighting and ex-
posure used by the cinematographer to capture the scene. Color correctors
provide a range of controls, generally grouped into primaries, secondaries
266
|
and windows. The primary color correction controls adjust parameters that
were originally implemented inside the telecine—these control the gain
(white), gamma (midscale) and lift (black) in each of the red, green and
blue records. The secondary controls mix the color primaries to create color
difference signals that allow individual hues to be selected and rotated.
Although the first color correctors were analog processors, digital tech-
nology was incorporated in the 1980s, allowing much more control and
flexibility with higher quality. One of the digitally enabled features was the
introduction of user defined windows that allowed the colorist to draw
mattes that applied selective color correction to the windowed region of the
frame. These mattes can be tracked and animated as the subject or camera
moves. Soft-edged mattes make the edges invisible. This powerful capability
allows the cinematographer to digitally “dodge and burn” shots, brighten-
ing or darkening areas of the frame to enhance the original photography. In
certain circumstances these tools can be used to save time in production,
reducing the need for relighting or waiting for just the right natural light.
Color grading does have its limits, however. You can’t enhance detail that
wasn’t captured on the original negative, so lighting and exposure are still
critical. And creating mattes to selectively grade regions of the picture takes
time and money. Extreme adjustments of color or contrast will amplify
grain, so these color changes don’t come free. It is always best to capture a
sharp, well-lit image on the set, because you can always darken or defocus
portions of the frame in post. It’s much harder to go the other way. The
larger the negative area, the finer the grain structure and the further you
can push the color without artifacts. This means that Super-35 and Cin-
emascope negatives provide more creative flexibility over standard 35mm
Academy or Super-16 formats.
color gaMut
DLP Cinema™ projectors were designed by Texas Instruments to support
a wider color gamut than standard video (ITU Rec. 709). This wider gamut
is commonly called P3, and was supported by DCI and SMPTE DC28 as
the minimum color gamut for digital cinema. This extended gamut is based
on practical mastering tests in which digital images were graded to match
a film answer print, and has served the industry for several years. Figure
3 shows the color gamut of a SMPTE Reference Projector, compared to
HDTV (ITU Rec. 709) and samples from Kodak Vision™ color print film.
There are some film colors that fall outside of the gamut of the digital pro-
jector, particularly dark cyans, but most of the film gamut is enclosed. It
is useful to note that the digital projectors are also capable of reproducing
brighter primary colors (red, green and blue) since print film is a secondary
color system with the image composed of cyan, magenta and yellow dyes.
| 267
0.9000
Measured Film Data
Compared to Reference Projector
0.8000
0.7000
0.6000
0.5000
y
0.4000
0.3000
0.2000
0.1000
0.0000
0.000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000
x
Figure 3. Color gamut of film compared to that of the digital cinema reference projector and
ITU Rec. 709. A series of measured film samples are shown in green, with the reference
projector gamut shown by the dashed redline and the ITU Rec. 709 gamut shown by the
dashed blue line.
FilM outPut
The finished digital master, called the Digital Source Master (DSM) in the
DCI specification and some SMPTE digital cinema documents, is sent to a
bank of film recorders for output to one or more inter-negatives for release
printing. The widely used ArriLaser™ film recorder runs at about 2 seconds per
frame, so a typical 20-minute reel requires 16 hours of continuous recording.
Color calibration is critical, and the 3-D color display LUTs that are used
for the grading process must be built for the specific film stock and pro-
cessing laboratory, because the variations between film stocks and labs are
significant. Film processing is a chemical process that varies somewhat from
day to day, and is a moving target that cannot be precisely matched. It is
important to also recognize that the precision of digital color correction is
much tighter than the process control and simple printer light adjustments
of the film lab.
Each internegative can produce from 1,000 to 2,000 release prints on a
268
|
high speed release printer before accumulating dirt or scratches that make it
unsuitable for further printing. Therefore, a wide release (3000–5,000 prints
or more) will require multiple internegatives. This can be accommodated by
using the conventional IP/IN photochemical duplicating process, but this
introduces another two generations of losses, which reduce the sharpness
and steadiness of the image. The preferred approach is to generate multiple
internegatives so that each release print is made directly from the digitally
recorded internegative, so that each print is essentially a “digital show print”.
Digital cineMa
Since the movie was graded on a projector that is calibrated to the DCI/
SMPTE Reference Projector specifications, the creative work on the Digital
Cinema Distribution Master (DCDM) is done. All that remains is to render
the 3-D color LUT that was used in the preview process into the digital mas-
ter, along with a simple color conversion to X’Y’Z’. This can be implemented
as a software batch process, or if working in 2K, it can be a real-time output
through an external 3-D LUT box.
The traditional practice in film scanning and recording is to scan and
reproduce the full camera aperture, so that the image remains the same size
on the internegative as the original negative without resampling, and the
resulting print can be projected with a standard projection aperture. The
standard film projection aperture is 5% smaller than the camera aperture,
providing some margin for hiding the frame lines, covering any dirt caught
in the camera aperture and also providing some “wiggle room” for picture
weave (unsteadiness) introduced due to mechanical tolerances in printing
and projection. The traditional practice works fine for film output. However,
the Digital Cinema Distribution Master does not distinguish between the
image structure of the master and the display. In fact, the DCI specifica-
tion explicitly states that the projector must be capable of displaying the full
DCDM, pixel for pixel, without any cropping.
If the same master is converted to a DCDM without resizing, the resulting
digitally projected image will be 5% smaller than the film print shown on a
projector with a standard aperture plate. While this may not be a big issue
aesthetically, it does become a problem if the camera aperture is uneven or
dirty, because now this distracting impairment may be visible on the digital
projector, depending on the adjustment of the masking. With most theatres
equipped with variable masking, it is standard practice to bring the masking
into the picture area a little to provide a good clean black boundary to the
picture. But this may not be enough.
Another option is to resize the picture for the DCDM, pushing in so that
95% of the original scan (1946 W) is resized to 2048 W, insuring that the digi-
tal picture is projected at the same size as the film print. However, this means
| 269
that all DCDMs will suffer a small sharpness loss due to the 5% enlargement.
Although this is probably acceptable operationally, it is inconsistent with the
objective of optimizing the quality of the digital cinema presentation.
At the time of this writing, a proposal has been made to SMPTE to define
a Safe Image Area for digital projection, similar in concept to the film pro-
jection aperture.
HoMe ViDeo Masters

The next step is to generate the video masters for DVD and television
distribution. Traditionally, home video masters were generated by trans-
ferring a color-timed IP on a telecine, with a colorist grading each scene
on a calibrated video reference monitor, under the supervision of the cin-
ematographer or director if they were available. This process often took an
additional two to four weeks. With digital intermediate, much of the work
of creating the various video masters can be automated, and that which re-
quires supervision and approval can be accomplished quickly in a couple of
days while the files are still on line and the cinematographer is still available.
Conversion to video for output requires two important processes. The first is
framing and composition for the narrower video screen, which involves a pan-
ning operation to extract the appropriate part of the widescreen frame for the
full-frame 4:3 standard definition video outputs. This process is called “pan/
scan”. With digital intermediate, this means overlaying a 4:3 framing rectangle
and moving it as necessary to capture the most important part of the action.
The second process is to correct the color and tone scale of the picture for
the very different characteristics and viewing conditions of a home video
display. Remember that the movie was color graded on a digital cinema pro-
jector in a dark theatre. The home viewing environment typically includes
some room light which fills in the shadows and increases the perceived
contrast of the image. SMPTE RP166 defines a Recommended Practice for
Critical Viewing Conditions for the Evaluation of Color Television Pictures
that prescribes a dim surround that is 10% of the peak luminance of the
monitor and matches the D65 color temperature of the monitor.
But grading for video is more than just a contrast correction for surround
illumination that could be implemented with a simple color transform. The
gamut of the HD video display (ITU-R Recommendation 709) is much
smaller than that of print film or the Reference Display for digital cinema. So,
it is necessary to map some colors that fall within the displayed color gamut
of reference projector, but are outside of the color gamut of the video monitor.
Although this could be implemented in a global 3-D LUT that compressed
out of gamut colors, this compression must be implemented as a soft-clip or
gradual ramp that also affects some of the in-gamut colors. So, it is always
best to look at each scene and selectively apply the gamut mapping only when
270
|
required. For these reasons, a color trim pass is often used to generate the
video masters. However, this trim pass can typically be performed in a day or
two, rather than the 2 to 4 weeks of the traditional mastering process.
arcHiVal eleMents
As we move from a film to a digital world, the question of what materials
to archive is stirring heated debates. The biggest problem is that there is no
accepted archival digital medium today, due to both media deterioration
and software and hardware obsolescence. Magnetic tape formats are not
expected to last for more than ten years or so, requiring an expensive migra-
tion strategy to protect the data. The lowest risk approach today to archive
the finished digital master is to make black and white film separations that
have a proven shelf life of over 100 years (if stored properly) and can either
be reconstructed photo-chemically or redigitized.
Which elements should be saved for the future? With traditional film
production, it is common practice to store the original camera negative,
a timed IP, and an answer print in addition to the black and white separa-
tions. A HD 1080/24p master tape is also stored as the home video master.
A similar approach should be taken in the digital world. A comparison of
archival elements for film and digital postproduction is shown in Table B.
TABLE B:
A comparison of archival elements
for traditional film and digital processes.
Element Film Process Digital Process
Original Original negative (o-neg) Original tapes
Color corrected Digital Source Master
Timed Interpositive (IP)
master (DSM)
Digital Cinema Distribution
Reference picture Answer Print
Master (DCDM)
Video master HD 1080/24p HD 1080/24p
Black and White Black and White
Long term storage
Separations Separations
The original negative or original digital tapes should be stored. The con-
formed and graded Digital Source Master (DSM) is the equivalent to the
timed IP. Although the color grading has been applied, it preserves much of
the range of the original photography if the movie needs to be remastered in
the future for a high dynamic range display. The most common file format
is DPX, with the data stored as 10-bit printing density. In the future, the
Open EXR file format may also be used, storing linear RGB data in floating
point form. It is important to tag the data with the color characteristics of
| 271
the Reference Projector used in mastering, and the 3-D color LUT used to
emulate print film.
Although the various distribution formats can easily be regenerated from
the DSM, it is also worth saving the output Digital Cinema Distribution
Master (DCDM) in TIFF-16 file format, storing 12-bit X’Y’Z’ color data,
as well as the HD 1080/24p home video master. The DCDM serves as the
answer print for color reference in future releases or restorations as well as
the master from which additional copies can be made. The HD 1080/24p
format is the master from which all other home video formats can be made.
aDDitional consiDerations
For Digital caMeras
Digital cameras are increasingly being embraced for motion picture origi-
nation. Just as with film, it is important to run preproduction camera tests
to explore the range of the camera and to optimize the setup. Calibration is
important with both the camera setup and the post facility that is generat-
ing the dailies or Digital Intermediate. Pioneered by the Panavision Genesis,
Thomson Viper and ARRI D-21 cameras, many cameras now offer a film-
style logarithmic storage mode, that captures an extended dynamic range
that provides grading flexibility in postproduction, much like negative film.
Data storage decisions affect production and postproduction. It is criti-
cal that you consult the postproduction facility that will be supporting your
dailies and finishing work. Many digital cameras store the pictures on
HDCAM-SR tape which is efficient to unload and reload on the set, as well
as quick to load into postproduction. However, this does require the rental
of relatively expensive HDCAM-SR recorders.
Another option is to temporarily store the pictures as data files on portable
flash cards, flash magazines or disk packs. Both compressed and uncom-
pressed storage solutions are available, but it is important to understand the
tradeoffs. Later, these digital files may be transferred to tape, or transferred
directly to working storage in the postproduction facility.
S.two and Codex offer uncompressed HD data storage on their disk-based
field recorders, which can be tethered to the camera via dual HD-SDI cables.
In 2009, S.two introduced the OB-1, providing dockable onboard storage of
uncompressed HD data via flash magazines. Also in 2009, Panavision in-
troduced their dockable Solid State Recorder SSR-1 for the Genesis camera.
The S.two and Codex field recorders also support the ARRI D-21’s
ARRIRAW format, which stores the full 12-bit signal range and full 1.33
frame area in a packed data format that can be transmitted across dual HD-
SDI cables, a proprietary mode that ARRI calls T-LINK. This means that the
ARRI D-21 camera can support traditional anamorphic lenses for widescreen
photography without compromising vertical resolution. ARRI supplies a free
272
|
software utility, the ARRIRAW Converter (ARC), which allows the ARRI-
RAW files to be converted to full frame 1.33 DPX files in postproduction.
Several cameras also support compressed data storage with onboard flash
cards or disk drives, including the RED One (r3d), Panasonic 3700 (P2),
and Sony EX-3 (XD-CAM) cameras. The RED camera uses a proprietary
wavelet compression codec, which produces reasonable results at a compres-
sion ratio of about 10:1. RED also provides software utilities for high qual-
ity rendering to DPX files for finishing. The P2 and XD-CAM formats use
much higher compression ratios, and while the resulting artifacts are not a
problem for editorial, they are not the best solution for finishing.
With any of these data formats, the production company must consider
the need to backup or archive the production footage to tape for storage. The
most widely used approach is LTO3 or LTO4 tape. For uncompressed data,
this tape backup takes about 3:1 run time and can be performed on location
or back at the post house that is providing dailies support.
| 273
ASC Color Decision List
by David Reisner and Joshua Pines
ASC Associate Members
I n most modern workflows, images—whether originated on film or with

digital cameras—spend most of their time and are processed in digital
form. Often, multiple facilities, with different color correction systems, are
involved in creating and processing elements for a show.
The American Society of Cinematographers Technology Committee, DI
subcommittee—with participants from the ASC, color correction system
vendors, postproduction facilities, and color scientists—created the ASC
Color Decision List (ASC CDL) to allow basic primary color correction
data to be interchanged between color correction systems made by differ-
ent manufacturers. This standardized interchange, when accompanied by
good communication and consistent viewing conditions, can create the
same results on multiple systems, substantially improving the consistency
and quality of the resulting images while also increasing efficiency and re-
ducing cost. The ASC CDL is widely used in motion picture production,
television production and visual-effects. The ASC CDL is the primary Look
Modification Tranform used to express artistic intent in the wide gamut,
high dynamic range Academy Color Encoding System (ACES).
The ASC CDL defines a small set of operations—transfer functions—that
provide the most basic set of color correction operations—operations that
can be implemented in all color correction vendors’ systems. They are Slope,
Offset, and Power which are applied to each of the R, G, and B signals inde-
pendently, and Saturation which operates on R, G, and B in combination.
Thus, ten parameters describe any ASC CDL color correction.
Most color correction systems natively provide vendor-specific Lift, Gain,
and Gamma functions. Slope, Offset, and Power are similar but mathemati-
cally purer functions. In most cases, vendor-specific Lift, Gain, and Gamma
can be easily translated into some combination of Slope, Offset, and Power
for interchange between systems.
The ASC CDL defines mathematical operations that are applied to all
image data, regardless of the format/encoding of that data. The ASC CDL
defines the math, not the encoding-specific interpretation of that math. Each
of the operations will have quite different results on log data than on “linear”
data. For example, a Slope of 2.0 multiplies the image code values by 2.0. If
the image data is linear, that will brighten the image by one stop. If the image
data is log, a Slope of 2.0 will double the contrast of the image.
274
|
Full Production Workflow Using ASC CDL
Film Film Calibrated

Aquisition Camera Telecine /
Scanner
Raw Image Data
Digital Digital ASC On-set

On-set Camera CDL Viewing
Raw Image Data ASC CDL
(XML or protocol TBD)(1)
Dailies Color
Dailies Correction Dailies
Dailies
ASC CDL (Look baked in)
(ALE, FLEx, XML)
Editorial Editorial(2)
ASC CDL
CMX EDL (CMX EDL, XML)
DI
Conform Color
Process Conformed Correction
Raw Image Data
Final Color Corrected Master
1) Currently, this communication is performed via the ASC CDL XML format or various vendor-specified methods.
A communication protocol for this step will be defined in a future ACS CDL release.
2) Editorial maintains association of ASC CDL parameters and shot.

ASC CDL parameters are passed through Editorial unchanged other than copying from one file format to another.
Figure 1. ASC CDL production workflow chart
The application of the ASC CDL provides no “interpretation” of the

underlying metric of the image code values. A given correction applied on
two systems that are assuming data in different formats, and therefore apply
different input or viewing transforms, will almost certainly produce incon-
sistent results. But awareness of the data metric and encoding-customized
operation is outside the scope addressed by the ASC CDL.
And a correction applied to the right data in the right way on two systems
| 275
must still be viewed on calibrated displays with the same characteristics and
in very similar viewing conditions to communicate the intended look. Co-
ordination of data metrics and viewing conditions is also outside the scope
of the ASC CDL and must be handled elsewhere in a project’s workflow.
Although the ASC CDL functions are intended for purposes of inter-
change—to communicate basic color correction operations from set to
facility and between systems from different vendors—many vendors also
provide user level controls that operate on ASC CDL functions directly.
In a workflow, the set of ten ASC CDL parameters for a correction is
interchanged via ASC CDL-defined XML files, by new fields in ALE and
FLEx files, or special comments in CMX EDL files. Most often the formats
shown in Fig. 1 will be used, but the exact methods used will be facility- and
workflow-dependent.
ASC CDL corrections are metadata that are associated with shots. Unlike
LUT corrections, the ASC CDL corrections are not baked-in to the image.
Because of this approach, corrections later in the workflow can be based
on earlier ASC CDL corrections without modifying the image data multiple
times—yielding highest possible image quality. And sharing an ASC CDL
correction gives information about how the earlier corrector was thinking
about the look. Corrections implemented with LUTs are fixed—they can be
viewed, or additional corrections can be layered on, but they cannot practi-
cally be adjusted or tuned.
The ASC CDL supports only the most basic color correction operations.
Not all operations of interest (e.g., log/linear conversions, 3D LUTs, window-
ing, tracking) can be expressed with these operations. It is possible that future
releases will support a somewhat expanded set of interchangeable operations.
The ASC CDL does not handle everything necessary to communicate a
look. A project must manage and communicate basic and critical information
like color space, data representation format, display device, and viewing en-
vironment. To communicate a look between on-set and post or between post
facilities absolutely requires that information be shared and used intelligently.
ASC CDL TRANSFER FUNCTIONS

Although generally similar to the common Lift, Gain, and Gamma
operations (which, incidentally, vary in detail from system to system and
manufacturer to manufacturer), the ASC CDL defines a set of three transfer
functions with unique names and simple definitions.
The ASC CDL’s three basic transfer functions are Slope, Offset, and Power.
They are applied in that order—Slope, then Offset, then Power—and are
sometimes referred to collectively as SOP. The transfer functions are in RGB
color space and are applied independently to each color component. These
three transfer functions for the three color components (assuming the cur-
276
|
SLOPE OFFSET POWER

1 1 offset = 0.5 1
slope = 2
power = 0.5
slope = 1 offset = 0
O ut p u t
O u tput
Output
power = 1
0 .5 0.5 0.5
slope = 0.5
power = 2
offset = -0.5
0 0 0
0 0.5 1 0 0.5 1 0 0.5 1
In put Input Input
Figure 2. Slope Figure 3. Offset Figure 4. Power

rent trichromatic systems) can collectively be described by nine parameters.
The vendor-specific Lift, Gain, and Gamma operations found on most
systems, individually or in combination, can be translated into Slope, Offset,
and Power.
The Saturation function was added to the ASC CDL in release 1.2. Unlike
the Slope, Offset, and Power functions, Saturation operates on all three color
channels in combination. The ASC CDL uses a common industry defini-
tion for Saturation (“Rec. 709” weightings). Saturation is applied after Slope,
Offset, and Power.
Vendors may still have their own proprietary saturation algorithms, but
they must support the ASC CDL Saturation algorithm when they are operat-
ing in “ASC CDL mode”.
Slope
Slope (see Figure 2) changes the slope of the transfer function without
shifting the black level established by Offset (see next section). The input
value, slope, ranges from 0.0 (constant output at Offset) to less than infinity
(although, in practice, systems probably limit at a substantially lower value).
The nominal slope value is 1.0.
out = in * slope 0 ≤ slope < ∞

Offset
Offset (see Figure 3) raises or lowers overall value of a component. It shifts
the transfer function up or down while holding the slope constant. The input
value, offset, can in theory range from -∞ to +∞ although the range —1.0 to
1.0 will fit most traditional use. The nominal offset value is 0.0.
out = in + offset -∞ < offset < ∞

If the underlying data is log, then Offset is an interpretation of printer
points—the most common method of color correction in film lab work.
Power
Power (see Figure 4) is the only nonlinear function. It changes the overall
| 277
curve of the transfer function. The input value, power, ranges from greater
than 0.0 to less than infinity. The nominal power value is 1.0.
out = in ^ power (where ^ is “raised to the power”) 0 < power < ∞
Saturation
Saturation provides a weighted average of the normal color (saturation
1.0) and all gray (fully desaturated, saturation 0.0) images. The saturation
operation modifies all color components. Color components are weighted
by the values used in most Rec. 709 implementations of saturation. Satura-
tion values > 1.0 are supported. Values > 4 or so will probably only be used
for special purposes.
Saturation is applied after the SOP (Slope, Offset, Power) operations.
sat is the user input saturation parameter. inR is the input red color com-
ponent value, G green, and B blue. outR is the output red color component
value, G green, and B blue. gray is the fully desaturated gray value, based on
the color compo nent weightings.
gray = 0.2126 * inR +
0.7152 * inG +
0.0722 * inB
outR = Clamp( gray + sat * (inR - gray) )
outG = Clamp( gray + sat * (inG - gray) )
outB = Clamp( gray + sat * (inB - gray) )
0 ≤ sat < ∞
BEHAVIOUR FOR DIFFERENT IMAGE ENCODINGS
The ASC CDL operations perform the same math regardless of the encod-
ing of the image data to which they are being applied. The resulting modifi-
cations to the image will vary a great deal for different image data encodings.
Management of the image encoding and appropriate application of the ASC
CDL operations is the responsibility of the project/show and outside the
scope of the ASC CDL.
VIDEO-GAMMA AND LINEAR (GAMMA 1.0)

ASC CDL operations will have similar effects on images in the common
encodings that are generally linear. Those encodings include linear light (pho-
ton count or energy) often used in CGI; Academy ACES scene-referred linear
which will be showing up more often in both film and digitally originated
material; and video signals which are linear but always have a gamma (power
function) applied—nominally 2.2 but numbers vary in practice. Some custom
278
|
“linear” encodings may have special han- Figure 5a and b.

dling near toe (blacks/crush) or shoulder Video Gamma Examples:
(whites/rolloff). In those cases, the interac- a) Slope
tion of ASC CDL operations and the non-
linear regions will have to be evaluated on
a case-by-case basis.
Slope
For linear encodings, Slope controls the Slope: RGB all 0.5
brightness of the image while maintaining
contrast—like adjusting the f- or T-stop.
Offset
For linear encodings, Offset controls
the overall “base fog” of the image. The
values of the entire image are moved up Slope: RGB all 1.0
or down together, affecting both bright-
ness and contrast. This is not traditionally
a common operation for linear data.
Power
For linear encodings, Power controls Slope: RGB all 1.5
the contrast of the image. b) Offset
Saturation
For all encodings, including linear, Sat-
uration controls the saturation—intensity
of the color of the image.
Offset: RGB all -0.2
The old telecine Lift function—raising
or lowering the darks while holding the
highlights constant—can be achieved via
a combination of Offset and Slope. Simi-
larly, the telecine Gain function can also
be achieved via a combination of Offset
and Slope. Offset: RGB all 0.0
Log
ASC CDL operations will have similar
effects on images in the common en-
codings that are generally log. Those
encodings include printing density (e.g. Offset: RGB all 0.2
| 279
Cineon, DPX), commonly seen from Figure 5c and d.

film scanners and created or imitated Video Gamma Examples:
by other sources; and the various log c) Power
modes output by various digital cameras
to present a more filmlike response with
a wider dynamic range (at least until
cameras put out the high dynamic range
Academy ACES “scene-referred linear”
floating point format). Some “log” en- Power: RGB all 1.5
codings have special handling near the
toe (blacks/crush) or shoulder (whites/
roll-off). In those cases, the interaction
of ASC CDL operations and the special
regions will have to be evaluated on a
case-by-case basis.
In digital intermediate, log images will Power: RGB all 1.0
usually have a film print emulation ap-
plied as an output display transform in
order to see the color corrected images
as they will be theatrically projected. For
this workflow, the ASC CDL is applied
before the film print emulation. (This Power: RGB all 0.5
procedure was applied to the example log d) Saturation
images shown here.)
Slope
For log encodings, Slope controls the
contrast of the image.
Saturation: 0.5
Offset
For log encodings, Offset controls the
brightness of the image while maintain-
ing contrast—like adjusting the f- or
T-stops. This is essentially the same as
Printer Lights but with different values/
units. Saturation: 1.0
Power
For log encodings, Power controls the
level of detail in shadows vs. highlights.
This is not traditionally a common opera-
tion for log data. Saturation: 2.0
280
|
Saturation Figure 6a and b.

For all encodings, including linear, Log Examples:
Saturation controls the saturation—in- a) Slope
tensity of the color of the image.
ASC CDL
INTERCHANGE FORMATS
The ASC CDL allows basic color cor-
rections to be communicated through Slope: RGB all 0.75
the stages of production and postpro-
duction and to be interchanged between
equipment and software from different
manufacturers at different facilities. The
underlying color correction algorithms
are described above.
When ASC CDL color correction Slope: RGB all 1.0
metadata is transferred from dailies to
editorial and from editorial to post-
production, provided that data rep-
resentation, color space, and viewing
parameters are handled consistently, the
initial “look” set for dailies (perhaps from Slope: RGB all 1.25
an on-set color correction) can be used b) Offset
as an automatic starting point or first
pass for the final color correction session.
ASC CDL metadata is transferred via
extensions to existing, commonly used
file formats currently employed through-
out the industry: ALE, FLEx, and CMX Offset: RGB all -0.1
EDL files. There are also two ASC CDL-
specific XML file types that can be used
to contain and transfer individual color
corrections or (usually project-specific)
libraries of color corrections.
ALE and FLEx files are used to trans-
fer information available at the time of Offset: RGB all 0.0
dailies creation to the editorial database.
New fields have been added to these files
to accommodate ASC CDL color correc-
tion metadata for each shot.
CMX EDL files are output from edi-
torial and used primarily to “conform” Offset: RGB all 0.1
| 281
the individual shots into the final edit. Figure 6c and d.

As there is a convention to include shot Log Examples:
specific metadata as comment fields after c) Power
the associated “event” in the EDL file,
it made sense to use this mechanism
to attach ASC CDL parameters to each
“event”. There are two ways of specifying
this in an EDL file—either “inline” or
“via XML reference.” Power: RGB all 1.25
Many vendors are currently using
XML to support their internal data struc-
tures. The ASC CDL includes a hook so
that CDL parameters can reside in simple
XML text files that can be referenced by
the CMX EDL format or other vendor-
specific formats. The two types of XML Power: RGB all 1.0
files are:
1. Color Decision List (CDL): files that

contain a set of color decisions (a
color correction with a reference to an
image) and that may also include other Power: RGB all 0.75
project metadata. d) Saturation
2. Color Correction Collection (CCC):

files which solely contain one or more
color corrections.
Each and every color correction de- Saturation: 0.5

fined in these files has a unique Color
Correction id. Any specific color correc-
tion defined in these XML files can be
referenced by its unique ColorCorrec-
tion id.
Saturation: 1.0
FOR IMPLEMENTORS
The ASC CDL release documents are
intended for implementers and vendors.
The release includes details on the inter-
change formats as well as example code
and test images. To get instructions on
how to get the current release, send an Saturation: 2.0
282
|
e-mail to asc-cdl@theasc.com.
The ASC CDL was created by the American Society of Cinematographers
Technology Committee, DI subcommittee. Participants included ASC mem-
bers, color correction system vendors, postproduction facilities, and color
scientists.
Discussion and questions may be directed to: Lou Levinson, ASC Technol-
ogy Committee, DI sub committee chair: joe.beats@yahoo.com; Joshua Pines,
ASC Technology Committee, DI subcommittee vice-chair: jzp@technicolor.
com; David Reisner, ASC Technology Committee and DI subcommittee sec-
retary: dreisner@d-cinema.us.
| 283
The Academy of Motion
Picture Arts and Sciences
Academy Color
Encoding System
by Curtis Clark, ASC and
Andy Maltz, ASC Associate Member
OvErviEw
T he Academy Color Encoding System (ACES) is an image and color

management architecture designed primarily for the production, mas-
tering and long-term archiving of motion pictures, although it is being suc-
cessfully used in television production as well. Developed by the Academy
of Motion Picture Arts and Sciences in collaboration with many motion-
picture production and color science experts, ACES provides a set of digital
image encoding specifications transforms and recommended practices that
enable the creation and processing of high fidelity images. It offers a larger
dynamic range of scene tones, a wider color gamut and greater mathematical
precision than is possible with 10-bit Cineon encoding or High Definition
Television standards such as ITU-R Rec. BT.709.
ACES also resolves ambiguities frequently linked to transforms between
so-called “log” and “linear” image encoding,1 as well as those associated with
using a variety of display devices during production, postproduction and
mastering. ACES supports flexible image pipeline development for many
different processes including film and digital acquisition, digital intermedi-
ate, visual effects production, remastering and on-set color management.
For cinematographers, ACES provides the following benefits:
◗ Eliminates uncertainty between on-set look management and down-
stream color correction through standardized viewing transforms and
equipment calibration methods
1. The terms “log” and “linear” as currently used in motion picture production are
loosely defined. “Log” can refer to some form of printing density values or log-
encoded RGB values, and “linear” can refer to some form of video code values or
linear-encoded RGB values. While the use of these terms is deprecated in ACES-
based workflows, it should be noted that log-encoded forms of ACES data are being
developed for use in certain color correction operations and for realtime transmis-
sion of ACES data over HD-SDI.
284
|
Academy Color Encoding System
◗ Preserves the full range of highlights, shadows and colors captured on-
set for use throughout postproduction and mastering
◗ Preserves the ability to use traditional photometric tools for exposure
control rather than having to compensate for custom or proprietary
viewing transforms in conjunction with a video monitor
◗ Simplifies matching of images from different camera sources
◗ Enables future expansion of the creative palette by removing the limita-
tions of legacy workflows
Key components and features of ACES include:
◗ A standardized, fully-specified high high-precision, high dynamic
range, wide color gamut color encoding specification (SMPTE ST2065-
1:2012 Academy Color Encoding Specification) encompassing the full
range of image color and detail captured by current and future digital
cameras
◗ A standardized, fully-specified high precision and high dynamic range
printing density specification (SMPTE ST2065-3:2012 Academy Densi-
ty Exchange and the related SMPTE ST2065-2:2012 Academy Printing
Density) encompassing the full range of image color and detail captured
by modern motion picture film stocks
◗ Standardized file formats for colorimetric and densitometric data based
on the popular OpenEXR and DPX data containers
◗ A methodology for display device-independent mastering that pro-
duces higher quality images than are possible with legacy workflows
◗ Recommended best practices for interfacing digital motion picture
camera “raw” data and film scanner output data to ACES
◗ Support for ASC CDL in on-set color management systems
The full benefits of ACES are achieved when cameras, color correctors,
image creation and processing tools, and display devices correctly imple-
ment the ACES specifications and recommended practices.
ACES Technical Description

The standardized Academy Color Encoding Specification (which also
is referred to as “ACES”) is a core component of the ACES architecture.
It defines a high-precision wide color gamut digital image encoding ap-
propriate for both photographed and computer-generated images. From
image capture to theatrical presentation and/or other forms of display,
ACES encodes the image data in a manner that makes it available for
downstream processing, which ultimately allows more flexible and cre-
ative color grading. This is in contrast to legacy 10-bit systems such as
Cineon and ITU-R Rec. BT. 709, whose precision, color gamut and dy-
namic range capabilities are below those of modern digital motion picture
cameras and film stocks.
| 285
ACES is capable of encoding all colors viewable by the human visual sys-
tem (see Figure 1). This greatly exceeds the range covered by HDTV and
Digital Cinema projectors.
ACES is derived from a hypothetical ideal recording device, designated as
the Reference Input Capture Device (RICD), against which actual recording
devices’ behavior can be compared (see Figure 2). As conceived, the RICD
can distinguish and record all visible colors, as well as capture a luminance
range exceeding that of any current or anticipated physical camera. The
RICD’s purpose is to provide a documented, unambiguous, fixed relation-
Figure 1. CIE 2-Degree Chromaticity Diagram with Gamuts showing ACES primaries com-
pared to other color encoding specifications
286
|
ship between scene colors (also called “relative scene exposure values”) and
encoded RGB values. This fixed relationship is integral to a fully specified
imaging architecture, as it provides a reference point for bringing any form
of image source data into the system.
Because ACES encodes scene colors, ACES values must be adjusted for
the target display environment and device characteristics to faithfully repre-
sent the recorded images. These adjustments are performed by the Reference
Rendering Transform (RRT) and a display device-specific Output Device
Transform (ODT) that are, in practice, combined into a single transform.
Original Scene Reference SMPTE Theater

Input Capture Reference Reproduction
Device Projector
Figure 2. ACES Reference Signal Chain
Figure 2 also shows the combined RRT and ODT along with the SMPTE
Digital Cinema Reference projector, a common configuration for mastering
theatrical motion pictures.
When an actual camera records a physical scene, or a virtual camera (e.g.,
a CGI rendering program) constructs a virtual scene, a camera-specific
Input Device Transform (IDT) converts the resulting image data into the
ACES relative exposure values as though the subject had been captured by
the RICD (see Figure 3).
As noted earlier, ACES images are not appropriate for final color evalu-
ation or direct viewing;
like film negative or digital
files containing scanned
images encoded as print- A
ing density they are only Original Scene Digital
Motion Picture
Camera
intermediate representa- A
tions. ACES images can

be examined to identify
image content, cropping
region, sequencing, the Original Scene Reference
Input Capture
amount of shadow or Device
highlight detail and other

qualities. For final color
evaluation, the RRT must B
be used in conjunction Original Scene Digital
Motion Picture
Camera
with a specific display B
device’s ODT to produce Figure 3. Input Device Transforms (IDTs)

| 287
a “color correct” viewable

image (see Figure 4).
Output Device
Figure 5 shows the full
ACES block diagram.2
Using the full range of
A color corrector tool sets,
Output Device A
Look Modification Trans-

B
forms (LMT - not shown
Output Device B for clarity) can be applied
Figure 4. The Referencing Rendering Transform (RRT) and to camera and/or CGI
Output Device Transforms (ODTs) images in ACES to record
basic color adjustments
(using ASC CDL, for example) or to achieve preferred creative looks during
color grading. Examples of preferred creative looks include bleach bypass and
film stock-specific looks, or any custom look that serves the visual storytelling.
Using ACES in practice, as with any production pipeline, requires advance
preparation and sufficient technical support to achieve satisfying results.
Key issues to consider when using ACES include:
◗ If shooting with digital cameras, selection of appropriate IDTs to produce
well-formed ACES-encoded images
◗ If shooting film, proper calibration of film scanners to produce well-
formed ADX-encoded images
Original Scene Film Camera Film Negative APD Calibrated Film Negative Film Print Film Projector Theater
Film Negative Academy Density Film Recorder Reproduction
Exchange
Encoding
ADX to ACES / ACES to ADX

Transforms
Reference Device SMPTE Theater

Transform Reference Reproduction
Projector
Original Scene Reference

Input Capture Academy Color Reference
Device Encoding Rendering
Specification Transform
Output Device Video Monitor

Transform
Original Scene Real Digital Input

Camera Device Transform
Figure 5. Academy Color Encoding System Block Diagram.
2. RDT (Reference Device Transform) is the ODT for SMPTE Reference Projector
Minimum Color Gamut (Nominal) (see SMPTE RP431-2-2007 Reference Projector
and Environment, Table A. 1).
288
|
◗ Use of appropriate RRT/ODT combinations for on-set viewing and mas-

tering, as well as consideration of any LMTs that might be used both
on-set and in the mastering suite
◗ Use of ACES-enabled color correctors and other ACES-enabled image-
processing equipment
With a correctly implemented ACES-based workflow, cinematographers
can rely on properly exposed source images as the best starting point for
visual effect integration and mastering.
For more information on ACES, please visit oscars.org/aces
Curtis Clark, ASC currently chairs the ASC’s Technology Committee and has
done so since it was revitalized in 2002.
Andy Maltz serves as the director for the Academy of Motion Picture Arts and
Sciences’s Science and Technology Council.
| 289
The Cinematographer
and the Laboratory
revised by Rob Hummel
T he relationship between a cinematographer and the laboratory process-

ing his or her film can mean the difference between the revelation of
seeing exactly the images conceived upon exposing the film, or a defensive,
frustrating time spent constantly seeking to achieve the right look. Remark-
ably, and not surprisingly, this all rests on effective communication between
the cinematographer and that person at the laboratory responsible for han-
dling your dailies. Since the production process begins with dailies and prin-
cipal photography, that’s where we’ll begin this discussion.
When working with a laboratory for the first time, it is advisable to con-
tact the lab well in advance of principal photography or any tests to be shot.
Upon locating the individual who will be handling your dailies, you should
discuss the film to be shot, the looks you are trying to achieve, and learn
from the lab what procedures they employ to ensure that your dailies are
delivered with a consistent look. At this time, you should find out what light
points the lab considers to be normal from a properly exposed piece of film,
and whether the lab utilizes RGB or YCM color-timing methods. Ideally,
this process begins with a visit to the lab for a face-to-face meeting.
For reasons discussed below, it should be understood that for the same
piece of film, light point values can vary many points one way or another
depending on where a lab has trimmed their printers. This needs to be un-
derstood so as not to be alarmed when using a lab for the first time and
discovering that your printer lights are 8 points lower than the lab you are
accustomed to using.
Laboratories routinely use the film manufacturers’ recommended speci-
fications for processing, modified to meet their particular equipment. (The
entire system—type of film, manufacturers’ EI recommendation, labora-
tory printing and processing range—is calibrated to produce a pleasing
rendition of fully lighted flesh tones under normal projection conditions.)
In addition to producing normal results on the screen, most laboratories
are able to offer several techniques that can modify the look of a film to a
certain degree; these processes would include flashing of negative or posi-
tive, push/pull processing, ENR and bleach bypass. You should speak to
your lab representative to understand all of the techniques that are avail-
able to you.
290
|
The Cinematographer and the Laboratory
Figure 1. LAD color chart
PrinTer PoinTs
The laboratory controls print density and color balance by increasing or
decreasing the intensity of each primary color of light in steps called printer
points. Since the development of the B&H Model C printer, most manufac-
turers have standardized on a range of 50 light points in 0.025 Log E incre-
ments. In addition to the light points, each printer usually also has 24 trim
settings (0.025 Log E), providing an available total of 74 lights.
When a lab changes the brightness/darkness of a print, they will say they
are adjusting the density of the print (which, quite literally, they are). Density
is adjusted by moving the values of all three light points (RGB or YCM) in
unison. Individual adjustment of the colors is obviously used for specific
color correction, or timing.
This means that at some laboratories, just because your light point is up
at 50, it may not mean there isn’t any more room to go. Sometimes you may
have the ability to go up to the equivalent to a 65 light point. That being said,
at some laboratories it can also mean that when you are at a 50 light point,
there isn’t any more room at all. Again, you must consult the lab you are
using and get advice about the flexibility of their light point scale.
One could argue that the ideal settings for scene-to-scene timing would
be at mid-scale (Trim 12 + Tape 25 = 37 lights). In actual practice the avail-
able range is considerably less. Printer lamps are usually operated under their
rated voltage. This reduces the light intensity in all three colors. For example,
| 291
lowering the voltage from 120 to 90 volts on a BRN-1200 watt lamp results in
a relative change in printer points equal to minus 12 Red, 13 Green, 17 Blue.
The trims are usually used to balance the printer for a given print-film emul-
sion. A typical emulsion might require 16 Red, 13 Green, 10 Blue or, in terms
of the ideal, plus 4 Red, plus 1 Green, minus 2 Blue. Other factors influencing
the available printer points are the operating speed of the printer, and the use
of neutral-density filters in the individual channels and the main light beam.
The sum of these variables explains why a given negative might be printed
28 Red, 29 Green, 22 Blue at one laboratory and 36 Red, 32 Green, 36 Blue at
another laboratory to produce matched prints. It is important to understand
that printer points relate only to how the printer exposes film. A one-stop .30
Log E change (12 printer points x .025 Log E ) is equal to a one-stop exposure
in the camera only if the film in the camera has a gamma of approximately 1.0.
The current negative films, both black & white and color, have gammas of
approximately .65. Therefore, in correlating camera and printer exposure,
one stop equals ⅔ x 12 = 8 printer points per stop. Much testing has borne
out the fact that one camera stop change in exposure equals 8 points of
printer density. At extreme ends of the printer scale it may vary a point or so
from that, but using 8 points is always the best rule to follow.
MoniToring The ProCess

It must be pointed out that all manufacturing processes are subject to vari-
ables and, as such, acceptable tolerances are set to ensure success without
being unreasonably demanding. The same principal applies to the process-
ing of film.
Through the use of a device called a densitometer, labs can measure the
processing of film and determine how far off of a predetermined standard the
negative is printing. By measuring a photographed gray scale, the accuracy
of original negative exposure can be measured. Most major laboratories have
the ability to expose a 21-step gray scale into an original negative, which,
when each step is measured and plotted, can produce a curve that will il-
lustrate a negative’s characteristics (sensitivity, gamma/contrast, highlight/
sshadow detail, etc.).
Labs have control departments whose principal purpose is to constantly
monitor the negative and positive processes, and calibrate the printers.
Printer calibration is, in fact, one of the areas most subject to drift, as the
lamp in each printer is constantly migrating color and brightness as it ages.
Through exposing predetermined exposure strips through the printers, the
control department can keep the printers on target by adjusting their trims.
While film manufacturers can always provide you with the characteristic
curve of any specific emulsion batch, if you’re uncertain about an emulsion
(due to age, transport conditions, possible exposure to radiation, etc.), the
292
|
lab will happily expose a section of your negative with a 21-step gray scale,
process the negative and plot the negative’s curve, so you will immediately
be able to see what condition your emulsion is in. When compared with
the curves supplied by the manufacturer of that emulsion batch, you can
precisely determine if this batch has shifted.
I’d like to make a comment here about professional motion picture nega-
tives. It is always assumed that motion picture negatives will be used within
six months or so of manufacture, and once exposed, will be processed within
72 hours. Negatives older than six months run the risk of a high base fog
density that will result in milky blacks in your print. Motion picture films
don’t remotely have the latent image stability of still-camera films, which is
why they need to be processed promptly or refrigerated at very low tempera-
tures when timely processing isn’t possible.
For consistent tracking and monitoring of processing from lab to lab, the
industry came up with Laboratory Aim Density, or LAD.
With LAD, film laboratories can determine how far off of a reference
standard they are when printing. We are principally concerned with LAD
for color positives printed from original negative; however, there are LAD
values for interpositives and internegatives as well. Again, a densitometer
is used to read a predefined “LAD patch” on a print. The RGB/YCM LAD
value is the same for all Eastman Kodak Motion Picture negatives. Fuji ele-
ments have their own LAD values.
LAD values for all Kodak negatives are: 0.80 Red, 1.20 Green and 1.60 Blue.
Cutting in of a LAD reference frame is standard operating procedure for
answer printing at film laboratories. This enables the answer-print timer
to know exactly where any given print stands vs. the LAD reference point.
Reading of the LAD patch lets the timer know if the print is leaning one
way or another in color or density. It’s always advisable before beginning a
screening of any answer print to ask the timer how the print is, or “What’s
the chart reading?” The timer will then express to you the values for each
color and how far off of standard, plus or minus, the print is.
While not standard, it’s not out of the question to ask to have a LAD patch
cut into your dailies each day, and have the patch read so you can know how
far off of standard your daily prints may be. If this is requested, also under-
stand that it adds one more part to the delivery of your print dailies, and
will delay their delivery ever so slightly. When printing one-light dailies, an
advantage of a LAD patch cut into your dailies is you will know immediately
if color drift is the cause of the laboratory or your own cinematography.
As for tolerance in LAD readings, understand that for camera negative,
0.025 Log E equals one point. For release prints, the lab should be allowed
a swing of 0.05 Log E (two points) on any individual color; however, you
shouldn’t have to be tolerant of anything more than 0.05 Log E sway in any
| 293
of the readings (especially for answer prints), but how tolerant you want to
be is at your discretion.
An example of how this procedure is implemented at a lab might be as
follows:
You travel to the laboratory to screen a reel of your movie with the answer
print timer. As you settle into your seat, you ask, “How did this print come
up?,” “What’s the chart reading for this print?” or “How are we doing today?”
While the timer is provided the precise Log E value plus or minus from the
standard, he or she will usually do some quick math and express its values in
light points. For example, you might hear, “It came up plus one, zero, minus
a half,” which would translate into plus one point red, on target for green
and minus a half point of blue (unless you are at a lab that uses subtractive
colors, where they would be referring to yellow, cyan, and magenta). That
being said, they will happily provide you with the precise Log E values if you
request it, remembering that 0.025 Log E translates into one printer point.
exPosure rePorTing
It used to be normal practice for laboratories to furnish one-light rather
than timed daily rush prints. However, most labs today find that, to be com-
petitive, they must offer scene-to-scene timing of dailies. That being said,
timed dailies usually yield an inconsistent workprint since they end up being
timed on an individual basis each day.
The term “one-light dailies” doesn’t mean that all negatives are printed at
the same light points. The laboratory establishes a day exterior, day interior,
night exterior and night interior light points for each film stock a cinema-
tographer may choose when he or she starts a picture, based on the first few
days of shooting. Each laboratory establishes its own method, but basically
all try to keep usable negative within the 1 to 50 light point scale (laborato-
ries do not necessarily agree on the numerical value of the preferred mid-
scale light point, but this is not critical as long as you know which system
your laboratory uses). A conference with your laboratory representative will
establish methods that fit your style of photography. After that, variation in
your exposure will show up as variation in the density of your solid black in
any area of the print; exposure must be kept at or above what the laboratory
tells you is their average light point for a well-exposed piece of film.
Should you feel uncomfortable committing to one-light dailies, at least
instruct your dailies timer/grader to try and use the same color and density
ratio light point, should you be shooting the same scene for several days at
a time. This will mitigate color shifts from timing the dailies on their own
merits each day.
Negative raw stock from different manufacturers may or may not have
the same base density, maximum density or density/exposure characteristic
294
|
(curve shape), although these differences are usually small. A daily print
made by the LAD control method shows the density and color ratio at
mid-scale on the printer. Negative from two manufacturers, both exposed
correctly, may or may not look the same at this printer point. If necessary, an
adjustment to the printer point may be made for the difference in raw stock,
and this new light point can be used for printing dailies on the subject.
sPeCiaL ProCessing
If special processing is requested, a conference with the laboratory represen-
tative and experimentation (or experience) is desirable. If special processing is
requested, or the cinematographer is using high or low exposure for effect, it is
desirable to test the effect by going through the entire release-print technique,
including the interpositive/duplicate negative generations, and to view the
result as nearly as possible under the anticipated release-print viewing con-
ditions. If the scene to be photographed will be used in an optically printed
visual-effect, it is wise to confer with the appropriate visual-effects experts.
reLease PrinTing ProCedures

After the picture negative and soundtrack negative have been assembled
in their final form, the laboratory will analyze the picture negative for scene-
to-scene color and density variations and make a print known as the first
trial answer print. As many trial prints are made as are necessary to resolve
all printing data. The final approved timing is also often known as a final
answer print. With the data thus obtained, one or more intermediates are
printed and from these the release prints are made. Modern film stocks
used to make the interpositives and internegatives or intermediate negatives
are of excellent quality, but they do entail added printing generations. The
appearance of scenes involving effects such as off-normal film exposure or
processing can suffer if they exceed the extremes the system can handle. (See
also Testing and Evaluation.)
CoLor reversaL FiLMs

Most of the above also applies to color reversal films; however, color re-
versal films are now usually used only when it is intended to project the
original, or for a desired visual effect. Exposure latitude is short compared to
that of color negative films. Proper exposure is therefore critical in order to
keep all scenes at a usable density.
BLaCk-and-WhiTe negaTive and reversaL FiLMs

The above also applies to black-and-white reversal films. Black-and-white
negative films, however, are an exception. Both their contrast and density
are more strongly affected by developing time than are color negative films.
| 295
While there is much more latitude in exposure with black-and-white nega-

tive films as compared to color negative films, both grain and acutance are
affected by exposure variations. Deviation from the manufacturers’ recom-
mended EI (exposure index) should be tested and evaluated.
ForCed deveLoPMenT oF CoLor FiLMs

With the color films most commonly used today, it is possible to compen-
sate for underexposure by extended development, or “pushing.” Similar to
the principles of traditional black-and-white sensitometry, forced develop-
ment of these color films increases their contrast, graininess and the fog level.
Therefore, forced development can never yield the same image quality pos-
sible when films are exposed and processed strictly according to the manu-
facturer’s recommendations. In many instances, however, the image quality
obtained with underexposure and overdevelopment is entirely satisfactory, and
a cinematographer may want to take advantage of this fact when shooting
under adverse light conditions. What pushing means, in effect, is that the cin-
ematographer can deliberately underexpose the film (sometimes by as much as
two stops) and request that the laboratory compensate in development.
With the introduction of high-speed color negative emulsions, there is
less call for pushing the moderate-speed films, except for a special “look”
or when underexposure is unavoidable and high-speed negative is not at
hand. It is possible to push one stop in development without appreciable loss
in image quality. The scenes produced in this manner can be intercut with
scenes exposed and processed normally.
If color negative is pushed two stops in development, the increase in the
graininess and the fog level is substantial, but the results are acceptable for
scenes involving night-for-night photography or available-light photogra-
phy under exceptional circumstances.
Extending development beyond two stops does not appreciably contrib-
ute to the image; rather, it increases the grain and fog level and should not
be attempted even as an emergency measure. It should be realized that with
color films the sensitometric balance of the three emulsion layers is only
achieved with normal processing and that forcing the development does not
accomplish a true compensation for underexposure. Forced development
does not result in a substantial increase in the exposure index of the nega-
tive as measured by accepted scientific methods. Nevertheless, it cannot be
denied that the technique proves to be of some practical value if it brings the
underexposed negative into an acceptable printing range.
Reversal films, unlike negative, derive their projection density from the
camera exposure. Forced processing of underexposed film can bring up the
projection density to normal. Eastman Ektachrome Films 7240 and 7285
(tungsten balanced), can be pushed one stop with acceptable results. In
296
|
emergency situations they can be pushed up to three stops with some loss
in quality. The ability to underexpose these films and still obtain on film a
usable image should by no means be regarded as a suitable substitute for
additional lighting when it can be provided.
If a cinematographer anticipates the need for deliberate underexposure
during a production, he or she should, if possible, shoot careful tests in ad-
vance using the same emulsion that to be used for the production and have
them processed by the lab that will be processing the production film. The re-
sults can then be analyzed with the help of a laboratory representative. Need-
less to say, underexposed rolls should be clearly marked with instructions as
to how much they should be pushed when they are sent to the laboratory.
FLashing
Even before discussing flashing, it must be said that one lab’s measure-
ment of a flash is not likely to match how another lab measures it. Thus, don’t
base a request for a type of flash on an experience at another lab; testing is
critically important.
Flashing may be described qualitatively as subjecting the negative film to
a weak, controlled, uniform fogging exposure prior to development—either
before, during or after photographing the desired subject. There is no mea-
surable difference in the effect if the flashing takes place before or after the
principal exposure. As a result, because of various unfavorable factors (such
as not being able to control the time interval between the flash exposure
and the time that development will actually take place, and not knowing
the actual conditions of photography in advance), preflashing is generally
avoided in favor of postflashing.
Simultaneous flashing during actual photography by means of a special
device attached to the front of the camera lens is described shortly under the
heading Lightflex. However, if not properly controlled, flashing on the cam-
era in the field runs the risk of an imprecise and improperly color-balanced
flash.
Since color negative consists basically of three emulsion layers sensitive
to red, green and blue light, the spectral composition of the light used for
flashing can be a neutral equivalent to tungsten light (3200°K) or daylight
(5500°K) which, depending on the film, would affect all three emulsion layers
equally. The fundamental reasons for using a neutral flash are to reduce the
contrast of the image and to increase shadow detail. This effect is accom-
plished because the flashing exposure affects principally the shadow region
of the negative image.
Another reason for flashing is to achieve certain creative effects by using a
nonneutral flashing exposure that would then alter the normal color rendi-
tion of the developed negative.
| 297
Flashing is also used sometimes to reduce contrast of positive or reversal

films when such films are to be used for special effects duplication purposes,
such as projection backgrounds or aerial image compositing with animation.
Some good examples of negative flashing can be seen in the films McCabe
and Mrs. Miller, Heaven’s Gate and Deliverance, all shot by Vilmos Zsigmond,
ASC. In Heaven’s Gate, Vilmos combined a 10% negative flash with a 30%
positive flash (as measured at Technicolor).
LighTFLex
Lightflex (and Arriflex Varicon) is an on-camera accessory, mounted in
front of the lens, which overlays a controlled amount of light on the scene
to be photographed at the time of exposure. It allows the cameraperson to
modify the gamma curve of the film in the camera during shooting. This
can extend the photometric range of the film, and provide more detail in the
shadow areas without affecting grain.
Lightflex works exactly the same with video cameras, providing more
shadow area detail with no increase in noise. A filter tray between light sphere
and reflector permits colored light overlay for special effects. The system’s
lamphouse holds three quartz bulbs that operate selectively or simultane-
ously to hold color temperature stable over the widest possible range. Com-
pound curved reflectors cover all standard 16mm and 35mm fixed lenses
and zooms.
Intensity control is by a handheld electronic dimmer with a digital 1⁄10-volt
line indicator. The lamphouse front features a built-in Obie light, with
changeable masks for direct or diffused light. The unit housing is mounted
on a swing-away bracket to facilitate lens changes and maintenance. There
are two front-mounted matte box shades for wide angle or normal lenses.
There are two lens filter stages: one for two 3" x 3" filters and one for two
4" x 4" or 4" x 5.6" filters or grads (rotatable). A single 4" x 5.6" filter may be
mounted inside the Lightflex housing.
Panavision’s Panaflasher achieves similar results with an internal device.
298
|
| 299
Emulsion Testing
by Steven Poster, ASC
T he object of these series of tests is to determine the best working exposure

index and the dynamic range for your original camera negative. This sys-
tem takes into account any processing techniques, print stocks, further duping
of the original camera negative and the effects of scanning and digital output.
Judging these tests should be done visually, although densitometer read-
ings should be taken for later reference. It is more important to train your
eye to see the various characteristics of the chain of events that result in
the presentation of images that we create during production than to know
scientifically all of the sensitometery that goes into the imaging system.
PrEParing BEnch TEsT nEgaTivE

The basic physical nature of the film stock (i.e., how much density there is
in the negative without any exposure) must be calibrated.
If you are going to test or use other film stocks and/or processing techniques,
these should also be calibrated at this time. If the negative is going to be push-
or pull-processed, or if it is going to be flashed, or skip-bleach processed, or
if you create a negative by cross-processing reversal stock, these special treat-
ments should also be done in the prescribed way at this time. You can measure
the specific densities of the base density plus fog levels on a densitometer for
reference. (This reference can be used later if there is a change of emulsion or
lab, or just as a simple check on your standard emulsion.)
We know that this specific density will be used to reproduce a black tone
on the final print. If this density on the negative is not printed deep enough
to reproduce a desirable black on the print stock, there will be no black tones
in the final print, and the images will be appear to have been underexposed.
If this is the case, the images can also develop a grainy appearance and will
not dupe well.
The lab should process a short length of unexposed negative no less than
50' in length. At this point you have a piece of unexposed, processed negative
that reflects any pre-exposure or special processing techniques done to that
negative. You should also have noted reference densities of that negative.
This leads us to the second part of the test.
PrinTEd Black TEsT

In order to determine the specific amount of light needed to print your
test negative to a desirable black tone, you must test the print stock and any
300
|
Emulsion Testing
printing techniques (flashing the print stock, ENR, CCR, optical printing,
digital scanning or print skip-bleach processing for example). This is done
by printing your piece of unexposed processed film stock at a succession of
printer lights increasing by 2 to 4 points of density (the equivalent of ⅓ of a
stop of printer exposure at your lab). If you are planning to use any unusual
printing techniques or print processing techniques, they should be applied at
this point. Any subsequent printing for these series of tests should have these
techniques applied as well.
A trick that I have often used to help me judge my optimum black density
is to punch a hole in the negative with a single-hole paper punch before it
is printed. This will give you a reference to zero density in the frame, which
can help determine the optimal visual black tone that you want. Your desired
black tone will never be as black as the portion printed through the hole, but
the reference helps to determine what density you will want to achieve with
your processing and printing techniques.
If your lab has film-strip projectors that they use for timing purposes, this
is a very good way to view these tests. Two identical prints can be made
which can be viewed side by side on these projectors, allowing you to study
the results and compare different densities. If no filmstrip projectors are
available, the length of each exposure should be long enough to allow you
time to view it sufficiently on the screen during projection.
Once you have determined which density you would like to represent black
in your final print, this should be read on the densitometer and used for later
reference. You can also read the densities of each level of printer lights to see
where reciprocity sets in, although this is not actually necessary because this
density will probably be deeper than you will actually be using to print.
A test for no-density print highlights can also be done at this time by print-
ing a piece of opaque leader at the determined printer lights and reading the
resulting density. The difference between your chosen black density and the
resulting white density will determine the dynamic range of the print stock.
In order to determine the speed and working range of your negative in rela-
tion to that print stock, further testing is necessary.
You should now have an optimum black density and a reference to the
intensity of the printer lights that will be required at your lab to achieve that
density with your chosen negative stock, as well as any unusual processing
methods and any variation in printing techniques that you choose to use.
This brings us to the third part of the test.
ExPosurE caliBraTion
This will be the first camera test that will provide the working speed or
exposure index (EI) that will allow you to judge the exposure necessary to
represent the values photographed as normal tones on the final print, when
| 301
Emulsion Testing
that print is made using the recommended density arrived at during the first
two parts of these tests. You must determine the amount of light that it will
require to properly photograph a midgray tone when the negative is printed
to the benchmark density.
There are several things I would like to mention at this point about testing
methods. Everybody has their own method of measuring light values. There
are probably as many methods as there are people taking exposure readings.
If your meter and method of reading works for you, it is correct.
I prefer to use a spot meter and take my neutral readings off of a Per-
manent Gray Card. I feel that this gives me a consistent and accurate way
of judging not only the light falling on a subject but the reflectance of that
subject as well. For these tests I also like to vary the amount of light falling
on the subject rather than changing the T-stop on the lens. I feel that this
method gives me a more accurate series of exposures because there is no
reliable way to vary the stop by fractions due to the variables and tolerances
of the lens iris. As you see later extreme exposures will be needed. In these
cases I may vary exposure speed but never T-stop.
The lighting for these tests require flat, even illumination over the surface
of the subject like copy light (light from two sides of the subject, at a 45°
angle from the camera). The color temperature of the light should be as close
to 3200° Kelvin as you can get, except in tests of daylight film, when 5500°
Kelvin should be used.
If you are planning to use filtration (such as diffusion of some kind), these
filters should be used in all subsequent tests. This is because some of these
filters can tend to absorb light. Even though this effect will be very slight, it
can affect the results of your tests by as much as 2⁄3 of a stop when you use
heavy filtration.
Make a series of exposures of an 8" x 10" gray card and a face with neu-
tral skin tone at a series of stops based on variations in the manufacturer’s
recommended exposure index. Start the series at one stop under the EI and
decrease the exposure by ⅓ of a stop until you reach one stop over the rec-
ommended speed.
For instance, if you were testing an emulsion with a recommended speed
of 500, you would start your test at an EI of 1000 and proceed to an EI of 250
in ⅓-stop increments, resulting in seven different exposures.
Remember, don’t vary the T-stop. Change the amount of light to give the
proper exposure at the T-stop you are using.
Print the negative at the benchmark density arrived at in the second part
of the test, adjusting the printer ratio (color balance) to reproduce a neutral
gray. Read the print density of the gray in each exposure. A proper mid-
gray print density for theater viewing should be R/1.09, G/1.06 and B/1.03
(status A filters).
302
|
Emulsion Testing
LABORATORY AIM DENSITY

RED GREEN BLUE
Color Negative .80 1.20 1.60 0
2383/2393 Color Print 1.09 1.06 1.03 0
5242 Master Positive 1.15 1.60 1.70 ± 0.10 Aim
2242 Dupe Negative 1.0 1.45 1.55 ± 0.10 Aim
View the print to determine which print is closest to that recommended

density. Look carefully at the quality of the color balance of the skin tones
in relation to the gray card. If an emulsion cannot reproduce skin tones
properly when the gray card is printed correctly (or vice versa), this is a
good indication that there are problems with either the emulsion or the lab
processes that have taken place. If this is the case when the skin tones are
printed properly in the final print, there will always be problems getting the
proper color balance in the shadows.
The print that is chosen as the best representation of the gray card and skin
tone will become the midpoint in the dynamic range of your negative. Check
which exposure index was used for this test. This EI will become your empiri-
cal emulsion speed. Most often I have found that the EI that is derived will be
within ⅓ of a stop of the manufacture’s recommended speed unless some form
of processing modification is used (such as push- or pull-processing).
TEsTing for dynamic rangE

This is the part of the testing process that will determine the usable dy-
namic range of your negative when exposed, processed and printed using
the information gathered in the previous tests.
Make a series of exposures using a Macbeth Color Checker color chart,
an 8" x 10" gray card, a small gray scale and a face with neutral skin tone.
Mount the color chart vertically with the gray card in the middle and the
scale vertically next to the gray card all on one piece of card. Mount this
card on a grip stand and place it over the head of the model. This allows
you to fill the frame with the cards and then tilt down to see the face. Shoot
the chart and the face each for a minimum of 10 seconds (more if you can
afford the film) so that you will have enough time to study the results on
the screen. If you are comparing emulsions or processing techniques, repeat
these tests for each variation.
Using the EI that you derived from the last test, start the series of ex-
posures at normal and underexpose successively until you reach five stops
underexposed. Do the same with overexposure.
| 303
Emulsion Testing
For example:
First Series Second Series
Normal Normal
1 Stop Under 1 Stop Over
1⅓ Stops Under 2 Stops Over
1⅔ Stops Under 3 Stops Over
2 Stops Under 3⅓Stops Over
2⅓ Stops Under 3⅔ Stops Over
2⅔ Stops Under 4 Stops Over
3 Stops Under 4⅓ Stops Over
3⅓ Stops Under 5 Stops Over
4 Stops Under
5 Stops Under
The reason for the uneven increments of exposure is based on experi-

ence. I know that the first shadow detail will fall somewhere in the range
of 2 and 3 stops underexposed and that the last highlight detail will fall
between 4 and 5 stops over. I also know from experience that the incre-
ments between 1 and 2 are very useful shadow densities to have as a visual
reference.
Print these tests again at the benchmark densities. View the work print
to make sure the color ratios are correct. At this point, if possible, an in-
terpositive, dupe negative and final print should be produced using any
special printing techniques or digital manipulation intended for the final
release (such as ENR, flashing the interpositive, or scanning). This will
allow you to view the results as they would be viewed in the theater. If this
is not possible, enough useful information can be gleaned by viewing the
work print.
When you view the results projected, either in motion or on strip projec-
tors, you will begin to see the effects of exposure on different tones and
colors. If you are comparing different emulsions or processing techniques,
the results should be viewed side by side for proper comparison.
The exposure difference between first shadow detail and last highlight
detail and their relation to midgray will determine the empirical dynamic
range of the negative, processing and printing combination.
conclusion:
TrusT your EyE morE Than your mETEr
It is important to remember that these tests are not scientific but em-
pirical. They are meant to train your eye to the dynamic range of your
emulsion under working conditions. The tests should be a good working
reference. In fact, I have often taken frames of each exposure and mounted
304
|
Emulsion Testing
them in slide mounts for viewing on the set if I want to know exactly where
to place a specific tone on the scale so that it will be represented exactly as
I want in the final print. To do this, you will need a small light box properly
color-corrected and with an illumination of 425 FC ±10%.
It is most important to learn to trust your eye rather than relying upon too
many exposure readings. These tests should give you a better understand-
ing of the results of exposing, processing and printing your original camera
negative so that you can predict exactly what the images you make will look
like. With this knowledge, you should be able to make more consistent, dra-
matic images that will help tell the story of your motion picture.
Steven Poster, ASC is President of the International Cinematographers Guild

and a vice chair of the ASC’s Technology Committee. He is a current member of
the committee for the Nicholl Fellowship of the Academy of Motion Picture Arts
and Sciences and a former member of the Executive Board of the International
Documentary Association.
| 305
Finding Your Own
Printer Light
by Richard Crudo, ASC
A mong the numerous tests to be performed before starting a project des-

tined for theatrical presentation, perhaps the most valuable one is the test
that establishes a single printer light setting to be used in dailies printing. Al-
though not connected to lighting in the literal sense, a direct relation does exist
in that a consistent printer light precisely determines how lighting will be ren-
dered on screen, and by extension influences the final release print’s texture.
The use of a single, predetermined set of numbers takes the guesswork out of
the laboratory’s dailies printing protocol and places total control of the film’s
look where it belongs—in the hands of the cinematographer.
Each night, dailies timers scroll an incredible amount of negative through
their color analyzers. These tens of thousands of feet are culled from a wide
range of productions shot under a variety of circumstances. Very often, their
only guidance in determining what information gets sent to the printer is an
unintelligible scrawl at the bottom of a camera report—print cool, print warm,
day-for-night, dawn effect and so on. While lab technicians are pretty good at
drawing meaning from the indefinable, when using a single printer light set-
ting—arrived at through one’s own choosing—the problems caused by relying
on vague, highly personal and subjective written or verbal descriptions are
eliminated. In addition, working in this manner allows the cinematographer
to introduce any amount of variation in color and density to the image in a
quantifiable and repeatable way. Whether these changes are effected through
filtration used on the lens, gels over the lamps or by ordering a measured shift in
the printer light itself is a matter of taste and experience. Then, when a deviation
from the norm occurs or a new batch of raw stock brings a shift in the film’s
look, the anomaly becomes easy to isolate and correct. Assuming the lab’s chem-
istry is up to code, the immediate payoff is day-to-day image consistency on
par with the still photographer’s vaunted “previsualization.” For the long term,
answer and release printing are much simpler affairs, since making corrections
is a matter of fine-tuning rather than a complete rebalancing of the entire film.
PreParatiOns FOr an emuLsiOn test tO

estabLish a singLe Printer Light (tungsten)
First, understand that the one-light print enabled by this test is markedly
different in concept from what dailies personnel commonly refer to in the
306
|
Finding Your Own Printer Light
same words. Left to its own devices, any lab can deliver a one-light print
every day. The difference is that since the dailies timer is making color and
density choices on his or her own, the lab’s version will inevitably change its
R-G-B Hazletine values from negative roll to negative roll or even among
different shots and setups within each roll. The printer light that ultimately
results from the following procedure is something chosen by the cinematog-
rapher, and is meant for use in all situations matching the lighting conditions
under which the test is performed. Thus, it is possible—indeed preferable—
to shoot an entire feature film on the same printer light. That said, it is also
viable to establish printer lights for defined situations, (i.e., day/exterior,
day/interior, night/exterior, night/interior, etc.).
Preliminaries
A. Make sure that camera body and lens tests are already completed and
approved.
B. Secure a pair of showcards flat against a wall, end-to-end in a horizontal
fashion. The black side of one of the cards should be placed on the left, the
white side of the second card on the right.
C. Recruit a model with a representative fleshtone. The term “representative”
is somewhat ambiguous in its use here, but that is part of what this test is
trying to determine. Avoid extreme complexions or coloring of any kind
unless you anticipate dealing with them during principal photography.
Also, make sure that your model dresses in neutral shades rather than
solid blacks or whites or any heavily saturated colors.
D. Using a lens from your tested, matched production set, compose the shot
so that the two showcards completely fill the frame. Place your model
directly facing the lens, just slightly in front of the center seam where
the two showcards meet. In 1.85 format, a 40mm lens works well at a
distance of about eight feet and renders a pleasingly medium close-up of
the model.
E. Create a series of individual flash cards that the model will hold within
the frame to clearly indicate the exposure information relevant to each
take:
NORMAL +1½ -½ -2
+½ +2 -1 NORMAL
+1 NORMAL -1½
F. In addition to a gray card, the following information should be mounted
in a plainly readable fashion somewhere within the frame but not in such
a way as to impede sight of the model:
• emulsion type and batch/cut numbers
• ASA/Exposure Index
• lens focal length
| 307
• T-stop
• development information (normal, push, pull, ENR, CCE, etc.)
• print stock type
• optional: color temperature of the light source you’re using
Lighting
A 2K fresnel is an ideal unit to use with this test. Placed at an angle of
about 20 degrees off camera right, make sure the light is evenly spread at
full flood across both the model and the two showcards—with no hot spots
or dropoff of any kind. From this position, the lamp serves a dual purpose
by not only properly illuminating the model but by throwing the model’s
shadow onto the black-sided showcard that covers the left half of frame. The
deep, rich, velvety darkness this provides will serve as an important point of
reference when judging the projected print.
Do not use any diffusion on the lamp and do not add any light to the fill side.
asa/exposure index
Since film speed is a relative concept, the best starting point is to rate the
negative at the manufacturer’s suggested value. Besides providing the infor-
mation necessary to choose a single printer light for the run of the show, this
test will also allow the setting of an effective ASA/EI rating for the manner
in which the film is to be exposed.
t-stop
In the interest of contrast uniformity and the elimination of as many
variables as possible, lock the iris ring at a predetermined T-stop and leave
it alone for the length of the test. It should rest precisely at or close to the
primary setting intended for use across much of the shoot. Measured expo-
sure shifts will be carried out through adjustments in lighting intensity and
a combination of neutral-density filters and shutter-angle changes. (For the
purpose of this article, the working stop for the test will be T2.8.)
Filters
If plans for principal photography include the use of filters in front of or
behind the lens, slip the appropriate grade into the matte box or filter slot
before you begin the test.
Laboratory instructions
The camera report should prominently display the following orders:
◗ Develop—(normal, push, pull, ENR, CCE, etc.)
◗ Print this negative roll two times.
◗ First pass: print on best light for gray card/normal exposure only.
308
|
◗ Second pass: correct each take back to normal in ½ stop (4 points each)*
increments.
◗ Note well: normal exposures should all print at the same light in all cases.
◗ Do not join these rolls together or to any other roll.
By keeping the two test rolls separate, you will be able to view them
in rapid succession without having to deal with any other distracting
material.
Basically, the one-light printing of the gray card and normal takes—
and thus the timing of the entire first roll—allows the dailies timer a
fighting chance at showing his or her interpretation of what will look
best onscreen with respect to the lab’s processing standards and the con-
ditions set up by the cinematographer. The second uncorrected printing
pass insures that you will see the effect over- and underexposure will
have on the emulsion in its purest state—without any assistance or aug-
mentation from the lab. Examining both rolls together essentially de-
fines a place from which the lab timer and cinematographer can begin
to deviate.
Note: For purposes of this test, it is given that the laboratory’s system
is calibrated so that 8 printer points equal one T-stop on the lens.
miscellaneous
◗ Beware of ambient light or anything else that might compromise the
test’s integrity.
◗ Be meticulous with meter readings. If you choose an iris setting of T2.8,
your normal exposure should read precisely T2.8 at the model’s face.
Measuring the increase in light level needed to support the overexpo-
sure parts of the test should be handled with equal care.
◗ Do a separate test for each emulsion you plan to use and each lighting
condition you plan on encountering.
◗ Be sure the model clearly displays the placards indicating the proper
exposure for the take being photographed.
◗ Don’t rush.
the test
First, fill the frame with a gray card. Light it to T2.8 and expose 20 feet at
the same value.
Next, recompose to fill the frame with the black and white showcards,
featuring the model at the center seam.
Following the notations in each column, expose 30 feet for each step as
noted:
(Note that after the first normal exposure the light level increases from
T2.8 to T5.6. This is done to facilitate the overexposure takes. The standard
| 309
iris setting here is T2.8, so before starting the test, simply light the model to
T5.6 and then use two double scrims on the 2K Fresnel to knock down the
intensity to T2.8 when needed.)
If overexposure is to be carried as far as +3 stops, the basic light level must
be increased to T8 to accommodate that portion of the test. Proportional
changes should then be made to the scrims and shutter angle/neutral density
filter combinations.
the results
When viewing the projected film, refer to the lab’s printer-light notation
sheet that corresponds to the test exposures.
You should speak to your lab contact as to what is considered a “normal”
printing light for the lab you are using; however, we will assume for this
article that a “normal” printer light would read 25-25-25. Roll 1 will now
obviously play all the way through at light 25-25-25. Any exposure changes
noted on screen will thus be a direct result of what was done at the lens. This
pass is especially helpful in gauging color drift as it relates to exposure. It
is also a good indicator of the emulsion’s ability to hold detail at the noted
extremes.
Roll 2 is merely a second printing pass of the same negative but with the
identical series of exposures corrected back to normal in measured incre-
ments by the Hazletine timer. Based on the concept of 25 across being nor-
mal, refer to the following boxed chart for the progression of printer lights
(assuming a laboratory printing scale of 8 points = 1 stop).
In this instance, the dailies timer has “helped out” by correcting “mistakes”
in exposure. The second pass thus provides an idea of how far the emulsion
can be stretched before it falls apart. Special attention should be paid to the
white and black showcards that make up the background behind the model.
Grain, color shift and variation in detail will be most readily apparent in
these areas. This pass can also provide information about contrast. If neces-
sary, it may be requested that the laboratory print “over scale” in order to
achieve correction on an overexposure.
Conclusion
Now that the data needed to decide all critical concerns has been revealed,
decisions that will directly affect the film’s look can be made in an informed
manner. Subjective judgement once again comes into play, but the difference
is that the cinematographer is the one doing the judging.
Usually, the model’s fleshtone will need some tweaking regardless of which
test exposure is most pleasing. Let’s say that Eastman 5213 was used at its
recommended ASA/EI rating of 200. While viewing results of the corrected
Roll 2 on screen, it is decided that the grain structure and shadow detail
310
|
of the take indicating ½ stop overexposure (printed back to normal) looks

best. By referencing the lab’s printer light notation sheet, this would render a
printer light of 29-29-29 to start and therefore an effective ASA/EI rating of
160. A desire for a different sort of fleshtone might lead the cinematographer
to order a small adjustment in the addition of perhaps 1 point red and 2
points yellow. The resulting printer light of 28-29-31 would then be the one
to use during principal photography. To verify the effect, it would be advis-
able to shoot an additional test under identical conditions with the same
model, while printing it at these new numbers.
Hereafter, by having the assistant cameraperson stamp 28-29-31 in the
camera report’s printing instructions box, the cinematographer can be cer-
tain of two things. Besides meeting a specific standard that depends solely
on the effort put into each shot, such items as silhouette effects will indeed
come back from the lab as silhouettes—each and every time. Instead of
communicating with such agonizing vagaries as “print for highlights,” this
simple set of numbers conveys to the dailies timer exactly what is needed in
a way that will stand up in court. This isn’t to say, however, that the printer
lights are by any means sacrosanct. Over the long haul of shooting a feature,
modifications are inevitable.
Ultimately, however, what is most important is that the cinematographer
is always the one who chooses how and when to do the modifying.
Richard Crudo, ASC currently serves on the Academy of Motion Picture Arts and
Sciences Board of Governors, and is a former ASC President and current ASC Vice
President. He has photographed the feature films Federal Hill, American Buffalo,
American Pie and Down to Earth.
| 311
Adjusting Printer Lights
to Match a Sample Clip
by Bill Taylor, ASC
H
filters.
ere’s a low-tech but usefully accurate method for modifying printer
lights “in the field” to match the color of a film clip, using CC Wratten
You’ll need a light box with a color temperature around 5400°K, a card-
board mask with two closely-spaced, side-by-side frame-size holes cut into
it and a set of color correction filters of .05, .10, .20, .30, .40, and .50 values
in Yellow, Cyan, Magenta, and Neutral Density. (Kodak Wratten CC filters
are expensive and fragile to use as viewing filters, however they will last
a long time if handled carefully. PC filters retired from darkroom or still
photo use are just fine. You may be able to find less-expensive plastic sub-
stitutes.) You’ll also need a loupe with a field big enough to see at least part
of both frames.
Put the mask on the light box, put the sample print film clip (the one
to match) over the left-hand mask aperture. Put the target print (to be
reprinted) over the right-hand aperture. (The mask keeps your eyes from
being dazzled by white light leaking around the film frame.)
It’s a two-step process.
MAtCh overALL brightneSS

Looking at the midtones of the two prints, add ND filters under one clip
or the other until the brightness is an approximate match. It’s much easier if
there’s a similar subject in both frames!
CoLor AdjuStMent
Estimate the color shift needed to bring the color of the target clip around
to match the sample. For example if the sample is a too blue, add yellow
filters. If it’s too cyan, add yellow and magenta. If the target clip now looks
too dark with the color filters added, remove some ND from the target (or
add it to the sample). With practice, you’ll be able to get a reasonable match.
From time to time, look away for a moment at a neutral color so your eyes
will not get tired.
Now add up the color filters (ignore the ND’s for now). Let’s say you
added a total CC .15 magenta and .30 yellow to the target print. Consult the
table below for the correction needed in printer points.
312
|
Adjusting Printer Lights
AdjuSting the Printer LightS

Make the first correction of the printing lights, adjusting color only. The
printer lights are usually listed in Cyan, Magenta, Yellow order on the light
card, or the corresponding printer filters Red, Green and Blue, respectively.
But don’t think about Red, Green and Blue when you’re dealing with color
print! The layers in color print produce Cyan, Magenta and Yellow dye, so
if you think in those terms, you can’t get confused.
Let’s say the hypothetical target print, which needs 2 printer points of
Magenta and 4 points of Yellow to bring it around, originally printed on C
25, M 22 and Y 34.
Adding 2 points of Magenta and 4 points of yellow yields an adjusted
printer light of C 25, M 24 and Y 38.
Now let’s get back to the ND’s. Assume you wound up with .30 ND over
the target print, which was too light as well as needing the color correction.
Per the table, you should add 4 printer points across. So the new, final print-
ing light is C 29, M 28, and Y 42.
If you had to lay the ND filters over the sample print instead, that meant
that the target print was too dark. So subtract the neutral printer point ad-
justment instead of adding. If it needs to be .30 daarker, subtract 4 printer
points across, and so forth.
This method requires mid-tones in both clips. If the target clip is mostly
highlights (on the “toe” of the print curve), it will take more correction to
“move” them to where you want. Obviously a burned-out specular highlight
on the print will not change at all from a modest printer light correction.
A very dark scene is also hard to correct. Adjust the lightest detail you can
find, preferably a skin tone.
With experience the filter adjustments will become second nature easy
and the arithmetic automatic; you’ll be calling lights like a film lab timer.
PRINTER POINTS = CC CHANGE

1 = .10 5 = .35
2 = .15 5 = .40
3 = .20 6 = .45
3 = .25 7 = .50
4 = .30 8 = .60
Why it works: color print stock has a gamma of about 3, meaning, as an exmple,
that a CC .10 (log E) change in exposure produces a change of .30 density. Each
printer point equals CC .025 change in exposure, so it takes 4 printer points to produce
that .30 change. The table shows calculations to the nearest whole printer point.
| 313
Cinemagic of the
Optical Printer
by Linwood G. Dunn, ASC
Lin Dunn, the author of this article, is one of the motion-picture industry’s most
accomplished pioneers. Involved with the design and origins of the first optical
printers, Dunn remained vibrantly active in the industry throughout his life and
applauded the introduction of CGI techniques that, in many ways, obviated many
of the needs for his optical printer. Dunn is responsible for the effects you can
and can’t see in Citizen Kane, and literally hundreds of film projects after that.
He was always on the cutting edge, and in his nineties founded of one of the first
companies to develop digital cinema, Real Image Digital. Dunn’s passing in May
1998 was a loss to all who were fortunate enough to be on the receiving end of his
generous knowledge and skill; his was also the loss of one of our remaining links
to Hollywood’s earliest days. Winner of the Gordon E. Sawyer Academy Award
from the Motion Picture Academy of Arts and Sciences, and a past President of
the ASC, Dunn penned an article that is as relevant today as when he first wrote
it. — Editor
T he earliest optical printers were custom-built by the major studios and

film laboratories, and were usually designed and made in their own
shops to fit their particular requirements. Modern standardized optical
printing equipment, capable of creating the innumerable effects heretofore
possible only in the major studios, became available to the entire motion
picture industry in 1943 with the introduction of the Acme-Dunn Optical
Printer, designed and built for the United States Armed Forces Photographic
Units. Later the Oxberry, Producers Service, Research Products and other
optical printers appeared on the market. Commercial availability of this type
of equipment greatly stimulated and widened the scope of the special-effects
field. Even the smallest film producers could now make motion pictures
with special effects limited only by their imagination and budgets, utilizing
the services of growing numbers of independent special-effects laboratories
which could now operate competitively using equipment available to all.
Developments over the years of more sophisticated equipment, new
duplicating films, special-purpose lenses and improved film-processing
techniques, as well as skilled technicians, have increased the use of the
optical printer to a point where its great creative and economic value is
common knowledge in the motion-picture industry. In more recent years,
314
|
Cinemagic of the Optical Printer
the adaptation of computer technology to the optical-effects printer has

basically simplified the control and accuracy of some of its important func-
tions, thus making it much easier to produce certain complex visual effects
at lower cost, as well as to greatly expand its creative scope. This has made
it possible to program, record and repeat the movement of certain of its
devices with such a degree of accuracy that area-blocking functions can
now produce traveling matte composite scenes heretofore highly impracti-
cal, if not impossible. One can truly say that the creative capability of the
modern visual-effects optical printer is only limited by the creative talent
and technical skills of the operator. In recent years, such major film produc-
tions as Star Wars, The Black Hole, The Empire Strikes Back and Cocoon
have all utilized the full capabilities of the modern optical printer to create
a whole new world of imaginative creativity through their extensive use of
very sophisticated motion-picture visual effects. The following list of some
of the work that is done on the modern optical printer will illustrate its vast
scope and tremendous importance to modern filmmaking.
TransiTiOnal EffECTs
Employed to create a definite change in time or location between scenes.
The fade, lap dissolve, wipe-off, push-off, ripple dissolve, out-of-focus or
diffusion dissolve, flip-over, page turn, zoom dissolve, spin-in and out, and
an unlimited variety of film matte wipe effects are all typical examples of the
many optical transitional effects possible.
ChangE Of sizE Or POsiTiOn

May be used to eliminate unwanted areas, obtain closer angles for extra
editing cuts, reposition action for multiple-exposure framing, including
montages and backgrounds for titles.
framE sEquEnCE mOdifiCaTiOn

Screen action may be sped up or slowed down in order to: convert old 16
fps silent films to standard 24 fps sound speed; change speed of action and
length of certain scenes or sections of scenes; provide spot frame modifi-
cation to give realism to specific action in fights, falls, chases, etc.; hold a
specific frame for freeze effects and for title backgrounds; add footage for
comedy effects; reverse direction of printing to lengthen action and for spe-
cial effects use; extend scenes through multiple-frame printing for action
analysis in instrumentation, training and educational films.
OPTiCal zOOm
Optical zoom is used to change frame-area coverage and image size during
forward and reverse zooming action in order to: produce a dramatic or impact
| 315
effect (according to speed of the move); counteract or add to the speed and
motion of camera zooms or dolly shots; reframe by enlargement and/or add
footage to either end of camera zooms or dolly shots by extending the range
of moves; momentarily eliminate unwanted areas or objects by zooming for-
ward and back at specific footage points (such as when a microphone or lamp
is accidentally framed in during part of a scene); add optical zoom to static
scene to match camera zoom or dolly in a superimposure. The out-of-focus
zoom also is effective to depict delirium, blindness, retrospect, transition, etc.
suPErimPOsurE
Superimposure is the capability used to print an image from one or more
films overlaid on one film. This is commonly done in positioning title let-
tering over backgrounds. Also used for montages, visionary effects, bas
relief; adding snow, rain, fog, fire, clouds, lightning flashes, sparks, water
reflections and a myriad of other light effects.
sPliT sCrEEn
Employed for multiple image, montage effects, dual roles played by one
actor, for dangerous animals shown appearing in the same scene with peo-
ple (as in Bringing Up Baby, which shows Katharine Hepburn working with
a leopard throughout the picture), where such split screens move with the
action. Matte paintings often utilize this technique when live-action areas
require manipulation within an involved composite scene.
qualiTy maniPulaTiOn
The quality of a scene, or an area within a scene, may be altered in order
to create an entirely new scene or special effect or to match it in with other
scenes. There are innumerable ways to accomplish this, such as adding or
reducing diffusion, filtering, matting and dodging areas, and altering con-
trast. Often library stock material must be modified to fill certain needs,
such as creating night scenes from day; reproducing black-and-white on
color film through filtering, printed masks or appropriately coloring certain
areas through localized filtering; and the combining of certain areas of two
or more scenes to obtain a new scene, such as the water from one scene and
the terrain or clouded sky of another.
adding mOTiOn
Employed to create the effect of spinning or rotating, as in plane and auto
interiors and in certain montage effects; rocking motion for boat action,
sudden jarring or shaking the scene for explosion and earthquake effects;
distortion in motion through special lenses for drunk, delirious and vision-
ary effects.
316
|
gEnEral usEs Of ThE OPTiCal PrinTEr

The preceding represent some of the special categories of effects that can
be produced on the optical printer. The following are a few of the more
important general uses employing this useful cinematic tool.
Traveling mattes
Used to matte a foreground action into a background film made at another
time. The various matte systems in use today require the optical printer in
order to properly manipulate the separate films to obtain a realistic-quality
matching balance between them when combined into a composite. Use of
this process has greatly increased as modern techniques produce improved
results at reduced costs. Motion control, referred to earlier, has greatly wid-
ened the scope of this visual-effects category.
Blow-ups and reductions

The fixed set-up optical printer is used for 16mm reduction negatives and
prints, and for certain limited release printing from 35mm originals. This
is utilized when small volume makes this procedure more economical than
through a converted negative, and when maximum quality is of greatest
importance. Enlarging from 16mm to 35mm color or black-and-white is a
very important function of the optical printer. Many fine theatrical films,
such as the Academy Award-winning The Sea Around Us, The Living Desert
and Scenes From a Marriage, have been photographed in 16mm, and have
enjoyed great financial success through 35mm release prints made from
35mm blow-up internegatives.
Special new lenses, raw stocks and immersed-movement printing have
enhanced the overall quality to a point where the 16mm–35mm blow-up
medium is presently enjoying very successful commercial usage. Conver-
sions between 65mm and 35mm are also an important function of the
optical printer. Productions made in almost any film format are being
release-printed in different types to meet certain theatrical distribution
requirements. The Concert for Bangladesh was the first feature-length film
to be enlarged from 16mm color internegative directly to 70mm theater
prints.
anamorphic Conversions
The standard optical printer equipped with a specially designed “squeeze”
or “unsqueeze” lens can be used to produce anamorphic prints from “flat”
images, or the function reversed. The possibility of the “flat” or spheri-
cal film being converted for anamorphic projection without serious loss
of quality has greatly widened this field of theatrical exhibition. The ma-
nipulations available on the optical printer also make it possible to scan and
| 317
reposition any scenes that require reframing when converted to or from

widescreen proportion.
dOCTOring, mOdifying and salvaging

Some of the important uses of the optical printer are not recognized as
special effects in the finished film, and often are not apparent as such even
to skilled motion-picture technicians. One of these applications is the field
of “doctoring” by modifying scenes that, for a variety of reasons, may not be
acceptable for use. This includes salvaging scenes that are completely unus-
able due to some mechanical failure or human error during photography;
and also the modification of stock film material through the various meth-
ods noted to fit specific requirements. Many expensive retakes have been
avoided by the ingenious application of such optical-printing reclamation
techniques. The liquid, or immersion, film gate produces dramatic results
in the removal of scratches.
Citizen Kane is an excellent example of scene modifications created on the
optical printer during the postproduction period. New ideas were applied
to existing production scenes for which new supplementary scenes were
photographed and integrated to enhance and create various new concepts.
In It’s A Mad, Mad, Mad, Mad World, an important scene was photo-
graphed in which a truck was supposed to back into a shack and knock it
over. The breakaway shack was rigged to collapse when wires were pulled
on cue. Signals became crossed, and the shack was pulled down well be-
fore the truck touched it. A very costly retake was indicated, so the optical
printer was called to the rescue. The task of correcting the error through a
split screen seemed relatively simple until it was discovered that the camera
panned with the falling shack. It then became necessary to plot and move
the split matching point frame-by-frame on the optical printer to follow
the pan. Through this traveling split-screen technique, the progress of the
shack’s falling action was delayed until the truck had reached the point
of impact. Perhaps the entire cost of the optical printer was saved by this
salvaging job alone. Such clever techniques have been used many times to
bring explosions close to people working in a scene, such as in One Minute
to Zero, where a line of so-called refugees were “blown to bits” by artillery
shelling. Split screens in motion and trick cuts, with superimposed smoke
and flame, did the job in a most effective manner.
nEw sysTEms
The optical printer is being used to develop new horizons in the creation
of special camera moves within an oversized aperture. This is particularly
effective in the creation of camera movement in a composite scene, such as
one involving a matte painting, thereby giving a greater illusion of reality.
318
|
VistaVision and various 65mm negative formats, including 15-perforation

IMAX and 8-perforation Dynavision as well as standard 5-perforation
frames, lend themselves to this technique.
Copying onto 4-perforation 35mm makes possible spectacular pans,
zooms, dolly shots, etc., without sacrificing screen quality, and with full
control over such movements, all of which are created on the optical printer
in the internegative stage and made during the postproduction period. Use
of this technique makes it possible to avoid time-consuming and compli-
cated setups during production, with the added advantage of flexibility in
later change of ideas.
Probably the most exciting new optical-printing development has been
in the field of electronics. The adaptation of video image transfer through
sophisticated high-resolution scanning systems, in conjunction with the
new developments in cathode-ray tubes, lenses, film-moving mechanisms,
special-purpose film stocks and the latest research in electronic image com-
positing, have opened up exciting new vistas in special visual effects. The
modification of filmed color motion-picture images through computerized
electronic transfer back to film is making it possible to create photographic
effects on film or tape faster, more economically, and with a scope of cre-
ativity heretofore not possible. The ability to easily and quickly transfer
areas or moving objects from one film to another through their instanta-
neous electronic isolation and self-matting will be of tremendous economic
benefit in this area of film production, as well as in stimulating creativity in
the wider use of special effects.
| 319
Motion-Control
Cinematography
by Richard Edlund, ASC
I n the 1970s the modern technique of motion-control liberated the long

locked-off effects camera from its tripod, giving it freedom to move, be-
coming an inseparable part of film grammar. Digital image tracking has
pulled out the remaining stops, giving complete freedom to the director of
photography who wants no restriction of camera movement.
Prior to the advent of digital technology, repetitve control of motion had
been attempted with various degrees of success by using analog wire record-
ers, Selsyn motors and gears, even by hand-cranking mechanisms using a
metronome for synchronization. Finally, solid-state digital electronics made
it possible to accurately record and play back motion with sufficient reliabil-
ity to achieve the robotic camera systems necessary to produce the dog-fight
and space sequences in Star Wars. How appropriate that this technology was
born of the space race in the late 1960s. The phenomenal success of Star
Wars brought on the modern renaissance of motion-picture visual effects,
and since that time, a clear majority of the top boxoffice grossers are “effects
films” that have relied on motion-control and image tracking for crucial
scenes.
To define it, motion control comprises an electromechanical system that
allows for the physical motions of a camera and/or other objects—such as
models mounted on motorized gimbals, tracks, etc., to be digitally recorded,
enabling successive passes to be photographed “on the fly” with the cor-
responding motion blur characteristics of normal motion picture cameras.
Thus, in combination with traveling matte systems of varying kinds, fore-
grounds and backgrounds of differing scales can be combined using a mov-
ing camera. Composites became possible that could combine separately
photographed actors, miniatures, lighting effects, background paintings,
pyrotechnics, and myriad other creative possibilities. With the use of syn-
chronized video playback systems, actors could perform within virtual or
separately photographed sets and locations, or can interact with themselves,
other performers, creatures or miniatures shot previously or subsequently.
Several companies have developed hybrid field motion-control systems.
These incorporate specially built or modified gearheads and dollies, which
have differing and various facilities, such as panning, tilting, booming,
follow-focus, remote operation, preprogrammability, ease of set-up, quiet-
320
|
Motion-Control Cinematography
ness of operation for sound, high tracking speed, long distance of travel, and
adaptability to various production cameras and visual effects cameras.
A visual effects supervisor who is familiar with this paraphernalia should
be available to collaborate with the director, director of photography, first
assistant director, production designer, and other appropriate crew mem-
bers to achieve the proper set-up for any given plate. Of course there is
responsibility to achieve any given plate within reasonable and predictable
set-up time, and for this reason careful preproduction planning is neces-
sary between the contracted visual effects facility and the unit production
manager. When shooting bluescreen or greenscreen scenes involving actors
within the principal production schedule, the wardrobe should be discussed
with the visual effects supervisor to make an attempt to avoid certain colors
that might cause matting problems in postproduction.
Motion-Control ExtEnds
Cinematic Capabilities
Motion-control systems are useful in many ways for visual effects. The
following list is certainly not exhaustive:
1. The ability to program model shots so that the motion of objects in an
effects scene is believable, and to preview these moves and modify them
as needed for approval.
2. The ability to repeat these scenes for front-light/back-light or front-light/
front-light matte passes if needed.
3. The ability to repeat these scenes for enhancement effects such as engine
passes, running lights, smoke-room effects, filtration, etc.
4. Precision flyby and extremely close approaches to objects can be accom-
plished smoothly and in perfect (programmable) focus.
5. Stop-motion and other forms of animation can be subsequently included
in live-action scenes that have field recorded moving camera data.
6. Massive rig removals on stage or on location can be easily accomplished
with moving split-screen techniques by shooting a clean pass of the back-
ground once the director is satisfied with any given take.
Motion tracking Further Extends Cinematic Capabilities

Modern digital tracking systems have emerged which enable the use of a
normal (nonmotion-control) production camera on set, and even Steadicam,
which enable the precise camera moves to be mathematically recaptured in a
digital post production environment. This technique, while not inexpensive,
offers much greater creative freedom on set and saves the production time
cost of setting up motion-control cameras, gearheads, dollies, etc. It is also
a technique which is in constant development by various effects houses and
equipment and software vendors, and that continues to advance. When this
| 321
technique is to be deployed, it is necessary to shoot geometrically perfect

grids, noting focus setting and object to camera distances to aid the computer
graphics technicians in discovering the precise distortion characteristics and
actual focal length of the particular taking lens that was used.
In most cases, the prospect of using such techniques requires a survey of
the set, and/or the use of precisely spaced witness points placed in the set.
On long shots, for example, red tennis balls are placed within the shot, and
will be painted out during the composite process. The exact positional re-
lationship of these markers must be recorded for use in the mathematically
intensive move regeneration process during post-production.
There are production cranes and gearheads available which have encoders
on the movement axes which enable the recording of move and position data
only for scenes that will have subsequent computer graphic enhancements.
In practice, the production company is best advised to contract with a vi-
sual effects supervisor and/or visual effects company to aid production in the
carrying out of specific shots and sequences that require these approaches.
Motion-Control Equipment: Visual-Effects studio

An ideal motion-control system for photographing miniatures consists
of a steady pin-registered camera with a motorized follow-focus system
mounted on a pan-tilt-roll gearhead (wherein the nodal point of the lens
can be situated at the vertex of all three axes) attached to a rotatable boom
arm that rides on a track of at least 50 feet in length.
Various model movers, rotators, or pylons are usually mounted on an-
other track of 20 feet or so set perpendicular to the camera track. There
are many variations on this basic theme incorporating various levels of
engineering prowess within the industry and the precision and reliability of
such systems provide the operators with different levels of creative freedom.
An electronic control system runs the motors, then stores the motion files
created by the operator. Usually stepping motors are used unless consider-
able speed or power is needed, in which case DC closed-loop servo motors
are used.
Studio motion-control equipment often has provisions to control the
camera shutter angle over a wide range in order to control the motion blur
of objects being photographed. The exposure range is from about ¼ second
to extremely long. Most systems have several ways to program moves and
any or all of the following methods may be used.
Joysticks (usually potentiometers or rotary optical shaft encoders) are
often used to manually program the motors that operate the various parts of
the system. The joystick might be used to program one motion-controllable
motorized function at a time, (i.e., pan, tilt, track, model rotator, etc.), and
as the programming proceeds, these motion files are recorded one at a
322
|
time and played back in concert until all of the functions necessary to the
particular shot are completed. The result is similar to remote controlling a
model airplane or car and making an exact record of what happened.
Using another programming technique, the joystick can be used to move
the components of the system to a series of start and end positions while
a record is made of these key positions, then subsequently commanding
the system to generate a mathematically smooth path through these points.
Much more complex methods of move generation are available using com-
puter graphics. There are many bells and whistles available which include
move-smoothing programs, even the use of a digital graphics tablet which
enables visual modification of graphically displayed motion curves; and
other specialized software ad infinitum.
The move files can be edited and modified in as many ways as there are
motion-control systems. A few commercial electronic motion-control sys-
tems are available, as well as mechanical systems, the most ubiquitous being
the Kuper system which recently received a Scientific and Engineering Oscar®.
Motion-Control Equipment: Field

If it is necessary to use motion-control in the field (defined here as outside
the walls of a visual effects studio, with the camera operating at sound speed),
there are different requirements. The director will almost always need a mov-
ing camera, and if this is to be done, much of the following will be required:
1. A steady camera, which is slavable to the motion-control electronics,
which will provide frame/shutter position accuracy in successive passes.
Though not always imperative, the camera should be silent so dialogue
can be recorded. It should have a synchronized videotap and playback
system that can be slaved to the motion-control electronics.
2. A reliable motion-controlled follow-focus system. Double-pass shots
must exactly repeat.
3. A pan-tilt motion-control gearhead, which by any of a variety of tech-
niques can provide scaleable lens entrance-pupil positions for postpro-
duction use on less than full-scale props or miniatures. This pan-tilt head
must have a remote operating console with hand-wheels and video moni-
tor, and in practice it is little different than using a remote head. Usually
such a head will use DC servo motors to provide a normal to high-speed
pan-tilt range. Again, silence of operation is often important.
4. A dolly with track, having a powerful tracking motor, and a motorized
boom, with positional encoders on both axes which enable dolly grip
control as in normal shots, or remote operation, or pre-programmed
moves. Ideally, the above equipment should be as standard as possible
in appearance and operational characteristics, and operate on quickly
deployable dolly track.
| 323
5. A motion-control electronics console, operated by a technician who can

efficiently log and store motion files shot-by-shot, and do this as invisibly
as possible to the rest of the production crew.
6. A videotap flicker-free system, which can store shots as they are made,
and play them back instantly for directorial scrutiny. It is often neces-
sary that this system be able to provide on-the-spot video composites for
comparisons of A to B scene action, and the ability to playback A while
recording B, etc. The video requirements will vary with the shot require-
ments.
7. A bookkeeping detail which can log actors’ positions, distances, etc.;
camera and track positions within the set; and other mathematical and
geographical information for use in postproduction. Again, this should
happen systematically and invisibly to the rest of the production crew.
Motion-Control technique
When working on Star Wars, we started with an empty building and had
to amass, modify and build our motion-control equipment before we could
produce any images. We had created visual ”violins” and had to learn to play
them. Fortunately, the picture hit and a large audience showed up for our
motion-control recitals. Since then, many innovations have come about in
the equipment and many excellent motion-control cinematographers have
appeared, and with many specialty techniques. In the studio there are two
main techniques for programming motion files: one is to use start and end
positions for each axis of motion and have the computer generate the moves;
the other allows the cameraperson to generate the move by joystick. It is my
opinion that the computer-generated method is superior for graphics and
animation purposes, and the human interface is best for most miniature and
model photography. If shots are created using a computer, the moves will
have mathematically perfect curves, slow-ins, slow-outs, etc., and will have
no heartbeat or verve, especially in action sequences, therefore becoming
subliminally predictable and less interesting to the audience. The human
operator is not interested in mathematical perfection, rather, they tailor the
camera move moment by moment to what is interesting in their viewfinder.
This human sense of curiosity is present in the work of a talented operator,
and it transfers to the audience.
Richard Edlund, ASC currently serves on the Academy of Motion Picture Arts
and Sciences Board of Governors and is a chairman of the Academy’s Scientific
and Technical Awards Committee as well as the Academy’s Visual Effects Branch.
He has been awarded Oscars® for visual effects in Star Wars, The Empire Strikes
Back, Raiders of the Lost Ark and Return of the Jedi.
324
|
| 325
Greenscreen and Bluescreen
Photography
by Bill Taylor ASC and Petro Vlahos
Overview
G reenscreen and bluescreen composites begin with foreground action

photographed against a plain backing of a single primary color. In a
digital postproduction process, the foreground action is combined with a
new background. The new background can be live action, digital or tradi-
tional models, artwork or animation, or any combination.
For the sake of simplicity, we’ll refer to “greenscreen” shots, with the un-
derstanding that the “screen” or “backing” can instead be blue or even red.
These composites (and others using related technology) are also called
“traveling matte” shots because they depend on creating an Alpha-channel
silhouette “mask” or “matte” image of the foreground action that changes
and travels within the frame.
The final composite is usually created in postproduction, although real-
time, full resolution on-set compositing1 is possible in HD video.
Fix it in POst
Production and post-production are not only separated in the budget,
they are too often separated by a huge cultural gulf. The driving force of pro-
duction is time, and in the name of speed it’s very easy to take shortcuts that
have a disastrous downstream impact. “Fixing in post” can cost hundreds of
thousands over the scope of a film and jam up an already tight post schedule.
Furthermore there is no “post fix” possible for some on-set mistakes. We
describe here what have proven to be the best practices, materials and equip-
ment available for this work.
For ease of reference, we’ll deal with the most-requested topic first:
HOw tO exPOse a Greenscreen sHOt, and wHy

Balancing screen (Backing) Brightness to the shooting stop
Let’s assume that the camera choices are optimal, screen materials and
lighting are ideal, and the foreground lighting matches the background
lighting perfectly. (All these topics are dealt with later in this chapter.)
1. In live composites, full-bandwidth HD video is fed to an HD Ultimatte hardware device

with a second camera, an HD deck or a digital wor kstation providing the background.
326
|
Greenscreen and Bluescreen Photography
Image Courtesy of BIll taylor asC

Shoulder
(highlights compressed)
Density (1/log opacity)
se
pon White
s
Re point
ear
Lin
Toe Green
(shadows compressed) Red
Blue
Exposure (log e)
Color Negative Response Curve (colors offset for clarity)
Figure 1. Schematic H&D curve
A common misconception is that backing brightness should be adjusted
to match the level of foreground illumination. In fact, the optimum back-
ing brightness depends only on the f-stop at which the scene is shot. Thus,
normally lit day scenes and low-key night scenes require the same backing
brightness if the appropriate f-stop is the same for both scenes. The goal is
to achieve the same blue or green density on the negative, or at the sensor, in
the backing area for every shot at any f-stop.
The ideal blue or green density is toward the upper end of the straight-line
portion of the H and D curve, but not on the shoulder of this curve, where
the values are compressed. Figure 1 shows an idealized H&D curve, a graph
which shows how the color negative responds to increasing exposure. In
film, each color record has a linear section, where density increases in direct
proportion to exposure, and a “toe” and a “shoulder” where shadows and
highlights respectively can still be distinguished but are compressed. Eight
stops of exposure range can comfortably fit on the H&D curve, a range only
recently achieved by digital cameras. The “white point” is shown for all three
records: the density of a fully exposed white shirt which still has detail.
Imagine a plume of black smoke shot against a white background. It’s a per-
fect white: the measured brightness is the same in red, green, and blue records.
The density of the smoke in the left-hand image ranges from dead black
to just a whisper. What exposure of that white backing will capture the full
range of transparencies of that smoke plume?
Obviously, it’s the best-compromise exposure that lands the white backing
at the white point toward the top of the straight line portion of the H&D
| 327
Figure 2. Normal and underexposed smoke plumes

curve in film (a white-shirt white) or a level of 90% in video, and brings
most of the dark values in the smoke up off the toe. If the backing was over-
exposed, the thin wisps would be pushed onto the shoulder and compressed
(or clipped in video) and pinched out by lens flare. If the backing was under-
exposed (reproduced as a shade of grey), detail in the darkest areas would
fall on the toe to be compressed or lost entirely.
You could make up for underexposure by boosting the image contrast
or making a levels adjustment (remapping the brightness range to a wider
gamut). As the right hand image in Figure 2 shows, boosting contrast makes
the backing white (clear) again, but tonal range is lost (the dark tones block
up), the edges of the smoke become harder, and noise is exaggerated. The ef-
fect of a levels adjustment is subtler, (and preferable), though extreme levels
adjustments betray the missing tonal values with banding or blocking.
Now imagine that instead of a white screen, we’re shooting the smoke
plume against a greenscreen and that the measured green brightness is the
same as before. What’s the best exposure for the greenscreen? Obviously,
it’s the same as before. The only difference is that the red and blue-sensitive
layers aren’t exposed.
Just like the smoke plume, greenscreen foregrounds potentially contain a
full range of transparencies, ranging from completely opaque to barely there.
Transparent subject matter can include motion blur, smoke, glassware, re-
flections in glass windows, wispy hair, gauzy cloth, and shadows.
To reproduce the full range of transparency, the greenscreen should be fully
but not overexposed. In other words, its brightness should match the green
328
|

Figure 3. White card in set, lit to shooting stop (incident reading)
component of a well-exposed white object like a white shirt; roughly defined
as the whitest white in the foreground that still has detail. (We don’t want to
expose that white shirt as top white, because we want to leave some headroom
for specular reflections, on the shoulder in film, 100% and over in video.)
settinG screen BriGHtness

Meter readings of blue- and greenscreens can be misleading. Some expo-
sure meters respond inconsistently to monochrome color, especially blue,
and some are affected by the high levels of UV coming from green and blue
tubes. A waveform monitor on a digital camera is the most accurate and
informative source of exposure information available. For film shoots, or if a
waveform monitor is not available, the most reliable method for balancing a
blue or green screen is still by eye, with the White Card Method:
white card Method for screen Balancing

1. Choose the f-stop at which the scene is to be shot. Let’s say it is f4. Position
a 90% reflectance white card in the acting area and light it to an incident
light reading2 of f4, keeping the spill light off the backing. The white card
2. Incident Light Reading is usually measured with an incident light meter. The
meter measures the light that is illuminating or falling on the subject. Incident light
meters typically add up all the sources falling on the surface of a white hemisphere,
which must therefore be shielded from the lights one does not want to include in the
measurement. Using the flat disk on the meter held parallel to a flat surface will give
a more accurate reading of the light falling on that surface than the meter’s sphere.
takes into account the angle, or geometry, of the light—what direction it is coming
from—and averages these two things together into a single reading.
| 329
is now lit to the brightest tone that still has detail (white shirt white) even
though the actual set lighting may not reach that level.
2. View the white card against the screen through a Wratten No. 99 green
filter. (Use a Wratten No. 98 blue filter for a blue backing.) In a pinch, Pri-
mary Green or Primary Blue lighting gels, folded to several thicknesses,
will serve.
3. Adjust the backing brightness so that the white card blends into the back-
ing. The circle overlay in Figure 3 shows the view through the filter. When
the edges of the card are invisible or nearly invisible, the green light com-
ing from the screen is now the same brightness as the green light compo-
nent coming from the f4 white card. (If you were to photograph the white
card now, the red, blue and green components coming from the card
would reproduce near the top of the straight line portion of the curve.
Since the greenscreen matches the brightness of the green component
coming from the white card, the green layer will also be exposed near the
top of the straight line portion of the curve, without overexposure.) The
backing will now expose properly at f4.
If it is easier to adjust set lighting than backing brightness, the procedure
can be reversed. Adjust the white card’s light until the card blends in, then
take an incident reading. Light the set to that f-stop.
The white card procedure needs to be done only once for a given screen
and illuminator setup. For the rest of the shoot, a quick spotmeter reading
will confirm the brightness and exposure needed. Once the backing bright-
ness is set, the spot meter may be calibrated for use with the appropriate
color filter to read f-stops directly: Wratten No. 99 + 2B for green; Wratten
No. 98 (or 47B + 2B) for blue. (The UV filters—built into the No. 98—en-
sure that UV from the tubes does not affect the reading.) Simply adjust the
meter’s ISO speed setting until the reading from the screen yields the target
f-stop (f4 in the example above). It’s advisable to use the same individual
meters (not just the same models) for the shoot as was used for testing.
Just as in the smoke plume example, more exposure is counterproductive;
it pinches out fine detail due to image spread and pushes the backing values
into the nonlinear range of the film or video sensor. Slight underexposure
is preferable to overexposure, but more than a stop of underexposure is
also counterproductive; the software will have to make up matte density by
boosting contrast or a levels adjustment.
LeveLs FOr diGitaL OriGinaL PHOtOGraPHy

Most of the same considerations apply as in film photography. It’s par-
ticularly important that none of the color channels be driven into highlight
nonlinearity or “clip”, allowing some headroom for specular highlights. If the
screen lighting can be adjusted independently of the set, light the screen to
330
|
Figure 4. Waveform display showing Green Figure 5. Greenscreen with white strip in
at 90% same light
an IRE video level of about 90% in the appropriate channel, as shown on the
waveform monitor.
The waveform display Figure 4 show the 4 channels of the image of a
greenscreen displayed in “parade mode”: Y or Luminance (brightness), Red,
Green and Blue. The display tells us a lot about the subject, which is a green-
screen filling the width of the frame and exposed to a level of 90% IRE. (Note
that 100% IRE represents the full scale of brightness from black to white. IRE
values are expressed as a relative percentage because a video signal can be
any amplitude.) The traces are flat in all channels, showing that the screen
was evenly illuminated across its width. Since this screen was lit with white
light, there is a substantial level of Red and Green contamination present,
and the relative amount is shown clearly. (Had it been lit with Kino Green
fluorescents, the Red and Blue values would be close to 0.)
The difference in level between Green and Blue (indicated by the white
double-headed arrow in Figure 4) is the headroom that the software exploits
to create the Alpha channel. The greater the difference, the less the software
will need to boost the contrast of the Alpha channel. The waveform monitor
makes it a simple matter to compare paints and materials,
However If the screen and the action must be lit together, a value of about
80 IRE on the waveform monitor is the maximum level at which the screen
can be exposed without clipping the whites in the foreground. Including a
strip of white card lit to the same level as the green screen, as in Figure 5, and
the corresponding waveform display Figure 6 shows why. With Green at 80,
the white card (indicated by the three white arrows in Figure 6) is at 90; more
exposure would push the white into clip. In practice a slight underexposure
is preferable to overexposure.
Too much underexposure is undesirable. In Figure 7, the Green level has
been set to IRE 45. There is now significantly less difference between Green
and Blue levels available to create the matte image in the Alpha channel.
Digital cameras like Arri Alexa and Sony F65 have even greater dynamic
range than film, with more latitude for exposure in the linear (nonclipping)
range. In these cameras the waveform monitor can be set to read the “log C”
| 331
Figure 6. Waveform display showing Green Figure 7. Waveform display showing Green
at 80%, White at 90% at 45%
transform (preferable to Rec. 709) of the camera’s “Raw” or data mode. Lack-
ing a waveform monitor, it’s a safe bet to set exposure for these cameras as
you would a film negative.
FunctiOn OF tHe BackinG – Green, BLue Or red

The purpose of the blue or green backing is to give the software an un-
ambiguous distinction between the color hues and values in the foreground
and the monochromatic backing. White and black backgrounds are used in
special circumstances to create “luminance masks” or “luma keys” but since
it is likely that similar luminance values will be found in the foreground,
these backings have limited use.
The degree to which the compositing software “sees” the backing deter-
mines the degree of transparency of the foreground in the final composite.
That transparency can be expressed as a percentage: Where the backing
value is zero in the Alpha channel image, the foreground is completely
opaque; where the backing value is 50% the foreground will be partly trans-
parent, and so forth for all values to 100%, the areas of the frame where the
foreground image is either not present or completely transparent. The goal is
to retain the foreground subjects’ edge transitions (including motion blur),
color and transparency in the final composite.
HOw tHe sOFtware creates tHe cOMPOsite

The software uses the difference between the backing color and the colors
found in the foreground to accomplish four tasks:
1. Optimally it will correct nonuniformity in the backing (the screen correc-
tion function, not available in all software packages).
2. It must create a silhouette matte (the alpha channel) of the foreground
action.
3. It must create a processed foreground in which all traces of the backing
color are suppressed (turned black or neutralized), while the foreground
color is carried through unchanged.
4. Finally, it must bring all the elements together into a believable composite.
332
|

Figure 8a & b. Orignal photography, silhouette (Alpha) matte
tHe aLPHa cHanneL

The alpha or matte channel (the channel which carries transparency infor-
mation) is a gray scale image in which the foreground image is a silhouette.
The silhouette may be imagined as black against clear (and all the values in
between) (Figure 9b) or with its values inverted, clear against black.
The alpha channel represents the difference in color between the backing
color and the colors in the foreground subject. The matte’s numerical level at
any pixel is proportional to the visibility of the backing.
the Processed Foreground

The original image contains the green backing and the foreground sub-
ject. Green is limited so that it cannot exceed red or blue, which has the
effect of turning the screen into a dark, near-neutral tone. The green back-
ing is further reduced to a black backing by subtracting a proportion of
the Alpha (matte) signal equal to the RGB content in each channel where
the green backing shows. All of the screen color seen through transparent
and translucent subjects likewise disappears. The subject appears to have
been photographed against a perfectly black backing. This is the processed
foreground image (see Figure 10a).
If the foreground subject (actor) is close to the backing or standing
on a green floor, the subject will have a green color cast due to reflected
(bounce) light from the floor and from lens flare. (This reflected light from
the screen is sometimes called “spill,” but it should not be confused with
the spill light from the subject’s lighting falling on the screen.) No attempt
should be made to remove this color with filters on the lights or camera,
or with color correction in transfer. All backing contamination is removed
from the subject by the compositing software’s white, gray and black bal-
ance controls.
Blue bounce is much harder to see on the set than green, but just as
troublesome in the composite. There is no difference between greenscreens
| 333
Figure 9a & b. Background image, background image multiplied by silhouette matte
and bluescreens of the same brightness as far as bounce is concerned. A dark

blue or green will bounce less, but dark colors have too little color saturation
to make a high-quality matte.
The order in which the foreground is processed is important. Noise reduc-
tion to the foreground should be applied when the processed foreground is
made, before the Alpha is created, while color correction should wait until
the final composite.
tHe cOMPOsite
Once the Processed Foreground and the Alpha Channel are made, the
final composite is a straightforward two-step operation. First the values in
the Alpha Channel image are multiplied pixel-by-pixel with the background
image. The result is a silhouette of the foreground combined with the back-
ground image (see Figure 9b).
Then the processed foreground is added pixel-by-pixel to the combined
Alpha and background. If good practices have been followed, the final re-
sult is a seamless combination of foreground and background images (see
Figure 10b).
Figure 10a & b. Processed foreground, final composite

334
|
distinctiOns FrOM OtHer

cOMPOsitinG MetHOds
Projection composites
Rear projection or front projection creates a projected background on-set
behind a live action foreground. No postproduction work is needed. Creating
a high quality background image with projection is challenging. (Incidentally,
a green or blue backing can be created by front-projecting the appropriate
color.) All the issues of projection such as shadow lines and screen uniformity
apply.
rotoscoping
“Rotoscoping” or “roto” is another method for making alpha-channel
composites. The alpha-channel masks are created by hand-tracing the out-
line of the foreground action frame-by-frame. This technique of tracing live
action (named by its inventor, the pioneer cartoonist Max Fleischer) is now
more refined and widely used than ever, because 3-D conversion depends on
it. It is extremely labor intensive but very flexible, since no special backing
or lighting is required.
Present-day computer-assisted rotoscoping can produce excellent com-
posites. Nearly all the composites of foreground actors in Flags of Our
Fathers (2006) were created by rotoscoping. Good planning was the key to
avoiding difficult foreground subjects. It helped a lot that many of the actors
wore helmets, simplifying their silhouettes.
Some foregrounds are poor candidates for rotoscoping. For example if the
foreground has a bright, burned-out sky behind it, it will flare into and even de-
stroy the edge of the foreground. Flare can be very difficult to deal with (unless
of course the final background is also a bright, burned out sky.) Fine detail and
blurred motion are also difficult or impossible to roto, often requiring paint-
work replacement. The skill of the artists is the single most important factor.
Rough roto masks (“G mattes”) are often used with green screen com-
positing, to clean up contaminated areas of the backing, remove unwanted
supports, and so forth.
screen cHOices: FaBric, Paint

and PLastic MateriaLs
The best materials currently available are the result of years of research to
create optimal combinations of lamps. dyes and pigments.
Fabrics
Even an indifferent backing can give good results if it is lit evenly with nar-
row-band tubes or LEDs to the proper level (within plus or minus ⅓ f-stop)
Spill from set lighting remains a concern.
| 335
Composite Components Co.3 offers a fabric that is highly efficient, very

light, stretchy and easy to hang. It must be backed by opaque material when
there is light behind it. The green fabric is fluorescent, so it is even more
efficient under UV-rich sources like skylight. CCC also make a darker mate-
rial for direct sun use,.
Following Composite Components’ lead, many suppliers now provide
“digital-type” backings of similar colors. While similar in appearance, some
of these materials are substantially less efficient, which can have a great cost
impact when lighting large screens. Dazian Tempo fabric, a fuzzy, stretchy
material, has a low green or blue saturation when lit with white light, so it
isn’t recommended for that application. Dazian’s Lakota Green Matte mate-
rial is a better choice for white-light applications like floor covering; it is
resistant to fading and creasing, and can be laundered. Another major sup-
plier, The Rag Place in North Hollywood, CA, supplies Optic Green Tempo
Fabric, which has a built-in opaque backing. Its reflective value is between
that of Digital Green© and chroma-key green.
Measured reflective values relative to an 18% gray car (N) are: chroma-key
Green = N, Optic Green =N+2⁄3, Digital Green© +12⁄3 EV.
When using fabric backings, minimize the seams and avoid folds. Stretch
the fabric to minimize wrinkles All green materials will sun-fade with time.
If a day-lit screen must be kept up for several days it should be protected
when not in use.
Paint
Composite Components’ Digital Green© or Digital Blue© paint is the
preferred choice for large painted backings. As with fabrics, there are other
paint brands with similar names that may not have the same efficiency.
Equivalent paints made specifically for this purpose are also available from
Rosco. Paints intended for video use, such as Ultimatte chroma-key paints,
can also be used with good illuminators (lights). A test of a small swatch is
worthwhile with materials whose performance is unknown.
Plastic Materials
Plastic materials are a good alternative to fabric or paint for floor covering.
Fabric can be hazardous if loose underfoot. Painted floors scuff easily and
quickly show shoe marks and dusty footprints.
ProCyc’s Pro Matte plastic material is a good choice for floors or for entire
limbo sets. The material is a good match to Digital Green© and Digital Blue©
3. A noted researcher and pioneer in the field, Jonathan Erland of Composite Com-
ponents Co. in Los Angeles, won an Academy Award for CCC’s line of patented
Digital Green© and Digital Green© lamps, fabric and paint.
336
|
Image Courtesy ultImatte Corp.

Figure 11a & b. Unevenly lit Blue set, actor in the set with shadows and reflections
paint and fabric. It is tough, scuff-resistant and washable. It is available in

sheets, preformed coves and vertical corners in several radii. It is relatively
costly, but the cost is at least partly offset by time and materials saved in
shooting.
Great American Market in Hollywood, CA supplies GAMFloor, a self-
adhesive vinyl which is more durable than paint and easily cleaned and
repaired.
BackinG uniFOrMity and screen cOrrectiOn

Since the luminance level and saturation of the backing determines the
level of the background scene, it is important to light the backing as uni-
formly as is practical, ideally within plus or minus ⅓ f-stop.
While a perfectly uniform backing is desirable, it may not be achievable in
the real world. (Please refer to screen lighting section.) If the backing itself is
irregular in brightness, the background image seen “through” the irregulari-
ties will become equally irregular in the composite.
In Figure 11b, an actor appears in a “Virtual Set” reclining on unevenly lit
blue steps. She casts shadows on three of the surfaces and is reflected in the
bottom plane. In the final composite, the actor, her shadow and her reflection
will appear on some marble stairs; for demonstration purposes, we will substi-
tute a gray limbo for the background stairs that the blue set was built to match.
In Figure 12a, the software has faithfully reproduced the actor shadows
and reflections, along with the unevenly lit set. It is possible to clean up the
Alpha channel by increasing the contrast (gamma) in the nonuniform areas
until those values are pushed (clipped) to 1.0 (see Figure 12b). Although
clipping the alpha values eliminates the nonuniformity in the backing, the
same values on the subject and her shadows are clipped, too, resulting in
hard edges and lost shadows, reflections and transparencies.
Virtual sets like these often contain set pieces that correspond to objects
in the final background, so that the actor can climb stairs, lean against a
| 337
Figure 12a & 12b. Straight composite, composition with clipped Alpha
doorway, and so forth. The props all cast shadows on themselves and the
blue or green floor, and the actor casts shadows on everything. With lighting
alone it’s impossible to eliminate set piece shadows without washing out the
actor’s shadow.
Several software packages have features to cope with nonuniform back-
ings. Ultimatte Screen Correction software can compensate for backing
luminance variations as great as two stops.
Screen correction is easy to use: After lighting the set, shoot a few seconds
before the actors enter. This footage is called the “clean plate” or “reference
plate”. All the backing and lighting imperfections are recorded on those few
frames. Now shoot the actors as usual.
When making the composite, the artist selects a well-lit reference point
near the subject. Software derives a correction value by comparison with the
clean plate and corrects the rest of the backing to the measured level. Software
compares the clean frames pixel by pixel with the action frames, and inhibits
the correction process in the subject area (the actor) and proportionately
inhibits the correction in transparencies. Even though the set had a wide
variation in color and brightness, the fine hair detail, the actor’s shadows,
Figure 13a & b. Alpha with screen correction, composite

338
|
her reflections and the full range of transparencies in the shawl have been
retained in the Alpha (Figure 13a) and in the composite (Figure 13b).
Backing defects such as scuffed floors, set piece shadows, and color varia-
tions in the backing as well as minor lens vignetting all disappear. Note that
the actor’s shadows reproduce normally, even where they cross over shadows
already on the backing.
There is a significant limitation: if the camera moves during the shot,
the identical camera move must be photographed on the empty set for the
length of the scene. While it is reasonably quick and simple to repeat pan-
tilt-focus camera moves with small, portable motion control equipment, that
equipment is not always available. Fortunately top camera operators have
an almost uncanny skill at repeating previous moves. Skilled match movers
can bring a “wild” clean pass into useful conformance around the actor, and
remove discrepancies with rotoscoping. Some matchmovers prefer the clean
footage to be shot at a slower tempo to improve the chances that more wild
frames will closely match the takes with the actors.
When it’s not possible to shoot a clean plate, Ultimatte AdvantEdge soft-
ware can semiautomatically generate synthetic clean frames. The software
can detect the edges of the foreground image, interpolate screen values in-
ward to cover the foreground, and then create an alpha using that synthetic
clean frame. There are some limitations; it’s always best to shoot a clean plate
if possible.
iLLuMinatOrs
The best screen illuminators are banks of narrow-band green or green
fluorescent tubes driven by high-frequency flickerless electronic ballasts.4
These tubes can be filmed at any camera speed. The tube phosphors are
formulated to produce sharply cut wavelengths that will expose only the
desired negative layer while not exposing the other two layers to a harm-
ful degree. These nearly perfect sources allow the use of the lowest possible
matte contrast (gamma) for best results in reproducing smoke, transparen-
cies, blowing hair, reflections, and so forth.
Kino Flo four-tube and eight-tube units are the most widely used lamps.
They are available for rent with “Super Green” or “Super Blue” tubes from
Kino Flo in Sun Valley, CA, and lighting suppliers worldwide. The originators
4. Flickerless electronic ballasts prevent the light from being unevenly exposed on
film at speeds that are faster or slower than 24 frames per second. If one does not use
them and shots at any other speed than 24 frames per second the image will appear
to flicker. At 180° shutter, other frame rates which are evenly divisible into 120 will
be flicker free with standard balasts, assuming 60 Hz AC power. For 50 Hz, divide
fps into 100. For frame rates that are not evenly divisble, other shutter angles that
eliminate lighting flicker can be confirmed by testing.
| 339
of narrow-band tubes, Composite Components still supplies Digital Green©

and Digital Blue© tubes tailored specifically to film response. Mac Tech and
others have introduced extremely efficient LED sources with similar charac-
teristics to narrow-band fluorescent globes.
All these lamps have very high output and can be set up quickly. The
light from these tubes is nearly perfectly monochromatic; there is almost
no contamination. Flickerless power supplies run these units. Some high
frequency fluorescent ballasts and all the LED sources can be dimmed, a
great convenience in adjusting backing brightness.
Large lightweight fixtures like Kino Flo make it easy to evenly illuminate
large backings, and the doors built into most units simplify cutting the col-
ored light off the acting area.
A good scheme for front-lit backings is to place arrays of green fluorescents
above and below the backing at a distance in front equal to approximately ½
the backing height. The units may be separated by the length of the tubes, or
brought together as needed to build brightness. The lamps must overlap the
outer margins of the screen. Keep the subjects at least 15 feet from the screen.
The goal is to eliminate direct green light falling on the actor. Figure 14b shows
side and top views of an actor standing on a platform that serves to hide the
bottom row of lights. If the actor’s feet and shadow are to be in the shot, the
platform may be painted green or covered with green cloth or plastic material.
Note that if a platform is not practical, mirror Plexiglas or Mylar on the
floor behind the actor can bridge the gap from the acting area to the screen,
extending the screen downward by reflection.
A backing can be evenly lit entirely from above by placing a second row
of lamps about 30% further away from the screen and below the top row.
The advantage of lighting from above is that the floor is clear of green lamps.
Lighting from above requires careful adjustment to achieve even illumina-
tion. The overhead-only rig requires about 50% more tubes and spills sub-
stantial green or green light onto the foreground in front of the screen. To
film 180-degree pan-around shots on Universal’s The Fast and the Furious
(2001), the ace rigging crew lit a three-sided backing 30 feet high and more
than 180 feet long entirely from above.
The number of tubes required depends on backing efficiency, the film
speed and the desired f-stop. As an example, six four-tube green lamps are
sufficient to light a 20 by 20 foot Composite Components Digital Green©
backing to a level of f4 at ISO 200. Eight four-tube blue lamps yield f4 with a
20 by 20 foot blue backing from the same maker.
aLternative LiGHt sOurces

In a pinch, commercial daylight fluorescent tubes or Kino Flo white tubes
wrapped with primary green or primary blue filter sheets can produce good
340
|

UPPER GREEN
LIGHT BANK
GREEN SCREEN
Approx. 1/2
Screen Height
LOWER GREEN
LIGHT BANK
15’ minimum actor to screen
SIDE VIEW
(no scale)
Approx. 1/2
UPPER AND LOWER GREEN LIGHT BANKS
Screen Height
CAMERA
ACTOR GREEN SCREEN
TOP VIEW
(no scale)
Figure 14a & b. Side view and top views of screen lit with 6 light banks
| 341
results. The Rosco CalColor line of primaries works best. Balance to the
desired brightness in the screen color as described below. The downside is
great loss of efficiency; it takes about four filtered daylight tubes to equal the
output from one special-purpose tube.
Regular 60Hz ballasts can be used with commercial tubes at the cost of
weight and power efficiency. As with any 60Hz fluorescent lamps, 24 fps
filming must be speed-locked (nonlocked cameras are fortunately rare) to
avoid pulsating brightness changes, and any high-speed work must be at
crystal-controlled multiples of 30 fps. These tubes are somewhat forgiving of
off-speed filming because of the “lag” of the phosphors.
Backings can also be front-lit with Primary Green or Primary Blue-filtered
HMI lamps. The only advantage is that the equipment is usually already
“on the truck” when a shot must be improvised. Getting even illumination
over a large area is time-consuming, and filters must be carefully watched
for fading. Heat Shield filter material is helpful. Because of high levels of
the two unwanted colors, HMI is not an ideal source, but it is better than
incandescent.
In an emergency, filtered incandescent lamps can do the job. They are an
inefficient source of green light and much worse for blue (less than 10% the
output of fluorescents) so they are a poor choice for lighting large screens.
Watch for filter fading as above.
A green or green surface illuminated with white light is the most chal-
lenging, least desirable backing from a compositing standpoint. White light,
however, is required for floor shots and virtual sets when the full figure of
the actor and his shadow must appear to merge in the background scene.
Advanced software can get good results from white-lit backings with the
aid of Screen Correction and a “clean plate” as described above. Difficult
subjects may require assistance with hand paintwork.
eye PrOtectiOn
A word about eye protection: Many high-output tubes produce enough
ultraviolet light to be uncomfortable and even damaging to the eyes. Crew
members should not work around lit banks of these fixtures without UV
eye protection. It is good practice to turn the tubes off when they are not in
use. The past practice of using commercial blueprint tubes was dangerous
because of their sunburn-level UV output.
cHOOsinG tHe BackinG cOLOr

The choice of backing color is determined by the wardrobe or subject
color. The range of permissible foreground colors is wider when the back-
ing can be lit separately from the actor, rather than when the actor must be
photographed in a white-lit green set (a “floor shot”) for example.
342
|
A blue backing is satisfactory for most colors except saturated blue. Pastel
blues (blue eyes, faded blue jeans, etc.) reproduce well. The color threshold
can be adjusted to allow some colors containing more blue than green (such
as magenta/purple) into the foreground. If too much blue is allowed back
into the foreground, some of the blue bounce light will return. Therefore,
if magenta wardrobe must be reproduced, it is prudent to take extra care
to avoid blue bounce and flare. Keep the actors away from the backing, and
mask off as much of the backing as possible with neutral flats or curtains.
Saturated yellows on the subject’s edge may produce a dark outline that re-
quires an additional post step to eliminate. Pastel yellows cause no problems.
A green backing is satisfactory for most colors except saturated green.
Pastel greens are acceptable. Saturated yellow will turn red in the composite
unless green is allowed back into the subject, along with some of the green
bounce or flare from the original photography. The same precautions as
above should be taken to minimize bounce and flare. Pastel yellow is accept-
able. Figure 15 shows a test of green car paint swatches against green screen.
The hue and saturation of the “hero” swatch was sufficiently distinct from
the screen color to pose no difficulties in matting or reproduction. Bounce
light from the screen was carefully flagged off in the actual shot. Note that
none of the colors in the MacBeth chart are affected except for two saturated
green patches, which have become semi-transparent.
High bounce levels are unavoidable where the actor is surrounded by a
green floor or virtual set: one should not expect to reproduce saturated ma-
genta or saturated yellow on a green floor without assistance in post.
If the foreground subject contains neither saturated green nor saturated
blue, then either backing color may be used. However, the grain noise of
the green emulsion layer on color negative and the green sensor in a digital
camera is generally much lower than the grain noise of the blue layer. Using
a green backing will therefore result in less noise in shadows and in semi-
transparent subjects. Black smoke in particular reproduces better against a
green backing.
Obviously, it is important for the cinematographer and his ally the vfx
supervisor to be aware of wardrobe and props to be used in green screen
and blue screen scenes. Sometimes a difficult color can be slightly changed
without losing visual impact, and save much trouble and expense in post. If
in doubt, a test is always worthwhile. Video Ultimatte Preview (see below)
can be invaluable.
Some visual effects experts prefer blue backings for scenes with Caucasian
and Asian actors, finding it somewhat easier to achieve a pleasing flesh tone
without allowing the backing color into the foreground. Those skin tones
reflect mostly red and green and relatively little blue. For dark-skinned ac-
tors, either backing color seems to work equally well.
| 343
Figure 15. Green paint swatch test against greenscreen, composite
In extreme cases (for example if the foreground contains both a saturated

green and a saturated blue, troublesome foreground colors can be isolated
(with rotoscoping if necessary) and color corrected separately.
BackinG tyPes and LiGHtinG

The color and illumination of the backing are crucial to a good result.
A perfect green backing would expose only the green-sensitive element of
the color negative or digital sensor. Cross-color sensitivity in the negative
or sensor, imperfect illuminators and spill light from the set all compromise
this ideal. It’s no surprise that the best combinations of backing, illuminators
and camera type yield the best quality composites.
Back-lit Backings
Backings can be backlit (translucent) or front-lit. Big translucent backings
are almost extinct due to their high cost, limited size and relative fragility.
Translucent Stewart blue backings gave nearly ideal results and required
no foreground stage space for lighting. Due to lack of demand, Stewart has
never made translucent greenscreens. Front-lit backings are more suscep-
tible to spill light, but with careful flagging they can produce a result every
bit as good as back-lit screens.
344
|
Translucent cloth screens can be back-lit effectively but when back-lit,

seams limit the usable size.
Fluorescent fixtures with UV (black light) tubes will cause Digital Green©
and Digital Red© fabrics and paint to fluoresce without affecting adjacent
sets. The fabric can be lit from the front or the back, seams permitting. Ac-
tors and crew should not be exposed to high levels of UV light.
Front-lit Backings
If the actor’s feet and/or shadow do not enter the background scene, then
a simple vertical green or green surface is all that is needed. The screen can
be either a painted surface or colored fabric. Any smooth surface that can be
painted, including flats, a canvas backing, and so forth, can be used. Fabrics
are easy to hang, tie to frames, spread over stunt air bags, and so on. Please
see the section on Illuminators in this chapter for spacing and positioning
of lamps.
day-Lit Green and Blue Backings

For big exterior scenes, authentic sunlight makes a very believable com-
posite that can only be approximated with stage lighting.
Daylight is the ultimate challenge, requiring the best quality backings and
Screen Correction compositing for good results. Thanks to those advances,
there are no limits to the size of a traveling matte foreground aside from the
size of the backing.
Figures 16a and 16b show the first daylight green screen shot made for
a feature film. Every trick in the book including Ultimatte Screen Correc-
tion was used to achieve this result. Original photography was on 8-perf
VistaVision. The two reflector boards shown plus a third out of frame on the
right foreground are the only supplementary lighting.
Coves are to be avoided; as seen in Figure 16a, there is usually a wide
band of overexposed, desaturated glare in the cove. Later experience has
shown that a clean, straight line is much easier to deal with in post. A raised
platform, painted or covered with backing material, with a separate vertical
backing well behind it is ideal.
If a platform is not practical, the cinematographer of Journey to the Center
of the Earth, Chuck Schuman, recommends a flat 45-degree join between
green floors and walls.
When the screen is to be lit by direct sunlight, using the darker of the
Digital Green© fabrics or chromakey green fabric or paint makes it easier to
avoid overexposure. When the actors are in backlight and the screen is lit
by skylight, the lighter, standard Digital Green© fabrics (which fluoresce in
UV light scattered from the sky) are the best choice. Tipping the screen back
about 15° will help catch skylight. In some cases the screen must be used in
| 345
Image Courtesy of unIversal studIos
Figure 16a & b. Daylight greenscreen shot from Greedy (1993)
both front light and back light. In that instance the more reflective Digital
Green© material is the best compromise, with exposure controlled carefully
to avoid overexposure.
Limits of day-Lit Backings

Because the green backing set must be oriented to achieve the sun direc-
tion matching the background plates, one can shoot relatively few set-ups in
a day. At some times of year, the sun on the set may never get high enough to
match the background sun, thus requiring a replacement source.
FLOOr sHOts, virtuaL sets

If the actor must be composited head-to-toe into the background scene, then
the floor must also be the color of the backing. (Green is preferred for floor
shooting since the shadows will be less noisy.) The same type of white light and
lighting fixtures that light the actor are also used to light the floor and backing.
A shadow cast on a green-painted wall or floor by the subject can be transferred
(when desired) into the background scene together with the subject.
Floors may be painted or covered with fabric. Fabric can be hazardous if
loose underfoot. Painted floors scuff easily and quickly show shoe marks and
dusty footprints.
Pro-Cyc’s Pro Matte plastic material is a good alternative for floors. The
material is a good match to Digital Green© and Digital Green© paint and
fabric. It is tough, scuff-resistant and washable. It is available in sheets, pre-
formed coves and vertical corners in several radii. Permanent sets are good
candidates for this material, due to cost.
Lighting uniformity problems (within plus or minus one f-stop), color
contamination of the floor, scuff marks and green set piece shadows can
be dealt with in compositing when Screen Correction frames are available.
Sheets of 4' x 8' mirrored Mylar or mirrored Plexiglas may also be used as
a walking surface. (Please see Illumination and Reflections from the Backing
section). Of course no shadow is cast on a mirror surface, and the reflection
must be dealt with.
346
|
The choice of fabric and paint affects not only the quality of the composite,
but also lighting cost. Some screen materials are much more efficient than
others, and require many fewer lamps to light to the correct level. In general,
green screens and tubes are more efficient than blue screens and tubes. Sav-
ings on lamp rentals can amount to tens of thousands of dollars per week on
large backings.
Limitations of Floor shots and virtual sets

Floor shots and virtual sets are both difficult and rewarding, because the actor
can walk or sit upon objects in the background, climb stairs, and walk through
doorways, even when the background scene is a digital model or a miniature.
When the actor’s shadow appears in the background scene, it adds believability
to the shot. To help tie in foreground action to the plate, screen-colored shapes
can be used to either cast a shadow of something in the plate, or to catch an
actor’s shadow so that it conforms appropriately to a shape in the plate.
Alpha channel (matte) contrast must be high in a floor shot to achieve
separation from the contaminated color of the floor. Even the finest green
pigment or dye reflects significant quantities of red and blue The problem
is often compounded by glare from backlighting. Since the matte is created
by the difference between the backing color and the colors in the subject,
and since there is inherently less difference because of red and blue con-
tamination, the alpha values must be multiplied by some factor to yield an
opaque matte that will prevent the background from showing through. This
multiplication raises the gamma (contrast) of the matte image.
If the real shadow can’t be reproduced, it can sometimes be simulated
within limits with a distorted copy of the alpha channel. If necessary, the
shadow can be hand animated.
underwater PHOtOGraPHy
In addition to underwater diving or swimming shots, underwater green-
screen photography creates a zero-G environment for actors with an all-axis
freedom of motion impossible on wire rigs.
The biggest challenge is keeping the water clear of sediment and particu-
lates. Underwater diffusion causes the screen to flare into the foreground and
vice-versa; It’s ruinous to the matte edges. High capacity pumps, good water
circulation and a multistage filter are necessary to keep the water clear. It’s
also important that all personnel have clean feet when they enter the tank.
Composite Components’ green material stretched on a frame works well
under water in a swimming pool or a tank. Tip the screen back to catch light
from above, with Rosco diffusion material floating on the water surface to kill
caustic patterns on the screen. Build up the screen lighting level with green
fluorescent units above the water. Underwater Kino Flo lamps are also available.
| 347
High chlorine levels common in swimming pools bleach out the screen
quickly; pull the screen out of the tank daily and rinse it off with tap water.
FOreGrOund LiGHtinG On tHe staGe

creating the illusion: Lighting to Match the Background
Inappropriate lighting compromises a shot the instant it appears on screen,
while imperfect compositing technique may be noticeable only to experts.
Obviously, the foreground photography must match the background lens
and camera positions, but lighting considerations are just as important. This
is why it is generally preferable to shoot live action backgrounds first. (If the
background hasn’t been shot yet, the job depends on everything from careful
map-reading to educated guesswork. Even the best guesses can be defeated
by unexpected weather.)
Foreground lighting must match the background in direction, shadow
hardness and key-to fill ratio. True sunlight has nearly parallel rays coming
from a single point at a distance that’s optically equivalent to infinity. To
simulate the sun, use the hardest source available, as far away as the shoot-
ing space will allow. Multiple sources cast multiple shadows—an instant
give-away. Sometimes folding the light path with a mirror will allow the
hard source to be further away, and a better representation of the parallel
rays of the sun. Skylight fill and environmental bounce light must be shad-
owless. Therefore surrounding the actors with the biggest, broadest sources
of light available is preferable. The perfect skylight source would be a dome
like the real sky, which can be approximated on stage with big silks or white
bounces.
To help in lighting the foreground subject while the screen is lit, a contrast
viewing glass of the complementary color can be used to “black out” the
screen area. The viewing glass also reveals spill light on the screen.
envirOnMentaL BOunce LiGHt

Since the software drops out the backing and the backing reflections from
the foreground object, the subject is “virtually” surrounded by black. The
black surroundings cause no problem if the composite background is an es-
sentially dark night scene.
However, if the eventual background is a light day scene, and if the subject
had really been in that day environment, the environmental light would light
up the hair and provide the normal edge brightness along arms, sides of
the face, and so forth. The cinematographer must restore the light that the
environment would have provided from the sun, the sky, the ground and
so forth. Large, white bounces are useful in creating back cross reflection
sources just outside the frame. Otherwise, edges of arms, legs and faces will
go dark causing the foreground to look like a cutout.
348
|
Simulated light from the environment can be added digitally to replace

the suppressed screen color with color derived from the background. It’s a
slight glow around the edges that can look good when tastefully applied. The
real thing is to be preferred.
High levels of fill light in wide day exteriors while sometimes desirable
for esthetic reasons, hurt the believability of day exterior composites. Movie
audiences are accustomed to seeing more fill in close-ups, a common practice
in daylight photography.
Local color
Of course, skylight is intensely blue, so fill light supposedly coming from
the sky should be blue relative to the key. Likewise, if actors and buildings in
the background are standing on grass, much green light is reflected upward
into their shadows. If the actor matted into the shot does not have a similar
greenish fill, he will not look like he belongs in the shot. Careful observa-
tion is the key. In a greenscreen shot, the bounce light from grass is low in
both brightness and saturation compared to the screen color, so that color
cast can be allowed in the composite foreground while still suppressing the
screen. The same is true of sky bounce in a bluescreen shot.
A day exterior shot will often shoot in the f5.6 to f11 range or even deeper.
Fortunately, efficient lighting and high ASA ratings on films and sensors per-
mit matching these deep f-stops on the stage. In a day car shot, for example,
holding focus in depth from the front to the rear of the car contributes to
the illusion.
Figure 17 shows a 28'-wide screen lit with 16 four-tube Kino Flo lamps, plus
two HMI “helper” lamps with green filters on the sides. This combination made
it possible to film at f11 with 200 ASA Vision 2 negative. Curtains at left, right,
and top made it easy to mask off portions of the screen outside the frame.
Of course, when it’s possible to film foregrounds like this one in daylight,
so much the better.
color Bias in Foreground Lighting

In the past, some cinematographers used an overall yellow or magenta
color bias in foreground lighting to “help” the composite, with the intent that
the bias be timed out later. This practice is counterproductive, resulting in
false color in blurs and transparencies. If an overall bias is desired, it’s easy
to achieve in post.
illumination and reflections from the Backing

Colored illumination and reflections from the backing on the subject
must be minimized for top-quality results. Illumination and reflection are
separate issues!
| 349
Figure 17. Greenscreen lit to f 11
Green illumination from the backing can be made negligible by keeping

the actors away from the backing (at least 15 feet, 25 feet is better) and by
masking off all the backing area at the backing that is not actually needed
behind the actors. Use black or neutral flags and curtains (The rest of the
frame can be filled in with window mattes in compositing.) Any remaining
color cast is eliminated by the software.
Screen reflections are best controlled by reducing the backing size and
by tenting the subject with flats or fabric of a color appropriate to the back-
ground. In a common worst case, a wet actor in a black wetsuit, the best
one can do is to shoot the actor as far from the screen as possible, mask the
screen off as tightly as possible, and bring the environmental bounce sources
fully around to the actor’s off-camera side, without, of course, blocking the
screen. A back cross light will of course wipe out any screen reflection on
350
|

COOPS
SILK DIFFUSION
Upper blue
light bank
GRADUATED TEASERS
BLUE
SCREEN
WEIR &
SPILLWAY
Lower blue
light bank
EXISTING STAGE TANK
Figure 18. Water tank diagram

the actor but it will look false if it’s not justified by the background lighting.
Big chrome props and costumes present similar challenges. Since they
present the whole crew with a huge headache (every light shows, and
sometimes the camera crew as well) it is usually not too difficult to arrange
modifications to these items. When the visual effects team is brought in early
on, problems like these can be headed off in the design stage.
A common reflection challenge is a Plexiglas aircraft canopy or a com-
pound-curve spacesuit helmet which, depending on the lighting angle and
camera position, can show every lamp and bounce source. A bounce source
for a canopy shot must be uniform and surround the canopy 180 degrees on
the camera side. Sometimes the best solution is to shoot without the Plexi-
glas and track a CG model back in, in the composite. An advantage to CG
Plexiglas is that it can reflect the composited background.
Reflections can be disguised with dulling spray, but sometimes they can-
not be eliminated. In the worst case, reflections make “holes” in the matte
that must be filled in digitally in post. Pay particular attention to the faces
of perspiring actors, which can be very reflective. Of course, when the actor
must stand in the middle of a green-painted virtual set, some green contami-
nation is unavoidable; it will be removed by the compositing software.
Sometimes reflections are desirable: Sheets of mirror Mylar or Plexiglas
can extend a screen by reflection, even below the stage floor. Actors can walk
on mirror Plexiglas to be surrounded by the screen’s reflection. (Of course
their own reflection must be dealt with.)
| 351
Image Courtesy unIversal studIos
Figure 19. Water tank in use
In a scheme devised by the late Disney effects wizard Art Cruickshank,

ASC, an actor on a raft in a water tank was shot against a Sodium matte
backing. The backing and the actor reflected strongly in the water. This
enabled the Disney optical department to matte the actor and his reflec-
tion into ocean backgrounds. Cruickshank’s method was revived and used
effectively in blue screen shots in Daylight (1996) and more recently in
greenscreen shots in Bruce Almighty (2003) where Jim Carrey and Morgan
Freeman seem to be walking on Lake Erie, while actually standing in a shal-
low water tank on the back lot.
The spillway at the back of both tanks ensure a seamless transition be-
tween the screen and its reflection in the water.
controlling spill Light

Attentive use of flags and teasers on set lighting and black cloth on bright
surfaces outside the frame will eliminate most spill light on the backing.
(To see spill light when the backing is lit, look through a red filter.) A small
amount of white spill light from the set inevitably hits the backing. It usually
comes from the large, nearly unflaggable soft sources that simulate skylight.
Since the skylight is typically two or three stops down from the key light,
the spill has little effect on the backing. Realistic lighting should be the para-
mount concern.
If there is white light contaminating an area of the backing, a higher level
of the alpha channel can be applied in post to darken it. Since there is no
difference in color between, say, transparent white smoke or mist, and white
light of the same brightness falling on the backing, it’s clear that the less
white light contamination there is to be cleaned up, the better. Otherwise, as
352
|
©unIversal studIos
Image Courtesy BruCe almIghty,
Figure 20. Bruce Almighty water tank composite
the contamination disappears, so do all the transparent foreground pixels of
the same color. Screen Correction is invaluable in extracting the maximum
detail from smoke and spray shot against white-lit backings.
If the foreground must be flat-lit to simulate overcast, a good approach is
to bring most of the light in from overhead through a large, translucent silk.
On stage, much of the overhead soft light may be kept off the backing with a
series of horizontal black teasers hung directly beneath the silk, running its
entire width parallel to the backing. The teasers are progressively longer top
to bottom as they get near the backing, preventing the backing from “seeing”
the silk (see Figure 18 above).
LiGHtinG virtuaL sets

Inescapably, if one is lighting an actor and the surrounding floor with
white light, there is no way to control the floor brightness independently
of the actor, other than changing the floor paint or floor fabric. The only
control available is the balance between the actor’s shadow and the rest of
the floor and backing.
Lighting Procedure for Holding the shadow (Petro vlahos technique)

1. Turn on the key light so as to cast the desired shadow.
2. Measure the brightness on the floor just outside the shadow (use a spot
brightness meter and green filter assuming that it’s a green floor).
3. Light all the rest of the green floor to this measured brightness while add-
ing as little light as possible to the shadow area.
4. Light the green walls to achieve the same brightness as the floor.
5. Shadow density may be increased by blocking fill light from the shadow
area, or lightened by adding fill light to the shadow area.
Shadow density is controlled by adjusting the fill light, not by adjusting
the key light. Outside the shadow, the entire green set should appear to have
equal and uniform intensity as seen from the camera position. Strive to stay
within ±⅓ f-stop; Screen Correction can deal with brightness variations as
great as plus or minus one f-stop.
| 353
Figure 21. Blue virtual set lit to retain shadow

The human eye quickly compensates for small light changes; it is not a
good absolute measuring device. (It is however superb at comparisons.)
It is necessary to use a spot brightness meter and green filter to check for
uniform brightness. A digital camera with a computer display is also useful
for making a quick check of lighting uniformity in the three color channels.
In backlight, because of the shallow angle between the camera and floor,
the floor will not appear as green as the back wall. A diffused, polarized
white-light “glare” component is reflected by the floor because of the shallow
angle. For holding good shadows in backlight it is essential to use a polarizing
filter over the camera lens. The HN38 is recommended. Rotate the filter until
the floor glare is canceled. Ideally, the backlights should be polarized too, but
it is rarely done. Large sheets of polarizing plastic are available up to about
19" wide; they can be protected against heat with Heat Shield reflecting filter
material. Of course, HMIs emit less heat than tungsten lamps to begin with.
The composite in Figure 21 might be improved by adding a slight greenish
density and image shift to the background where it is seen through the thick
glass table top
LiGHtinG tO eLiMinate tHe sHadOw

(vLaHOs tecHnique)
1. Light the entire green set uniformly with large area diffused light sources.
2. Check uniformity as noted above.
3. Place the actor in position. If he casts a shadow, add additional low-level
lighting to return the light level in the shadow to its original level.
4. Add a modest key light to create the desired modeling, and ignore the
shadow it casts. The added key light will cause a shadow to be visible to
the eye, but because the key light did not affect the green intensity of the
floor in the shadow it has created, the shadow can be made to drop out in
compositing.
tracking Markers
When the foreground camera moves, the background must move ap-
propriately. Unless foreground and/or background can be photographed
354
|
unIversal studIos
Image Courtesy
Figure 22. Three frames from “Bruce Almighty” Steadicam shot
with a motion-control camera, tracking data must be extracted from the
foreground image and applied to the background during compositing. This
process is called Matchmoving.
Tracking marks applied to the otherwise featureless screen give the
matchmovers fixed points to track. These marks must obviously show in
the photographed scene, but ideally they should clear the foreground actors,
or at least avoid their heads, since they must be removed in the composite.
Marks are typically laid out in a rectangular pattern, with about 3' to 5' be-
tween them—depending on the lens used, the action and the distance to the
backing. Black or white tape crosses will usually suffice, though uniquely
identifiable markers are very helpful if there is much tracking to do.
Figures 22 shows the continuation of the elaborate Steadicam shot that
begins in Figures 8 through 10 earlier in this chapter that includes a pan of
about 140 degrees The black tape marks on the screen provided the data to
track in the panoramic background (Figure 23), which was seamed together
from three static BeauCam VistaVision plates shot from a tugboat with a
gyro-stabilized head. In the process the American Falls and the Canadian
Falls were moved closer together.
If camera shake or other sudden motion is required in the foreground
photography, motion blur can obliterate the tracking marks. The Aerocrane
Strobe Tracking System created by Alvah Miller provides target arrays of
LED lamps which strobe in sync with the camera shutter, giving well-defined
marks on every frame even if they are not in focus. Cylindrical LEDs have
uniform brightness even when viewed off axis.
Sometimes it is desirable to light the tracking LEDs continuously, allowing
them to blur in motion. Valuable tracking information can be derived from
the length of the blur. Consult the tracking team for their preference.
On-set Preview
On-set preview composites made with a still camera and calibrated monitor,
like the Kodak/Panavision Preview System, or a live composite made with a
hardware Ultimatte device will alert the crew to problems before they are com-
mitted to film. A few video assist companies provide this specialized service.
Using the digital Ultimatte previewer (hardware device or software on a
computer) on the motion picture set eliminates much guesswork and un-
| 355
Image Courtesy BruCe almIghty, Greenscreen and Bluescreen Photography
©unIversal studIos
Figure 23. Panoramic background from three VistaVision plates

certainty. It’s a great assist in photographing actors who must be realistically
integrated with people and objects in the background scene. Previewing
with Ultimatte also immediately identifies the acceptable limits in lighting
irregularities and wardrobe color.
If it’s a digital shoot, an output video stream must be available that’s
compatible with the Ultimatte. An outboard processor may be needed. This
yields the best preview available with all the foreground-background rela-
tionships visible at full quality.
For film shoots, a small, outboard color camera feeds the previewer. (Film
camera color taps, even when they can be switched to 100% video, are so
starved for light that they usually cannot make good composites, although
if their geometry is properly adjusted, they are fine for alignment purposes.)
Playback from disk or tape provides the background scene.
tv Monitors
It’s often necessary to matte TV images into monitors when the desired on-
screen material is not available before shooting. When the monitor is live, the
best approach overall is to feed it a pure green signal and adjust its brightness
to match the shooting stop. With this approach the room reflections in the
monitor surface can carry over believeably into the composite, the camera
can operate freely, and actors can cross over the screen without difficulties
in post. Watch for reflections of the screen on actors. Where the monitor is
just a prop, it’s often possible to rig a backlit green fabric screen in or behind
the monitor. If set lighting permits the monitor to be front-lit to a sufficiently
high level, the monitor can be painted green or covered with green fabric
behind the glass face plate. The edges of the monitor usually provide all the
tracking data needed in operated shots; no on-screen markers required.
cOsts and cOnsequences

Film Photography: choosing a camera negative
Some camera negatives are better suited to composite work than others.
Ideally, one would choose the finest grained, sharpest film available. It is also
important to have low cross-sensitivity between the color layers. Foreground
356
|
and background film stocks do not have to match, but of course it’s helpful if
they have similar grain and color characteristics.
Kodak Vision 2, 100T and 200T (tungsten balance), films are ideal for
green and blue backing work. The dye clouds are very tight and well defined.
Vision 3, 500T, the latest in a series of remarkably fine-grain high-speed
films, as one would expect is still grainier than the lower speed films. While
the 500T film is not ideal, a well-exposed 500T negative is much better than
a marginally exposed 200T negative!
An interlayer effect in these films produces a dark line around bright fore-
ground objects (such as white shirts) when they are photographed against a
green screen. Software can deal with this effect.
Kodak Vision 2, 50-speed daylight film produces superb results in sun-
light, with very low shadow noise, but require high light levels on stage.
If these 100T and 200T films cannot be used for aesthetic reasons, one
should still pick the finest grain emulsion compatible with lighting re-
quirements. Be aware that additional image processing (and cost) may be
required. A few negative emulsions have so much cross-sensitivity between
the color layers that they should not be used.
Film emulsions are constantly evolving. As an example, recent improve-
ments in red sensitivity in some emulsions have been accompanied by more
sensitivity to infra-red reflected from costumes, altering their color notice-
ably. This effect is easily dealt with by filtration—if you know it’s there. A
quick test of actors and costumes is always worthwhile.
digital Photography: choosing a camera

Since all three color channels are used in creating the composite, an ideal
camera would have high resolution and uncompressed color (bandwidth)
in each channel.
There are there are thee major factors affecting color recording:
1. Spatial resolution
2. Captured bit depth
3. Recorded bit depth and compression
spatial resolution
Spatial resolution is broadly related to the number of photosites (light-
sensitive elements) available for each color. In single-chip cameras, the
green, red, and blue photosites are on a single plane in a mosaic geometry.
Depending on the camera, groups of four to six adjacent photosites are
sampled and interpolated to create each full-color pixel. (The variation in
sampling methods is the reason that there is not necessarily a relationship
between pixel count and actual resolution of a given camera.) Most digital
cameras use a mosaic called a Bayer Array on which there are half as many
| 357
blue photosites as there are green photosites. Likewise there are half as many
red photosites as green photosites. The “missing” values are derived through
interpolation from adjacent pixels in the “de-Bayering” operation. Since
human visual acuity is greatest in the green wavelengths, Bayer’s array gives
excellent visual results from an optimally small number of photosites.
Even in the best high-resolution Bayer arrays the blue and red image is
still half the resolution of the green image, which limits the resolution and
fine detail of the mask image.5 To address this and other image quality issues,
a few high-end single-sensor cameras (Panavision’s Genesis, Sony F35) have
a 35mm-film-sized sensor with full resolution in all three colors. (Although
the sensors in the two cameras are nearly identical, at this writing the F35
has the edge in dynamic range.)
In three-chip cameras like Sony F23, the color image is split into green, red,
and blue images by a beam-splitter behind the lens. Each component color is
imaged on one of three full-resolution chips, so there is no resolution loss in the
red and blue channels, and no need to interpolate color values. The F23 uses
2⁄3" HD resolution sensors, smaller than 35mm film, which results in greater
depth of field (similar to that of 16mm), which some filmmakers love and oth-
ers find undesirable. F23’s native output is a 4:4:4 pixel-for-pixel uncompressed
image, which when correctly processed yields first-class composites.
“4:4:4” does not refer directly to RGB bandwidth, but rather to “YUV”.
The Y channel carries the luma or brightness information while U and V
are the channels from which the color information is derived (similar to Lab
color space in Photoshop). In a 4:4:4 recording, every channel is recorded
at the full color depth. “4-4-4” is actually a misnomer, carried over from
standard definition D1 digital video. Because it’s well understood to mean
full bandwidth in all three channels, its use has continued into the high-
definition-and-higher digital cinema world.)
Arri Alexa, with a 35mm-film-sized Bayer Array sensor, on paper isn’t the
winner of the Pixel Count contest. Nevertheless Alexa has produced some
of the best composite results to date, thanks to dynamic range at least the
equal of present-day film negative and extremely high quality on-board image
processing.
RED Epic and Sony F65, examples of a new generation of 4K-and-higher
Bayer Array cameras, have produced state-of-the-art composite work. F65’s
huge dynamic range is particularly useful. At 4K and above, detail loss due to
de-Bayering is less of a factor. Any untried camera should of course be tested
with the actual subject matter.
5. It should be noted that film builders use a roughly equivalent compromise: green-
and red-sensitive negative layers have more grain and less resolution than the green
layer.
358
|
color Bandwidth and compression

Assuming your camera can produce a full-bandwidth, uncompressed
RGB signal, much information can be lost when that signal is compressed
and recorded. Many HD VCRs are limited to 4:2:2 recording, which includes
rolling off the green channel’s high frequencies and applying half-bandwidth
MPEG compression to blue and red.
Just as the classic Bayer array has a negligible effect on images intended for
viewing but can adversely affect composite quality, well-compressed images
designed to look good on screen can have serious limitations when compos-
ited. Good software engineering can recover some of the lost bandwidth,
but edge detail (fine hair and so forth) and shadow noise still suffer from
compression artifacts. A laundry list of compression artifacts includes dark
or light lines trailing or leading moving objects, banding in dark areas, and
so forth. These problems are even more pronounced in DV and SD cameras.
With new cameras constantly coming on line, testing on the actual subject
matter is always worthwhile.
edge enhancement/sharpening/detail settings

Camera edge enhancement/sharpening should be turned off. The artificial
edges that sharpening produces will otherwise carry into the composite. If
sharpening is needed, it can be done during compositing.
recording
Recording in “data mode” gives maximum flexibility and best quality in
post. “Data mode” records the uncompressed data (as directly off the cam-
era sensor as the camera’s design allows) to a hard disk. This is often called
“Raw” mode, but beware: at least one camera’s (RED) “Raw” mode is in fact
compressed. Since Raw data cannot be viewed directly, a separate viewing
conversion path is required to feed on-set monitors.
If recording in data mode is not possible, shoot material intended for
postcompositing as uncompressed 4:4:4 full-bandwidth HD (or better)
video onto a hard drive or a full-bandwidth VCR, such as Sony’s 4:4:4 SR
format machines. While Arri Raw is the preferred output from Alexa cam-
eras, Alexa can also record in Apple ProRes 4444, a remarkably high quality
compressed format that has produced good composite results.
To sum up, resolution numbers are not the whole story, since some cam-
eras trade off resolution for color depth. Test your available camera and
recorder choices.
an imperfect world
You may have no choice but to shoot or record with 4:2:2 equipment.
While 4:2:2 is not ideal, don’t forget that the last two Star Wars films were
| 359
shot with 2⁄3" 4:2:2 cameras, cropping a 2.40 slice from the center, including
thousands of green screen composites. Test the camera on the subject mat-
ter. 4:2:2 can produce a satisfactory result in greenscreen (since the green
channel has the highest resolution in these cameras), but one should not
expect the ultimate in fine edge detail. (Consumer cameras typically record
4:1:1, and are not recommended for pro visual effects use.)
Whatever the camera, it can’t be overemphasized that any edge enhance-
ment or sharpening should be turned off. The artificial edges that sharpening
produces will otherwise carry into the composite and cannot be removed. If
sharpening is needed, it can be added during compositing.
FiLtratiOn
In general no color or diffusion filters other than color-temperature cor-
rection should be used on the camera when shooting green or blue screen
work. Compositing can be called “the struggle to hold edge detail”; obvi-
ously low-con, soft effects or diffusion filtering that affects the edge or allows
screen illumination to leak into the foreground will have an adverse effect.
For that reason, smoke in the atmosphere is not recommended; it can be
simulated convincingly in the composite.
To ensure that the filter effect you desire will be duplicated in the compos-
ite, shoot a short burst of the subject with the chosen filter, making sure it is
slated as “Filter Effect Reference”.
negative scanning and digital conversion

The film frames, data recording, or video recording must be converted
into frame-based digital files the software can use. It’s important not to lose
information at this step.
The three layers of the color negative are sensitive exclusively to the red,
green and blue portions of the color spectrum. When the negative is scanned,
the RGB densities of each pixel in the image are translated into red, green
and blue numerical levels in a digital memory. The three color records of
each frame are referred to as the red, green and blue channels. They are usu-
ally recorded as Cineon or DPX frames, which are uncompressed formats.
Video and data must be similarly converted into frames. This step is some-
times called “digitization”, really a misnomer since the source is already digital.
These frames are usually recorded in the DPX format. The Academy ACES
format, a universal format which is coming on line at post facilities around the
world, can accommodate DPX and any future wider-gamut color space.
cOLOr cOrrectiOn
Color correction at the scanning/conversion stage can be a major source
of data loss. It should not be built-in to image files intended for compositing.
360
|
On the other hand, a few frames recorded with the desired color and filtra-
tion will be invaluable reference in the composite step.
wOrkinG witH tHe visuaL eFFects suPervisOr

The vfx supervisor is the cinematographer’s ally on the greenscreen set.
So much depends on the quality of the original photography. Befriend that
person early in the game, and keep him or her in the loop!
The supervisor will help you achieve the photographic result you desire.
Be sure he or she understands the final look required, if (for example) you
need to shoot without diffusion on the lens. Be sure to shoot properly slated
reference with the desired diffusion, filtration and so forth, so that it can be
matched in the composite stage.
For more information, refer to the Visual Effects Society Handbook, which
has a similar chapter on this topic, with an expanded sections dealing with
setting up large greenscreens on location.
aPPendix: cOMPOsitinG sOFtware

While it’s rarely necessary for cinematographers to create composites
themselves, it’s very useful to be aware of the tools in common use.
The software described below is in wide use. All save Nuke IBK are avail-
able as “plug-ins” for most of the leading digital compositing packages,
including After Effects, Nuke, Flame/Inferno, Shake, and so on. All contain
filters to deal with less-than-ideal video like DV.
Each package has individual strong points; all are capable of first class
results with well-shot photography. Sometimes the best results come when
two programs are used on a single shot. This list is by no means inclusive.
keylight
At this writing, Keylight is the most-used package, thanks to its bundling
into After Effects Professional and Nuke. A straightforward interface makes
it very easy to use.
Keylight was developed originally at London’s pioneering Computer Film
Company by Wolfgang Lempp and Oliver James. It is marketed worldwide
by The Foundry.
ultimatte
Ultimatte and Ultimatte AdvantEdge, are still the tools of choice for dif-
ficult shots. AdvantEdge borrows from Ultimatte’s knockout concept by
processing the edge transitions separately from the core of the foreground
image, blending them seamlessly into the background without loss of detail.
The deep and rich user controls require an experienced operator to get
the most from the software. The interface works as a “black box” within the
compositing package, which can complicate workflow. One benefit of this
| 361
architecture is that the interface is identical in the wide range of supported

software packages. In 2010, Ultimatte AdvantEdge software was bundled
into Nuke, the first implementation in 32-bit color depth.
Ultimatte software was the first of its kind; it was derived from the original
film Color Difference logic created by Petro Vlahos. The digital implemen-
tation won multiple Academy Awards. Ultimatte real time video hardware
compositing devices are also available from the company.
Primatte
Primatte was orginally developed at Imagica Japan by Kaz Mishima. The
unique polyhedral color analysis allows fine-tuned color selections between
foreground and background. The user interface is intuitive and uncompli-
cated while offering many options.
nuke iBk (image Based keyer)

The Nuke IBK was developed by Paul Lambert at Digital Domain. It em-
ploys Ultimatte code carried over from earlier Cineon and Rays compositing
software packages. Like Ultimatte, it can deal with a wide variance in back-
ing color by creating a synthetic clean plate
dedicated witH Gratitude tO PetrO vLaHOs

This document draws heavily on the Traveling Matte chapter that Petro
Vlahos and I wrote for previous editions of the ASC manual. Mr. Vlahos,
a multiple Oscar® winner, perfected traveling matte systems for film with
several revolutionary inventions, including the Color Difference blue
screen technique in 1958. In the years that followed, he created analog and
then digital hardware and software versions of Ultimatte, the first high-
quality electronic compositing tools. At their core, all digital bluescreen
and greenscreen compositing software employs variants of the original
Vlahos algorithms.
with thanks to:

Chuck Schuman, Cinematographer of “Journey to the Center of the
Earth”, whose detailed notes on the previous version have been incorporated
into this article.
Paul Vlahos for images from Ultimatte Corp.
Jon and Kay Erland of Composite Components Corp. for decades of friend-
ship, advice and their tireless efforts to improve visual effects composites.
Petro Vlahos is the inventor of the film Color Difference Travelling Matte System,
and both video and film Ultimatte compositing. He has won three Oscars® for his
work.
362
|
Bill Taylor ASC is a cinematographer specializing in visual effects and co-owner

of Illusion Arts, Inc.
Persons interested in details of the purely photographic travelling matte process
should consult the 7th edition of the American Cinematographer Manual.
| 363
Photographing Miniatures
by Dennis Muren, ASC
T he recent increase in the use of miniatures in motion pictures means

that live-action cinematographers may now be called upon to photo-
graph miniatures, an area usually handled by specialists. Today’s pinpoint-
sharp lenses, very fine-grain color negatives and crystal-clear film stocks
can reveal flaws, and the solutions require the utmost attention to detail by
every member of the effects team. The effects cinematographer should talk
to the director, the live-action director of photography and the effects crew.
He or she should look at as much footage from the job as possible, especially
immediately preceding and following the miniature shot. Based on this ma-
terial, he should then visualize how the shot would have been photographed
had it been built full size and apply that information to the following:
1. The notion that miniatures look big when photographed with wide-
angle lenses from a low viewpoint is somewhat true. But when cut into
a sequence filmed from above or with long lenses, the shot may look out
of place.
2. A small f-stop is usually necessary to hold the depth of field needed to
keep the model in focus.
3. The entire model and set must appear to be in focus, as it probably
would have been if the scene had been built full size.
4. When shooting a fully miniature shot, a diffusion filter on the camera
can give an artificial atmosphere that enhances the sense of reality.
5. Match the preceding and following live-action photography as closely
as possible. Lighting units should be placed at the scaled distance from
the model to duplicate natural light falloff. Small units help the scale.
6. Artificial smoke can be used to slightly cloud the atmosphere in a min-
iature and give a realistic aerial haze. In instances where more control
is needed, bridal veil material can be tightly stretched within a set and
separately lit.
7. Panning, tilting, trucking, and even jolts and shakes can add greatly to a
shot if they are appropriate to that moment.
8. High-speed film stocks allow for extra stopping down. Perforation size
and location can be checked on each roll to help ensure rock-steady
images, if necessary.
9. For high-speed shooting, rental cameras should be loaded and tested
by the assistant who will use them. Registration steady tests should be
made at the chosen speeds, if necessary.
364
|
Model Size
Water, fire and exploding models should be as large as the budget and
safety allow, even half size if possible and shot high-speed. Intense wind can
help break up out-of-scale water droplets and, in some cases, fire. Exploding
models should be prebroken, reassembled and exploded within slow-mov-
ing, low-powered and colorful pyrotechnics, preferably with two or more
blasts. Other types of models can be built just big enough to be adequately
detailed and still carry depth of field.
Miniature explosions and fire can be dangerous because the camera may
need to be in close proximity to the miniature. Plan accordingly.
Shooting SPeedS
If there is no motion on the miniature, it can be photographed at any
speed. Water, fire, explosions and falling effects are usually done with large
models and camera speeds of up to 360 fps. The exact speed depends upon
the scale of the model and the effect desired. The chart on page 896 is a
starting point, but for the best results, tests should be made.
High-speed shots can often be expensive and unpredictable events because
of the uncertainty of required camera speeds, pyrotechnics, winds, mechani-
cal equipment, human error and the need to sequence events in much faster
succession than they will be viewed. It is not unusual to shoot miniature ex-
plosions at 300 fps to achieve a huge scale. One second of shooting time will
be seen as 12.5 seconds of screen time (300⁄24). If the pyrotechnician wants to
sequence many explosions, he may ignite the charges using a computer and be
able to specify each timing by the millisecond. Movements of the camera and
any objects in the shot will need to move 12.5 times faster than in real time.
This can make for some very fast-moving riggings. Be sure to work closely
with the rigging crew so that they will understand your problems (reach-
ing speed, operating, event timings, aborting, etc.). Achieving an adequate
level of good-looking lighting can be very difficult if shooting high-speed at
a small f-stop. If using HMIs, make sure that there will be no flicker at the
filming speeds. Scenes which are supposed to take place outdoors should be
shoot outdoors if weather permits.
With stop motion, shooting is accomplished at one frame at a time with
the object being slightly moved by hand between each frame. One fourth-
second exposures or more per frame allow for great depth of field in low
light levels. Stop-motion photography is used to give a freedom of move-
ment and expression to an object or figure.
Motion-control photography is used when the camera or an object or figure
is moved by computer-controlled motors at very slow speeds. Long exposure
times per frame allow for very small f-stops. The computer can repeat the
movements of the motors, which allows for multiple exposures. Any facet of a
| 365
shot can be isolated and wedged for intensity, color, filtration and atmosphere.
The image can be built up through multiple exposures made from the chosen
wedge frames, while the computer repeats the same motions each time.
Go-motion shooting is used when shooting animal or creature models.
The major body parts are attached to rods that are moved by computer-con-
trolled motors. Detail movements are animated by hand each frame. Single-
frame shooting allows for small f-stops at long exposure times. Coverage at
various angles and camera speeds is especially useful to help cushion the
risks on high-speed shots.
CalCulating CaMera SPeed

Explanation of table (Page 896)
The scale of the model may be stated as “inches per foot,” or as a fraction
of full size. In photographing a miniature, portraying any motion when the
speed of that motion depends on gravity, the frame rate of the camera is
governed by the scale. This includes falling objects or water, wave action, fire
or smoke, explosions in which objects are thrown into the air, etc. On the
other hand, any object (for instance, an automobile) moving at a controllable
speed can be related to the selected camera speed; the camera frame rate is
increased as the inverse square root of the scale fraction (the square root of
the relation of full size to miniature). For instance, for a miniature 1⁄16 full
size (3⁄4 inch = 1 foot), the inverse of the fraction is 16. The square root of 16
is 4 and the frame rate should be: 4 x normal = 96 fps.
In the same set, an automobile portrayed as traveling 60 mph should move
1⁄16th that speed because of the scale, but increased four times because of the
frame rate.
(Scale fraction) x (portrayed speed) x (frame rate)
(normal frame rate)
∕ x 60 x ∕ = 15 mph (or 22 ft./sec.)
Dennis Muren, ASC is the senior visual-effects supervisor at Industrial Light &
Magic. Recipient of eight Academy Awards for Best Achievement in Visual Effects,
Muren is actively involved in the design and development of new techniques and
equipment.
366
|
| 367
In-Camera Compositing
of Miniatures with Full-Scale
Live-Action Actors
by Dan Curry
“H anging” or “Foreground” miniatures have been a valuable method

of creating production value since the earliest days of filmmaking.
In-camera compositing makes the use of miniatures an attractive alternative
when postproduction compositing may not be possible. These techniques
may be of special interest to student filmmakers working with microbudgets.
With the exception of 3-D, photographed images are two-dimensional. The
camera, and the audience, cannot tell how far or near an object may actually
be. The only clues are linear perspective, aerial perspective (caused by natural
atmospheric haze) and focus. Filmmakers can take advantage of this by placing
a miniature within the frame to create the illusion that it is part of the full-scale
world being photographed. There are many ingenious ways to incorporate live
actors into foreground miniatures using platforms, ladders, and devices to cast
shadows. Once the basic principles of photographing miniatures are under-
stood, filmmakers can expand upon them to suit their specific needs.
Some advantages of hanging miniatures

◗ In-camera compositing eliminates postproduction costs.
◗ One miniature can be photographed from many different angles, and
therefore used for different shots.
◗ Good miniature builders are easier to find than matte painters.
◗ When shot in daylight, light on the miniature will naturally match the
light in the scene.
◗ Nodal pan/tilts can be utilized.
◗ When carefully aligned, people can be positioned “inside” or “on” the
miniature.
Important considerations when photographing

hanging miniatures
◗ Scout and shoot reference photos of location in advance and prepare
miniature for specific location requirements.
◗ Adequate depth of field must be established.
368
|
Figure 1.
Illustration is not to scale The actual height of
the lens above ground
level should be the same
as the height in scale
above the ground level
miniature.
E.g., If the actual lens
height is 8 feet, and the
scale of the model is ½
Mixed-Scale Compositing
inch equals 1 foot, then

the lens height above the
miniature’s ground level
should be 8 feet in the
scale of the miniature:
(½" x 8 = 4").
Depth of field must be
sufficient to carry focus
from the nearest point on
the miniature to infinity.
Use the Depth of Field
Chart in this manual to
= sh determine the aperture
needed.
h = actual height
above ground
h
sh = equivalent
scale height
| 369
Figure 1a. Approaching distant structures
◗ Plan proper set-up for accurate perspective and miniature stability.

◗ Make sure that light angles and cast shadows work with the illusion. A
backlit structure that would naturally cast a shadow on the ground in
front of it (which it cannot do unless the ground is built into the minia-
ture set) will shatter the illusion of reality. Key light should be forward
of the miniature. When scouting locations note time of day for the best
light angle.
◗ If actors are to appear “inside” or “on” the miniature, provisions must be
made to cast shadows on them where needed.
◗ Short focal length lenses offer the greatest depth of field.
◗ Aerial perspective (natural atmospheric haze) can be simulated with
diffusion sprayed onto a foreground glass in the appropriate areas. Clear
Krylon spray works well.
It is impossible to predict every situation that may arise, but the following
examples may provide useful guidelines:
Example 1: Actors approach a distant city or structure.
◗ If the actual height of the lens is 10', the ground level on the miniature
must be set below the lens an equal distance in scale. If the scale of the
miniature is ½" = 1', then ground level on the miniature should be 5"
below the lens.
◗ Depth of field must be determined (use the chart elsewhere in this
manual) to carry focus to include the miniature. If the nearest point on
the model is 4' from the camera and an 18 mm lens (on a 35 mm camera)
an f-stop of 5.6 with focus set at 6½' will carry focus from 3' 6" to infinity.
Example 2: Using a miniature as a foreground cutting piece.
Example 3: Perspective from a tall structure.
370
|
Illustration is not to scale
Figure 3. Floor level of the miniature should be the same elevation it would have if full
scale. E.g., If the set was full sized and the floor was supposed to be 25' above ground
level, the floor of the miniature should also be 25' in elevation to keep the people on the
ground scale with the miniature. The lens should be the correct scale height above the
model’s floor.
Illustration is not to scale
Figure 4. The same guidelines apply as in Figure 3

| 371
Example 4: Hanging miniature as a ceiling extension.

◗ This can be useful on soundstages where lighting grids must be used to
illuminate set and there is no room for a practical ceiling.
NodAL PANS ANd TILTS

Pans and tilts are possible if the camera is mounted so that the pivot point
is set at the nodal point of the lens. A simple way to confirm that the camera
is properly mounted at the nodal point is to set two objects (c-stands, stakes,
human stand-ins, etc.) directly in line with one another in front of the cam-
era. Try a test pan. If the foreground object appears to move in the opposite
direction of the pan, then the camera is mounted in front of the nodal point,
slide it back on the head until the two objects remain in alignment when the
camera pans.
ForCed PerSPeCTIve
Situations may arise where forced perspective or constantly changing
scales may provide the best solution to production needs. If this is the case,
close coordination between director, director of photography, visual effects
supervisor, production designer, and a highly skilled model maker is re-
quired, as there is no room for error.
MIxed SCALeS
To gain a greater illusion of distance, elements within a miniature scene
can be built, working from larger scale in the foreground to smaller scale the
farther away objects are intended to appear.
AdjuSTINg SPeedS oF MovINg objeCTS

Useful for situations such as moving miniature trains to be photographed
as an in-camera composite with full-scale elements at 24 fps.
Example:
◗ A miniature elevated train, scale ¼" = 1', is to pass through frame as part
of a live-action scene.
◗ At 4 (¼" units) per foot 4 x 12" = 48 (¼" per second)
◗ The fractional equivalent of this scale is 1⁄48. (scale expressed as a frac-
tion) x (depicted speed) = scale speed
◗ If the speed depicted is 30 mph, then 1⁄48 x 30mph = .625mph
◗ The speed of the miniature must be 1⁄48 of the desired speed or .625 mph.
◗ It is more practical to calculate in feet per second.
(5280'/3600 sec) x .625 = .92 ft/sec
◗ In a practical situation, mark off as many .92 ft increments as is conve-
nient, and adjust the speed of the miniature train to match. If you mark
372
|
Figure 4.
The actual elevation of the lens
Illustration is not to scale below the level of the top of
the set wall should be the same
as the distance in scale below
the level of the equivalent part
(bottom) of the miniature.
The actual distance to the back
wall of the set should equal the
scale distance to the back of the
le = actual lens elevation measured
from top of set wall miniature.

E.g., If the actual lens height is
d = actual distance to back wall set 6' below the top of the set
sd = equivalent scale distance wall, and the scale of the model
sle = equivalent scale measurement is ½" equals 1', then the bottom
from bottom of miniature of the miniature shoudl be set
6' in scale above the lens height
sd (½” x 6 = 3”).
Depth of field must be sufficient
to carry focus from the nearest
point on the miniature to the
back wall. Use the Depth of
le Field Chart in this manual to
determine the aperture needed.
h h Set lighting and lighting within
sle the miniature must be matched
to create a successful illusion.
The miniature can be suspended
by any means invisible to the
camera.
d
| 373
off four .92 increments, then it should take 4 seconds for your train to
cover the distance.
(Please refer to “Calculating Camera Speed” on pages 365, 682-683 and 896.)
Dan Curry is an Emmy-winning master of visual effects. Working as an art

director, matte painter, effects supervisor and producer on more than sixty televi-
sion productions. He has many classic shows under his belt, including Star Trek,
Deep Space Nine and Star Trek Voyager. He holds a MFA in film and theater.
drawing Photographic Perspective
There is only one horizon line.
All vanishing points are on the horizon line.
All elements must fall

parallel to the rays
of the vanishing point.
374
|
| 375
Light Sources, Luminaires
and Lighting Filters
CoLor TemperaTure
Color temperature describes the “true” temperature of a “black-body
radiator” and thereby completely defines the spectral energy distribution
(SED) of the object. When the object becomes luminous and radiates energy
in the visible portion of the spectrum, it is said to be incandescent. Simply
stated, this means that when an object is heated to an appropriate tempera-
ture, some of its radiated energy is visible.
The color temperature is usually described in terms of degrees Kelvin (°K).
The first visible color when an object is heated is usually described as “dull
cherry red”. As the temperature is increased, it visually becomes “orange,”
then “yellow,” and finally “white” hot.
One of the most important features of incandescent radiators is that they
have a continuous spectrum. This means that energy is being radiated at
all the wavelengths in its spectrum. The term “color temperature” can only
be properly applied to radiating sources that can meet this requirement.
When the term “color temperature” is applied to fluorescent lamps (or other
sources that do not meet the criteria for incandescence), it really refers to
“correlated color temperature.”
CorreLaTed CoLor TemperaTure

The term correlated color temperature is used to indicate a visual match
where the source being described is not a black body radiator. The term
is often abused, an example being its application to such light sources as
mercury-vapor lamps.
From a photographic standpoint, the correlated color temperature can be
extremely misleading. It is important to keep in mind that its connotations
are visual. It is a number to be approached with extreme caution by the
cinematographer.
See the Correlated Color Temperature chart on page 821.
The mIred SySTem

When dealing with sunlight and incandescent sources, the MIRED sys-
tem offers a convenient means for dealing with the problems of measure-
ment when adjusting from one color temperature to another. This system is
only for sources that can truly be described as having a color temperature.
The term MIRED is an acronym for Micro Reciprocal Degrees. The MIRED
376
|
Light Sources, Luminaires and Lighting Filters
number for a given color temperature is determined by using the following

relationship:
Sunlight should not be confused with daylight. Sunlight is the light of the
sun only. Daylight is a combination of sunlight and skylight. These values are
approximate since many factors affect the correlated color temperature. For
consistency, 5500°K is considered to be Nominal Photographic Daylight.
The difference between 5000°K and 6000°K is only 33 MIREDs, the same
photographic or visual difference as that between household tungsten lights
and 3200°K photo lamps (the approximate equivalent of ¼ Blue or ⅛ Orange
lighting filters).
1,000,000 106
Mired Value = =
Color Temperature (degrees Kelvin) ºK
As a convenience, refer to page 835 to determine the MIRED values for
color temperatures between 2000°K and 10,000°K in 100-degree steps.
Filters which change the effective color temperature of a source by a
definite amount can be characterized by a “MIRED shift value.” This value is
computed as follows:
MIRED shift values can be positive (yellowish or minus blue filters) or
negative (blue or minus red/green filters). The same filter (representing a
single MIRED shift value) applied on light sources with different color tem-
peratures will produce significantly different color-temperature shifts. Oc-
casionally, the term “decamireds” will be used to describe color temperature
and filter effects. Decamireds are simply MIREDs divided by 10.
Mired Shift Value =

([ ] [ ])
106
T2
–
106
T1
CoLor renderIng Index

The Color Rendering Index (CRI) is used to specify the stated characteris-
tic of a light source as it might be used for critical visual color examinations,
such as in color matching or inspection of objects. The CRI is established by
a standard procedure involving the calculated visual appearance of standard
colors viewed under the test source and under a standard illuminant. The
CRI is not an absolute number, and there is no relative merit to be deter-
mined by comparing the CRIs of several sources.
The CRI is of importance photographically only when it is between 90 and
100. This is accepted to mean that such a source has color-rendering proper-
ties that are a commercial match to the reference source. For example: the
| 377
HMI lamps have a CRI of 90 to 93, referred to the D55 standard illuminant
(D55 is the artificial match to standard daylight of 5500°K).
deaLIng wITh ILLumInaTIon daTa

1. Lighting Quantities — Intensity
There are two ways that intensity information is normally shown. Most
lighting manufacturers supplying instruments to the motion-picture in-
dustry tend to present their data in a rectangular format. The polar presen-
tation is more likely to be encountered with commercial/industrial-type
fixtures.
Where the intensity distribution of a lighting source is known, the illumi-
nation produced by the unit can be calculated using the inverse square law.
This is expressed as follows:
Intensity (candelas)
Illumination (footcandles) =
D2 (D=distance in feet)
Intensity (candelas)
Illumination (Lux) = 2
D (D=distance in meters)
Example: A fixture is described as having a center intensity (or center beam
candlepower) of 50,000 Candelas. What is the illumination at 25 feet? (What
is the illumination at 10 meters?)
at 25 feet: = 50,000 50,000
(a) = = 80 footcandles
Illumination 25 x 25 625
50,000 50,000
(b) at 10 meters: = = = 500 Lux
Illumination 10 x 10 100
2. Lighting Quantities — Coverage
Beam Coverage: This is described as the limit of the area covered to
within 50% of the maximum intensity.
Field Coverage: This is described as the area covered to within 10% of the
maximum intensity.
Of the two areas described above, the beam coverage is the more impor-
tant photographically. It describes the area that is illuminated at a level that
is not lower than one stop down from the center intensity. The assumption
is made, where a single distribution is shown, that the distribution pattern
is essentially circular.
Calculating Coverage from Beam Angle: The following expression allows
the computation of the coverage diameter (W) for any distance (D) and a
given beam angle. (See tangent function table on page 679.) The expression is:
378
|
W = 2 x (D) x [Tangent (1/2 Beam Angle)]

Example: For a distance of 50 feet and a known beam angle of 26 degrees,
what is the coverage diameter of the beam (50% of the center)?
D = 50 feet; Beam Angle = 26 degrees

1/2
Beam Angle = 13 degrees
Tangent of 13 degrees = .231
W = 2 x 50 x .231 = 100 x .231 = 23.1 feet
1/2 Beam Angle

Beam
Angle W
Fig. 1. Lamp Beam

3. general Comments on Calculations
In the event that it is necessary to convert from footcandles to Lux, the
value of footcandles should be multiplied by 10.76. To convert Lux to foot-
candles, divide Lux by 10.76. Lux is the metric measurement of illumination,
whereas footcandles is the English method.
phoTographIC LIghT SourCeS

The sources covered in this section include the more familiar types, such
as incandescent, and the AC and DC types of discharge lamps. The initial
section deals with the properties and characteristics of natural daylight.
The general characteristics of each type are delineated in moderate detail,
including spectral energy distributions and electrical characteristics. In addi-
tion, any special considerations for the cinematographer are noted. (See the
chart on page 837 for characteristics of typical photographic light sources.)
naTuraL dayLIghT
Daylight conditions are highly varied from a photographic viewpoint
and are based on the local atmospheric conditions, location on the earth,
altitude, time of year, hour of the day and the amount of atmospheric pol-
lutants that may be present. A brief summary of some of the possibilities are
presented in the Correlated Color Temperature chart on page 821.
| 379
Least Diffuse – In clear cloudless sunlight, the sun as the main lighting
source (key) is truly a point. This produces the hardest, most distinct shad-
ows. The incident light level from the sun on such a day can be as much
as 10,000 footcandles. The skylight contribution (fill) is about 1,500 foot-
candles. This produces a lighting ratio of about 7:1 (key to fill). Lighting
control in these situations may require booster lighting or the use of certain
grip devices such as large overhead scrims.
Most Diffuse – A completely overcast day is essentially shadowless light-
ing. The entire sky, horizon to horizon, becomes the light source. The inci-
dent level may be as low as 200 footcandles.
FILTerS For ConTroL oF naTuraL dayLIghT

A family of materials supplied as deep-dyed polyester films are made in a
variety of subfamilies for application in a number of different lighting situ-
ations. One of these subfamilies is designed to deal with natural daylight
situations. The materials are supplied in rolls ranging from 48" (122cm) to
60" (152cm) wide, and in rigid acrylic panels ⅛" (3mm) thick and in the
same widths as noted and 8' (244cm) to 10' (305cm) high.
When properly applied, image clarity is maintained through windows
treated with either the plastic film or rigid panel filters. The panels are par-
ticularly useful where wind or strong air movement may cause the plastic
film to move and produce visible highlights.
There are three kinds of filters available in these materials: neutral density
filters, color-correction filters, and combination filters. (See the filters for
daylight correction on page 829.)
InCandeSCenT LIghT SourCeS

All incandescent light sources consist of a tungsten filament in a sealed
bulb which is designed so that the terminals of the filament are accessible.
By applying a voltage across these terminals, a current flows through the
filament and raises its temperature to a level that results in the production
of visible light.
For many years, the standard incandescent globe has been replaced in
motion-picture photography with tungsten halogen incandescent lamps
that have a longer life and more consistent color temperature.
InCandeSCenT Lamp
operaTIonaL CharaCTerISTICS
Filters for Incandescent Lamps
These filters are placed in front of incandescent sources to change the
color temperature to an approximation of daylight. The filters may be plastic
film types, or dichroic filters. The dichroics are usually only utilized as filters
380
|
which fully convert 3200°K sources to an approximation of daylight (ap-

proximately 5500°K). Where the conversion is to be used for fill lighting, the
final color temperature would usually be 5000°K, which permits a higher
light transmission level.
Care must be exercised in the use of dichroic filters because they do not
have uniform filtering characteristics. When used on some types of focusing
lights (particularly some of the open reflector tungsten halogen types) as
well as wide beam floodlights, there may be color shifts at the outer edges of
the field. There also may be sufficient difference between dichroics so that
if used on multiple keys in the same scene, there could be significant dif-
ferences in the various areas being lit. A color-temperature meter should be
applied in these situations.
A broad range of very good conversion filters are available. The blue filters
(CTB) are used to convert 3200°K sources to higher color temperatures.
There is a range of conversion levels. The ⅛, ¼, ½ and full CTBs will allow
for adjustment in light sources due to voltage variations and to correct the
standard incandescent source to an approximation of 3200°K. The full CTB
correct 3200°K to daylight (5500°K). The orange filters (CTO) are used to
lower the color temperature. The full CTO will convert daylight (5500°K)
to 3200°K. The ⅛, ¼, ½ and ¾ complete this set of filters. However, the pri-
mary reason for their existence is to permit the cinematographer to achieve
artistic effects.
In order to maintain these artistic effects, the importance of testing any
filter conversion system, cannot be over emphasized. There are many small
variations that can add up to disappointment. The use of a carefully cali-
brated three-color color temperature meter is essential. Variations that could
cause trouble include: incorrect voltage, lamp globe, lamp reflector, lamp
lens, lamp filters, bounce surfaces, atmospheric conditions (open shade,
overcast, sunset, sunrise), camera lens, camera filters, negative and positive
processing, and finally poor exposure.
dC Carbon arc Sources

The open carbon arc remains in use, in particular the 225 ampere “Brute”
Fresnel lens spotlight. The tables on pages 833 and 838 summarize the vari-
ous carbon arc units, as well as the type of carbons necessary for each type.
There is also a summary of the electrical characteristics of these arcs when
properly operated.
Boosted Voltage operation

It is possible to overvoltage a wide range of standard 120-volt, 2800°–
2900°K lamp types and convert them effectively to photographic lamp types.
This system (“Colortran” boosting) was widely used in many places around
| 381
the world until the advent of the tungsten-halogen lamp. Little used in the
United States now, it is still in wide use in other parts of the world and offers
some interesting advantages. There are many situations in which this system
may be both cost-effective and functionally desirable.
Typically, when 120-volt lamps are operated at 165 volts, the color temper-
ature should be approximately 3100–3200°K. It is possible to continue the
boosting operation, and some lamp types will actually yield 3300–3400°K
when operated at approximately 185 volts. Due to the low pressure in the
standard incandescent, long-life lamps, this is a safe type of operation.
A further advantage of this system is that the standard incandescent types
utilized in it tend to be much less expensive than the photographic lamp
types rated at 3200°K at the operating voltage. Further, the expected life of
many of these lamps at 3200°K operation is directly comparable to the life
that can be expected from 3200°K-type photographic lamps operated at
their rated voltages.
Fluorescent Lighting
There is now a considerable selection of professional lighting fixtures, some
of which are very portable and compact, that utilize a range of fluorescent
lamps which closely approximate 3200°K and 5500°K lighting. These utilize
high frequency (25,000Hz) electronic ballasts which completely eliminate
any concerns regarding possible “flicker” problems from operation at usual
line frequencies. This system was perfected by Kino Flo. The fluorescent
tubes used in these systems have a typical life of 10,000 hours.
Noncolor-correct fluorescents may be corrected by using a series of either
minus green (magenta) or plus green filters on the lamps or camera. Some
color-correct fluorescent tubes still may have some green spikes in their out-
put when they get too hot. This can be easily taken care of with these filters.
(See charts pages 832 and 839.)
enclosed aC arc Lamps

Most of these lamps are operated from alternating current sources only
and require the use of a high-voltage ignition device to start and restrike
them when hot, as well as a ballasting device to limit the current.
aC arc Lamp Flicker problem

The potential for a flicker problem exists when simple inductive or
reactive-type ballasting is used at line frequency, and where there is no pro-
vision for modifying the modulation characteristics of the light source. In
such cases, it is necessary to be sure the power source is precisely at 50Hz or
60Hz, and the frame rate and shutter angle of the camera are held in certain
specific relationships.
382
|
Lamps that can exhibit flicker problems include: fluorescents, mercury-

vapor, metal-halide additive types and high-pressure sodium, as well as
photographic types such as HMI, CID or low-pressure AC Xenon arcs.
All of the lamps listed require the use of a ballasting system to provide
current limiting after the arc is struck. In the past few years, an increasing
number of reliable, electronic “flicker-free” ballasts have become available
for all HMI wattages. However, a significant number of inductive (magnetic)
ballasts are still in rental.
The time-related factors involved in ensuring a uniform exposure from
frame to frame using these types of light sources (i.e., flicker-free) are the
following:
1. Stability of the power frequency to the lamp ballast
2. Camera frame rate
3. Stability of camera speed
4. Camera shutter angle
5. Phase of shutter relative to light (particularly at high camera speed)
Simply stated, it is necessary to be sure that the same number of light pul-
sations are present during each exposure interval of the film. The amount of
variation permitted is different for different values of the parameters noted
above.
The light output of an HMI lamp modulates 83%. This means that the
light output is pulsing from 100% to 17% at two times the line power fre-
quency. The result is that there are two light pulsations for each full cycle
of the power line fundamental frequency (for 60Hz systems there are 120
pulses per second; for a 50Hz system there are 100 light pulses per second).
For 60Hz power, the only valid shutter speeds are any that can divide into
120 evenly (e.g.1, 2, 3, 4, 5, 6, 8, 10, 12, 15, 20, 24, 30, 40, 60, 120). For 50Hz
power, the shutter speed must divide into 100 evenly (e.g.1, 2, 4, 5, 10, 20,
25, 50, 100).
hmI™ Lamps
The most widely used of the new types of photographic enclosed AC dis-
charge lamps are known as HMIs. They are made in wattages ranging from
125–18,000. The chart on page 836 illustrates the various versions of this
light source.
These are considered medium-length arc types and are fundamentally
mercury arcs with rare earth halide additives. The color-temperature range
is normally quoted as being between 5600°K and 6000°K with a tolerance of
±400°K. The CRI of all these types is 90 or more, and they are dimmable and
capable of being restruck hot.
As the power to the lamp is reduced further, color temperature increases
and the CRI decreases. Where the light output needs to be reduced, it is
| 383
preferable to use neutral density filters on the luminaire in order to avoid any
possibility of a shift in color characteristics.
CalColor™ Filters
Rosco Laboratories in conjunction with Eastman Kodak has recently cre-
ated a family of filters for motion-picture lighting for which they were jointly
awarded an Academy Award for Scientific and Technical Achievement. This
is Rosco CalColor™, the first system of lighting filters specifically related to
the spectral sensitivity of color negative film.
These filters are very precise equivalents to the established range of the
very familiar “CC” filters. The Series I colors include the primaries blue,
green and red, along with the secondaries yellow, magenta and cyan. The
Series II will include six intermediaries, two of which are available at this
writing, pink and lavender. All colors are produced in the familiar 15, 30, 60
and 90 designations (½, 1, 2 and 3 stops).
All of the colors are produced on a heat-resistant base. During manufac-
ture the CIE references are continuously monitored by online computerized
colormetric equipment, which ensures the consistency of product from run
to run. The CalColor™ products are available in sheets (20" x 24") and rolls
(48" x 25') .
The principle of this system is that each color enhances the individual
color elements at each light source to which they are applied. For example,
CalColor™ 90 Green selectively enhances green transmission by reducing the
blue and red transmission by three stops. A CalColor™90 Magenta enhances
the blue and red transmission by reducing the effective green transmission
by three stops. See CalColor chart on page 834.
Another feature of the CalColor™ system relates to the colorant, selections
that were made with concern for the purity of each color. The colors finally
presented are so “clean” that they can be combined with fully predictable
results (i.e., combining 30 Cyan (-30R) and 15 Blue (-14G, -16R) results in a
Light Steel Blue filter (-14G, -46R)).
high-pressure dC Short arc xenon Light Sources

These are the best commercially available light sources for use in pro-
jection systems. The excellent color-rendering properties of this nominal
6000°K source (CRI = 98), its very compact source size and very high lumi-
nance (often referred to as “brightness” or “brilliance”), and the stability of
the arc location due to the DC operation make it the light source of choice
for motion-picture projection.
Xenon has recently found application in spotlights for motion-picture
photography due to its long throw capability.
A feature of these lamps is that the color temperature is virtually inde-
384
|
oF note…
The SMPTE standard for motion picture projection is 5600°K at 16

Footlamberts, ±2Fl. Unless you are using a dedicated Footlambert meter
to measure the screen brightness, you should use an analog reflective
spotmeter (digital meters will be fooled by the flickering display). Set
the meter to 100ASA for 1⁄50 of a second.
32Fl = f/4.8 or EV 10 16Fl = f/3.2 or EV 9 12Fl = f/2.8 or EV 8 2⁄3
10Fl = f/2.5 or EV 8 1⁄3 4Fl = f/1.5 or EV 7
pendent of lamp type, lamp current, wattage, dimming or boosting, lamp

voltage or the age of the lamp.
Low-pressure aC/dC xenon arc Lamps

These are low-pressure long arc light sources. DC units are available in
fixtures at power levels of 1.1K, 2.2K, 3.3K and 10K.
The AC Lamps represent the highest-powered modules available in the arc
discharge field (AC or DC operated), with modules ranging from 10K to 50K.
It is claimed the DC units do not have any flicker problems, while the AC
units require the same caution as with AC discharge lamps.
SpeCIaL purpoSe phoTographIC LIghTIng

eQuIpmenT and SySTemS
Lightning and other Flashing-Type Lighting effects
Several companies now supply electronically controlled Xenon flash tubes
for simulating lightning and other types of flashing effects.
The lightning-effect fixtures range from a handheld unit (20K) used to
simulate strobe flashes associated with still photography, up to a 2,000,000
watt unit. These sources all produce light at 5600°K, which is dimmable from
100% to approximately 20%. An attractive feature of this equipment is that
it is not necessary to synchronize the flashes because the shortest flash dura-
tion of any of the heads is 1⁄24 of a second.
A broad range of functions can be accomplished with the control devices
that are available. The simplest and most often used is one that will vary
the light intensity of a unit down to 20% of maximum and control the flash
duration up to 1.5 seconds per flash. This unit also permits random intensity
changes of up to 3 f-stops in level as many as 24 times per second to approxi-
mate the look of natural lightning. Another unit with the same functional
controls manages up to four lighting units.
A number of other specialized controllers are available. These include a
precision fader that allows programming of flash time and intensity frame
by frame up to 99 frames, and has four stored programs, “Photoflash,” “Ex-
| 385
oF note…
CAUTIONS: Xenon lamps have high internal pressure, even when cold
(cold, the internal pressure can be up to approximately 150 psi, and
in operation this pressure can be as high as 450 psi). The lamps are
supplied with a protective jacket over the bulb, and this should not be
removed until the lamp is fully installed. It is required that suitable face
shield, body jacket and gauntlets be used any time that the protective
jacket is removed. The protective jacket should be installed before steps
are taken to disconnect and remove a lamp.
A further caution must be considered relative to the characteristically
high luminance of the arc in these sources. Direct viewing of the arc can
result in serious damage to the retina.
Xenon lamps produce a considerable ultraviolet component (up to
about 6% of the total lamp energy output). This can result in the pro-
duction of ozone which is harmful to health if breathed for extended
periods or in poorly ventilated spaces. This caution should be observed
even when using “ozone-free” versions of these sources.
plosion,” “Machine Gun” and “Undulate”. Custom programs can be created

and stored in memory. Optical and acoustic triggers initiate the flash effect
by sensing a bright light or by a loud sound. Another device allows timing
flashes by triggering from standard SMPTE time code. There is yet another
that can be programmed to chase up to sixteen lights and can be preset for
each light for number of frames on and frames off.
Synchronized Stroboscopic Lighting

These typically utilize Xenon flashtubes, which produce a good approxi-
mation of daylight (about 6000°K) and a relatively stable color temperature
throughout life.
The strobes must be synchronized to the camera shutter. It is imperative
that the units flash when the shutter is fully clear of the gate; otherwise a
partially exposed frame will result. To check camera synchronization, the
lens should be removed and the cavity illuminated with the strobe with the
camera turned on. The shutter should appear to be frozen in one position.
oF note…
CAUTION: People with photosensitive epilepsy should be informed

that strobe lighting will be in use.
386
|
The control equipment for these strobes permits the addition of delay to
the pulse in degree increments. The position of the shutter will move ei-
ther forward or backward in relationship to the gate until it is in the proper
position. For reflex cameras the strobe fires twice for each frame, once to
illuminate the subject and a second time to illuminate the viewfinder.
CommerCIaL/InduSTrIaL LIghT SourCeS

With today’s readily available color-corrected light sources, it is much
more cost-effective to change out the globes at a location rather than struggle
with correcting the existing sources with gels and filters.
However, most filmed television shows transferred and corrected elec-
tronically require only that the location illumination is the same color tem-
perature; correction to “normal” can happen in the telecine suite.
LumInaIreS Fresnel
Lens
Fresnel Lens Spotlights Filament
Spherical
Fresnel spotlights are made for Reflector
standard incandescent and tungsten

halogen incandescent sources, and a a Flood
Beam
also for the range of HMI, CID and Specular
Finish
CSI arc discharge lamps. The range
of wattages, taking into account all d
types, is from 100–24,000.
These luminaires represent the Fig. 2. optical system of standard Fresnel
most widely used motion-picture spot light when in full flood position
lighting units. They provide the
means for changing the beam diameter and center intensity through a rela-
tively broad range. Using standard incandescent lamps, the spot-to-flood
ratio may be 6 to 1 or so, and with a tungsten-halogen lamp it may be pos-
sible to extend this ratio to 8 or even 9 to 1 under some circumstances.
The optical system of
Fresnel
Lens
these luminaires is the
same for all the variations
Filament
Spherical
Reflector
that may be presented. The
light source and a spheri-
a a Spot cal reflector are located in
Specular Beam
Finish a fixed relationship to one
another. This combination
d
of light source and back
reflector is designed so
Fig. 2. optical system of standard Fresnel spot light when that the spherical reflector
in spot position reflects the energy being
| 387
radiated toward the back of the housing through the filament and toward
the lens. The effect intended is that the energy radiated to the lens appears to
come from a single source. The combination of reflector and light source is
moved in relation to the lens to accomplish the focusing.
One of the most important features of the Fresnel lens spotlight is its abil-
ity to barndoor sharply in the wide flood focus position. This property is
less apparent as the focus is moved toward a spot (at spot focus it is not
effective at all). The barndoor accessory used with this spotlight provides
the cinematographer with the means for convenient light control. The sharp
cutoff at the wide flood is, of course, due to the fact that the single-source
effect produces a totally divergent light beam. The action of the barndoor,
then, is to create a relatively distinct shadow line.
Occasionally it may be desirable to optimize the spot performance of these
units, and for this situation “hot” lenses are available. These tend to produce
a very narrow beam with very high intensity. It is important to remember
that the flood focus is also narrowed when these lenses are used.
open reflector Variable-Beam Spotlights

These are typically the tungsten-halogen open reflector spotlights. There
are also some low-wattage HMI-types available. These nonlens systems pro-
vide “focusing” action, and therefore a variable diameter beam, by moving
the light source in relationship to the reflector (or vice versa). These types
of units are available for sources ranging from 400–2,000 watts. One of the
drawbacks of this system, when compared with the Fresnel lens spotlights,
is that there are always two light sources operative. The illumination field
produced by these systems is the sum of the light output directly from the
bulb and the energy reaching the field from the reflector. The use of the
barndoor accessory with these lights does not produced a single shadow due
to this double-source characteristic. Typically a double shadow is cast from
the edge of the barndoor.
d d
Spot Wide Flood

Position Position
Fig. 4. non lens, variable beam spotlights

388
|
The great attraction of these luminaires is that they are substantially more
efficient than the Fresnel lens spotlights. Typical spot-to-flood intensity ra-
tios for these types of units is between 3:1 and 6:1.
Tungsten-halogen Floodlights
A variety of tungsten-halogen floodlighting fixtures take advantage of
these compact sources. Two of the more typical forms are treated here. These
fixtures are available in wattages from about 400–2,000.
There are types of “mini’’ floodlights using the coiled-coil, short-filament,
tungsten-halogen lamps, which provide very even, flat coverage with extreme-
ly sharp barndoor control in both directions. Due to the design of the reflector
in this system, the light output from this fixed-focus floodlight appears to have
a single source. This accounts for the improved barndoor characteristics.
Cyclorama Luminaires
These lighting fixtures were originally developed for lighting backings in
theater but have broad application in similar situations in film. Because of
the design of the reflector system, it is possible to utilize these fixtures very
close to the backing that is being lit and accomplish a very uniform distribu-
tion for a considerable vertical distance. Typically these units are made for
tungsten-halogen linear sources ranging from 500–1,500 watts.
Based on the variations in design, some of these may be used as close as
3' to 6' from the backing being illuminated. The spacing of the luminaires
along the length of the backing is in part determined by the distance of these
fixtures from the backing itself.
Soft Lights
Soft lights, which attempt to produce essentially shadowless illumination,
are made in wattages from 500 up to about 8,000 and typically utilize mul-
tiple 1000w linear tube tungsten-halogen lamps. The degree of softness is
determined by the effective area of the source.
The Aurasoft™ is unique, in that it produces a greater area of coverage than
a comparable conventional unit, at a full stop advantage in light level and
with comparable “shadow-casting” character. These units can be quickly
converted in the field between tungsten-halogen and HMI light sources.
Also available is a plastic tubular diffuser with a reflector at the closed
end which is fitted at the open end, with a tungsten-halogen spotlight. The
configuration allows for the unit to be easily hidden or placed in a corner to
provide soft light that can be used very close to the actors.
The helium filled balloons are designed to contain either tungsten-halogen
or HMI sources in various combinations. These balloons are tethered and
can be used up to an altitude of about 150 feet (45 meters). They range in
| 389
Shield Painted
Surface
Reflector
Tungsten- Shield
Halogen
Lamp
Painted Standard
Surface Incandescent
Lamp
Cone Light Folded "Soft" Light
Painted
Surface
Tungsten-
Halogen
Lamp
Umbrella
“Soft” Light Light
Fig 5. Reflector systems of various “soft” lights
size from approximately 4 feet in diameter (1.2 meters) to tubular shapes as

much as 22 feet long (6.6 meters) by 10 feet in diameter (3 meters). There is
a range of power levels, with tungsten halogen lights up to 16,000 watts and
HMI lights up to 32,000 watts.
“par” Fixtures and Lamps

The most popular of these are the Par64 and Par36 configurations. These
lamps have a parabolic reflector that has a high-reflectance aluminized coat-
ing and a prismatic type of front lens. Typically they are supplied in VNSP
(very narrow spot), NSP (narrow spot), MFL (medium flood) and WFL
(wide flood) lens systems. They are extremely efficient optical systems.
Fixtures are available which assemble multiples of these types of lamps
for daylight fill applications or for long-throw stadium and arena lighting
requirements. Both 3200°K and dichroic coated versions (approximately
5000°K) are available.
390
|
Beam projectors Beam of

A luminaire consisting of a large Light Baffel Rings
parabolic mirror with the globe Globe
filiment placed at the focal point of
the mirror, so as to produce a parallel
beam of light. Sizes are described by Parabolic
Reflector
the diameter of the mirror: 18", 24", Mirror
36". The lamp source is either HMI or
tungsten. In by gone years the source
was the carbon arc, thus the old name
Fig. 6. Beam projector
“sun arc”. The beam width can be var-
ied a small amount. A series of baffle Spherical Condenser
rings are placed to cover the filament Reflectors Lens
Filament
area to eliminate nonparallel rays of Meniscus
light. Lens
optical Lens System Lights

The most commonly used is the
ellipsoidal spotlight. It consists of an
ellipsoidal reflector and two plano
convex lenses of various focal lengths
Fig. 7. Dedolight
that project very sharp shadows.
Various shadow-making devices can
be introduced on the focal plane
in the interior of this light to create
sharp shadows (e.g., variable iris, cut-
ter blades, and pattern holder).
A variant of this design is the Dedo
light and the Focal spot attachment, Ellipsoidal Shutter Lens
which can be placed in the barndoor Reflector Pair
holder of Fresnel spotlights. Fig. 8. ellipsoidal Spotlight

The biggest option is the theatrical
follow spot. The size and intensity of the beam can be varied by an internal
lens and reflector system. Sources for shorter throws are either tungsten or
low wattage HMIs. For stadium use HMI, Xenon or Carbon arc source are
used. Many types of internal and external accessories are available: color
changers, pattern holders, irises and motorized wheels for moving effects.
Led (Light emitting diodes)

Flat panels consisting of colored or white LEDs that simulate tungsten
or daylight color temperature. They may be programmed via computer to
produce most colors in constant or chasing mode.
| 391
LIghT-ConTroL aCCeSSorIeS
Barndoors
The purpose of this accessory is to prevent the illumination beam from the
fixture from reaching certain portions of the set. A relatively well-defined
edge can be established that delineates the end of an illuminated area and the
beginning of an unilluminated zone.
Barndoors are most effective when used on Fresnel spotlights when the
spotlight is in the wide flood position. The effectiveness of the barndoor is
reduced as the focus is moved toward spot and is totally without effect at the
spot focus.
The effectiveness of the barndoor as an accessory on other types of lumi-
naires varies sharply with the design of the specific item. In a number of the
open reflector tungsten-halogen systems (particularly floodlights), barndoor
effectiveness is limited to the edge of the barndoor that is parallel to the source.
Snoots
This is a funnel-shaped device used to limit the beam of a Fresnel spot-
light. Available in various diameters.
Scrim
The type of scrim referred to here is placed directly in the accessory-
mounting clips on a luminaire. This type of scrim is normally wire netting,
sometimes stainless-steel wire, which is used as a mechanical dimmer.
The advantage of the scrim is that it permits a reduction in light intensity
in several steps (single and double scrims) without changing the color tem-
perature or the focus of the luminaire. Contrary to popular belief, it is not
a diffuser.
The half-scrim permits the placement of scrim material in only half of the
beam and is widely used on Fresnel spotlights. It overcomes the problem
encountered when the Fresnel is used at fairly high angles. The portion of
the beam striking the floor or objects near the floor closest to the luminaire
produces intensities that are too high to match the desired level at the dis-
tance associated with the center of the beam. The reason for this, of course, is
the substantial variation in the distances that the illumination energy travels.
The half-scrim applied on the portion of the beam impinging on the nearest
objects can overcome this problem.
gel Frames
Different forms of these holders are made and designed to fit into the ac-
cessory clips on the front of most luminaires. They permit the use of various
types of plastic filter materials to modify the characteristics of the beam.
Color media may be put in these holders to affect color, and a wide range of
diffusion products are available.
392
|
electrical dimmers
The old-fashioned resistance and autotransformer dimmers have given
way to the solid state SCR dimmer systems. Computer management DMX
controllers cannot only control the intensity of each luminaire, but can
switch cues, replug circuits, precisely control the duration of a dim, and con-
trol many other accessories (color wheels, lamp movement and focus). All of
these cues can be recorded and stored in the computer for replayability. With
arcs, mechanical shutters are motorized to execute the dim.
grIp aCCeSSorIeS For LIghT ConTroL

diffusers
There are various diffusion materials sewn on wire frames of different
types and size which permit the diffusion of both artificial and natural
sources.
They are translucent materials (various textiles) that truly act as diffusion.
When supplied in very large sizes and supported from a single point, they
are called butterflies; when the frame becomes extremely large and is sup-
ported from two or more points, it is called an overhead.
Solids, gobos, Flags or Cutters

These all come in the same form as the various scrims, dots, fingers, but-
terflies and overheads but are opaque. They are utilized to keep light from
falling in a given area and permit very fine adjustment of the lighting in a
large area.
The cucoloris is a cutout pattern placed in the path of the spotlight in
order to cast a shadow that might be comparable to the light coming through
the leaves on a tree.
reflectors
Reflector boards are widely used for redirecting sunlight and modifying
its characteristics so that it is suitable for use as set illumination and fill light.
These boards have been surfaced with various reflecting media, usually sign-
painter’s leaf, either silver or gold.
Most reflectors have two sides, a soft side and a more mirror-like surface,
commonly referred to as the “hard” side.
egg Crates
Large, fabric egg crates can be stretched in front of soft diffusion material
to control spill light.
| 393
Ultraviolet
Infrared
Near Far
Violet Blue Green Yellow Red

380 400 500 600 700 760
Wavelength in Nanometers
Fig. 9. Visible Light Spectrum

Violet: 380-430nm; Indigo: 430-450nm; Blue:450-480nm;
Green: 510-550nm; orange: 590-610nm; Red: 610-760nm
394
|
| 395
LED Lighting for
Motion Picture Production
by Frieder Hocheim
O ver the last four years LED lighting fixtures have been working their way
into the motion picture industry. Lead by companies such as Litepanels
Inc., products have been introduced that effectively exploit the inherent char-
acteristics of LED technology: low DC power and amperage draw, low heat,
and dimmable without color shift. LEDs are primarily powered by low-volt-
age DC with low-energy demands. This has enabled the design of some inno-
vative battery-operated fixtures. Untethered from a power cable, the instru-
ments have provided a handy, easy-to-rig fixture well suited to the fast-paced
shooting styles of today’s production environments. The use of multicolored
LEDs based on RGB principles or the newer multicolored LED mixing allow
for products that expand the potential for accurate spectral displays.
LEDs display a number of advantages over conventional lighting technology:

w Low power requirements, allowing for battery operation
w High Lumens per watt efficiency
w Color mixing
w Video Imaging display capability
w Full range dimming without shift in color
A number of companies have emerged over the past few years that are
providing innovative LED products: Color Kinetics, Litepanels, Mole-Rich-
ardson, Gekko Technologies, Element Labs, Kino Flo, Zylite, Nila, LEDz
among others.
Challenges however remain in this new technology. For a cinematogra-
pher it is all about the light. If a fixture’s light characteristics aren’t correct
for the scene, it will be passed over. Cinematographers strive for clean edge
shadows or soft light sources that display diffuse shadow lines. The LED will
have to deliver light as good as or better than existing tools provide. Energy
savings alone will not be the reason to embrace the LED.
The introduction of the LED has presented lighting designers with some-
what of a predicament as it pertains to motion picture lighting. The LED
is a point source much like Thomas Edison started with back in the 1880s.
Edison’s source was wrapped in a clear glass envelope. One bulb alone pro-
396
|
LED Lighting For Motion Picture Production
vided little light. Numerous light bulbs needed to be combined to provide

adequate light levels.
The same challenges that the early designers of lighting instruments dealt
with are now being addressed again in the twenty-first century. Given the
number of patent filings the U.S. patent office has seen over the LED, you
would think that Thomas Edison had never existed. We are essentially rein-
venting the light bulb.
LED design challenges can be broken down to a number of issues:

w Glare
w Multiple shadows
w Loss of efficiency when diffused
w Color rendering
w Heat management
GLarE
As a bright point source the LED is very glaring when viewed on axis. Large
quantities have to be assembled to provide adequate levels of illumination.
When an LED fixture is in the direct eyeline of an actor they tend to com-
plain of discomfort.
There is not much you can do to reduce glare if you need the light output.
Adding a diffuser to the fixture will reduce the glare but leads us into the
next problem: substantial light loss and loss of efficiency.
Diffusion anD LiGht Loss

In order to soften the light and reduce the glare from an LED, it needs to
be diffused. Diffusion breaks down the optical characteristics of the LED
and dramatically reduces its light output. Most of the heralded advantages of
lumens-per-watt efficiency are gone, along with any advantage over existing
technologies.
MuLtiPLE shaDows
Depending on the spacing of the LEDs, the light characteristics can display
multiple shadow lines. Much like the number of pixels per inch determines
the resolution in a digital image the density of LEDs in a fixture will define the
shadow line characteristics of an LED fixture. The tighter the spacing of the
LED, the more homogenous and clean the shadow. Open faced fixtures with-
out diffusion can give off very distracting multiple shadows from a barn door.
CoLor
Early adoption of LEDs was hampered by their spectral output. LED man-
ufacturers specified their color space as dictated by the demands of general
| 397
lighting applications or a larger customer such as the automotive industry.

Lumen-per-watt efficiencies dictated the color point targets. This inevitably
resulted in light that resembled that of early fluorescents. They were green
and did not render images on film or digital accurately. The introduction of
color binning or measuring and sorting LEDS according to their specific
color performance enabled fixture manufacturers to provide a better color
more suited for the imaging industry. This is an approach that adds more
cost and reduces the yields of each manufacturing run.
hEat ManaGEMEnt
Although the lamps are highly efficient it comes at the cost of heat. Put
a few 2 watt or 5 watt LEDs on a matrix and heat builds up rapidly. The
lightweight LED now has to be linked to a much heavier and bulkier heat
management system or heat sink. This gives the designers challenges as
to how to exploit the LED’s advantages of size and light output against the
weight and size of the heat-sink requirements. So if it ends up as hot and
bulky as a 1 K Fresnel and the color is not as good as an incandescent, will
the industry care to work with it?
The challenges are being addressed in innovative ways and are resulting in
new and useful lighting tools. Together with the inherent advantages of the
LED, we can expect it to play a major role in providing lighting solutions to
the motion-picture industry.
tyPEs of LED fixturEs

avaiLabLE to thE CinEMatoGraPhEr
A number of companies are offering LED products that have found accep-
tance by cinematographers. Some of the companies are designing fixtures
specifically for the imaging industries and some are hybrid offerings that
have been designed primarily for event lighting or architectural applications.
Architectural and event products for the most part do not display color cor-
rect or high CRI illumination.
Companies such as Litepanels, Gekko Technologies, Kino Flo, Nila and
Zylight are marketing directly to the motion picture and TV industry. Color
Kinetics and Element Labs come from a strong architectural and event light-
ing market.
Litepanels – www.litepanels.com
Product names: Micro Series, Mini-Plus Series,
1 x 1, 2 x 2, 4 x 4, Ringlite Series, SeaSun
One of the first successful companies to exploit the advantages of LEDs
was Litepanels. Founded by industry gaffers, they have provided innovative
lighting tools for on camera lighting as well as more general set lighting ap-
398
|
plications. Their battery operated Micro, Mini-Plus and Ringlite products

are designed to mount onto the camera as eye-lights or fill light. Their 1 x 1
fixtures provide a lightweight lighting tool that can operate on battery power
or AC. By carefully binning their LEDs they have been able to provide a high
CRI light quality essential for good imaging.
Color Kinetics – www.colorkinetics.com

Products: ColorBlast, iColor Cove, ColorBlaze,
Intelliwhite Lighting systems
Color Kinetics has had great success in the architectural markets and has
a substantial intellectual property portfolio relating to LED color controls.
Products such as the Blast series and Cove fixtures have found applications
incorporated into motion picture set lighting. RGB color mixing does not
provide high color rendering light but is widely used in color changing
applications such as backgrounds or effects lighting. Color Kinetics also
features a white light product range. These systems incorporate special bin-
ning techniques in selecting their LED components to ensure high color
rendering white light.
Gekko technology Ltd. – www.gekkotechnology.com

Product names: K-lite, Kicklite, kuelite lenslite,
kisslite, george, Kelvin Tile, karesslite
Founded by cinematographer David Amphlett, Gekko Technology has
introduced a very successful line of high color rendering LED lighting in-
struments. Taking advantage of the low DC power requirements and the
lightweight nature of the LED, Gekko produces a range of portable lighting
instruments that can be mounted to the camera or used for small area set
lighting applications.
Element Labs – www.elementlabs.com

Product names: Stealth Display, Versa Tile, Versa Ray,
Versa Tube, Helix Screen, Versa Tile T100, Helix P1 Screen, Cobra
Element Labs comes from a strong background in large area video dis-
plays for concert/event lighting and architectural applications.
The Stealth Display, Helix Screen, Versa Tile, Versa Ray and Cobra are
all variations on a means of creating large area architectural displays that
incorporate video images and patterned color changes.
The Versa Tube looks like a color changing fluorescent tube that can be as-
sembled into a large video wall. Individual tubes can be rigged into set designs
and controlled through a standard DMX512 protocol from a dimmer board.
The Versa Bank is a portable panel of LEDs designed for motion picture and
TV production. It has been effectively used in rendering sophisticated color
| 399
changing light effects for process shots. The Versa Bank can simulate the ever
changing source light in synchronization with a process background plate.
nila – www.nila.tv
Product names: Nila JNH series
Founded by gaffer/grip Jim Sanfilippo, Nila offers a high powered LED light-
ing system consisting of a module that interconnects to form larger fixtures.
Interchangeable lenses vary the beam angles in 10-, 25-, 45- and 90-degree
increments as well as vertical and horizontal Elliptical beam options. Other
accessories such as yokes and gels flesh out the system. The module is dim-
mable through a DMX protocol and is available in either daylight or tungsten
equivalents. The units operate on a universal 90-240 AC/DC power supply.
Zylight – www.zylight.com
Product names: Zylight IS3, Z90 & Z50, Remote using Zylink protocol.
Founded by Charlie Collias a veteran of video and documentary produc-
tion and his brother Jim from an electrical engineering background, Zylight
produces a range of color-changing RGB portable lights well suited for
on-camera and general studio lighting applications. An innovative wireless
remote-control system, ZyLink offers control over numerous fixtures at one
time. The units operate on AC or DC power. Given the high density of LEDs
on the light emitting surface the shadow characteristics are very clean not
multiples of shadow lines.
LEDz – www.led-z.com
Product names: Mini-Par, Brute 9, 16, and 30.
Founded by veteran HMI designer Karl Schulz, LEDz offers a range of
small portable lighting instruments.
The Mini-Par is a 12VDC on-camera light that offers various beam angles
using a set of accessory lenses. The Brute fixture family consists of fixtures
from 30 watts, 50 watts, and 90 watts. The fixtures are available in Daylight,
5500°K or Tungsten 3000°K.
400
|
| 401
An Introduction
to Digital Terminology
by Marty Ollstein
and Levie Isaacks, ASC
W hat follows is much more than a simple glossary of terms. This is a

comprehensive overview of the science behind motion imaging and
a cornerstone in forming a working knowledge of digital technology and
how it relates to analog image capture.
ANALOG – The natural world is analog. Light and sound are described as
waves whose shape varies with their amplitude and frequency in a con-
tinuously variable signal. An analog signal is understood to be formed
by an infinite number of points. Analog human vision perceives light as a
continuous gradient and spectrum from black to white. Film is an analog
medium and can record a continuous spectrum. Digital video cameras
capture a scene as analog voltage and then digitize it with an A/D (analog-
to-digital) converter to create linear digital code values.
A/D CONVERSION – Analog-to-digital conversion transforms analog data
(such as light intensity or voltage) into a digital binary format of discrete
values. Referred to as digitization or quantization.
ANAMORPHIC – An optical process in which a widescreen (wide aspect
ratio) image is recorded onto a narrower target (film or sensor) using an
anamorphic lens to squeeze the image horizontally. An anamorphic lens
will squeeze a 2.40:1 ’Scope scene onto a 1.33:1 negative frame or 4 x 3
camera chip. The format uses almost the entire image area with no waste
(of pixels or negative area), resulting in a higher-resolution image. For
display, an anamorphic projector lens is needed to unsqueeze the image.
ARTIFACT – A flaw or distortion in an image—a result of technical limita-
tion, incompatibility or error. An artifact can be introduced at any step in
which an image is altered or converted to another format.
ASPECT RATIO – Ratio of screen width to height.
Common ratios:
1.33:1 (4 x 3) Standard TV—or
35mm Full Aperture (Silent)
1.37:1 Academy Aperture—or
Regular 16mm Full Aperture
1.66:1 European Theatrical Standard
402
|
An Introduction to Digital Terminology
1.66:1 Super-16mm Full Aperture

1.78:1 (16 x 9) HDTV
1.85:1 American Theatrical Standard
2.40:1 Anamorphic or ’Scope
BAKED IN – Changing or re-recording recorded images to integrate a par-
ticular look. This action limits options in postproduction but insures that
the look is preserved.
BANDING – A digital artifact caused by digital processing in which lines
or bands appear in an image that previously displayed a smooth unbroken
gradient between light and dark or two different color values. Banding can
be caused by the down-conversion of the bit-depth or sampling ratio of an
image that results in a loss of data.
BANDWIDTH – Literally, the range between the lowest and highest limit-
ing frequencies of an electronic system. Commonly used to refer to the
size of the “pipeline” employed to transmit data, quantified by the amount
of data that can be transmitted over a given period of time, such as mega-
bits per second. Compression techniques are used to reduce the size of
image files to facilitate real-time display of high-quality images in systems
with limited bandwidth.
BAYER PATTERN – A chip design that allows a camera to record full color
using a single chip. The pixel array employs a pattern of 2 x 2 matrices of
filtered photoreceptor sites—two green, one red and one blue. Three-chip
cameras record red, green and blue on three separate chips. (See Figure 7.)
BINARY CODE – The mathematical base of digital systems that uses com-
binations of only 0 and 1 to represent all values. A mathematical represen-
tation of a number to base 2.
BIT – The smallest increment of digital information. A bit is a digit of binary
code which can define only two states—0 or 1, on or off, black or white.
BIT DEPTH – Bit depth determines the number of steps available to de-
scribe the brightness of a color. In a 1-bit system, there is only 0 and 1,
black and white. An 8-bit system has 256 steps, or numbers from 0–255
(255 shades of gray). Until recently, 8-bit was standard for video and all
monitors. Most monitors still have only 8-bit drivers, but many HD video
systems support 10-bit signals through their image-processing pipelines.
A 10-bit system has 1024 steps, allowing more steps to portray subtle
tones. A linear representation of light values, however, would assign a
disproportionate number of steps to the highlight values—the top range
of 512–1024 would define only one f-stop, while leaving 0–512 to define
all the rest. A logarithmic representation of code numbers, however, gives
equal representation across the full dynamic range of film negative—in
10-bit space, 90 code values for each f-stop. This allows for more preci-
sion to define shadow detail. For this reason, 10-bit log is the standard for
| 403
recording digital images back to film. Some publishing color applications

use a 16-bit system for even more control and detail. The cost of the ad-
ditional bits is disk space, memory, bandwidth and processing time.
The consequence of too few bits can be artifacts—flaws in the image
introduced by some image processing. Some artifacts include banding,
where a smooth gradient is interrupted by artificial lines, and quantiza-
tion, where a region of an image is distorted. If image data is recorded or
scanned in 10-bit color, converted to an 8-bit format for postprocessing,
then converted back to 10-bits for film recording, image information (and
usually quality) is lost and cannot be retrieved. Whenever possible, it is
preferable to maintain all of the image data and not discard information
through conversion to a more limited format.
BLACK STRETCH – Flattens the curve of the toe to put more steps or val-
ues in the lower part of the tone scale. Reveals more shadow detail.
BYTE (B) – A digital “word” composed of 8 bits which can describe 256
discrete values.
• kB = Kilobyte (100 bytes)
• MB = Megabyte (1,000 bytes)
• GB = Gigabyte (1 million bytes)
• TB = Terabyte (1 billion bytes)
CALIBRATION – The adjustment of a display device, such as a monitor or
projector, that prepares a device to produce accurate, predictable and consis-
tent results in accordance with a standard. Calibration is an essential factor
in color canagement and look management. It enables two people, in two
different locations, to look at the same digital image and see the same look.
CCD – (charge-coupled device) A type of light-sensitive chip used in cam-
eras for image gathering, either as a single chip or in a three-chip array.
Basically a grayscale device that measures light intensity, a CCD records
color in a single-chip camera by using color filters on each photoreceptor
site (see Bayer pattern). CCDs are analog sensors that convert light inten-
sity into voltage. An A/D converter then transforms the analog voltage
value to digital data.
CHIP – A device aligned behind the lens that contains an array of pho-
toreceptor sites, or light-sensitive sensors, for capturing image data. The
sensors convert light intensity into voltage levels, an electrical charge
proportional to the intensity of the light striking it. The voltage is then
sampled and converted to a digital code value—one for each pixel of the
image.
CHROMATICITY DIAGRAM – The two-dimensional plot of visible light,
commonly used to represent the three-dimensional 1931 CIE XYZ color-
space standard. It is effectively a 2-D “slice” out of a 3-D model. The horse-
shoe-shaped plot defines the range of human color vision, independent
404
|
of brightness. More limited color gamuts (ranges) of media (such as film

or HD video) or display devices can be clearly portrayed in the diagram.
Individual colors are defined as points in the diagram with chromaticity
coordinates (x, y). Color saturation is greatest at the outer rim of the color
plot; the neutal white point area is near the center of the plot.
CHROMATICITY COORDINATES – (x, y) Chromaticity coordinates de-
fine specific color values in the CIE XYZ color space. They describe a color
quality (hue and saturation only) independent of luminance (brightness
value).
CIE XYZ COLOR SPACE – The color space originally defined in 1931
by the Commission Internationale de L’Eclairage. (The CIE standard
has been updated, and is currently being re-evaluated.) CIE XYZ uses a
.9
.8
.7
.6
.5
2000º K
.4 rve
Cu
dy
B o 4000º K 1000º K
k
ac
Bl
6000º K
.3
10000º K
.2
.1
.0
.0 .1 .2 .3 .4 .5 .6 .7 .8
Figure 1. Triangles within CIE chromaticity chart define limits of color gamuts
| 405
three-dimensional model to contain all visible colors. It defines a set of

three primaries (specific red, green and blue colors) and a color gamut to
describe the range of human vision.
CINEON FILE – a 10-bit log image file format based on film density. De-
signed to capture the full dynamic range of the motion picture negative, it
is an industry standard for film scanning and recording.
CLIPPING – The total loss of image detail at either end of the scale—high-
light or shadow. Clipping occurs when the exposure moves beyond the
threshold determined by the capability of a camera or recording device.
On a waveform monitor the exposure thresholds for clipping are usually
0 and 100 IRE.
CMOS (Complementary High-density Metal Oxide Semiconductor) – A
category of chip with an array of photoreceptor sites. Its integrated cir-
cuitry has both digital and analog circuits. Unlike CCDs, where the charge
is actually transported across the chip and read at one corner of the array,
CMOS chips have transistors at each pixel that amplify and move the
charge using traditional wires. This approach is more flexible because each
pixel can be read individually; however, CMOS chips are less sensitive to
light because many of the photons hit the transistors instead of the photo-
diode. The distinguishing characteristic of the CMOS chip is that it allows
a large number of camera processing functions while using less power and
generating less heat than other types of chips. It can perform at various
frame rates and perform either progressive or interlaced scans. The Arri
D-20 uses a CMOS chip, while the Panavison Genesis uses a CCD chip.
CMY (cyan, magenta, yellow) —The additive secondary colors. Combining
two additive primary colors (R, G, B) produces an Additive Secondary
color.
CODEC – A compression/decompression algorithm. The software that ex-
ecutes the compression and decompression of an image.
CODE VALUE – The digital representation of the brightness and color of
a pixel.
COLOR DECISION LIST (ASC-CDL) – Designed by the ASC as the begin-
ning of a universal color-grading language, the CDL is a standardized lim-
ited expression of how an original image has been graded in digital post.
Currently addressing only lift, gamma and gain parameters, the CDL can
facilitate working in multiple facilities on the same production, and allow
basic color-correction data to be interchanged between color-correction
systems from different manufacturers. Within the areas addressed, using
color-correction systems that accommodate the CDL code, the CDL will
process an original image and produce the same results on different sys-
tems and in different facilities. It is a useful tool for the cinematographer
in look management.
406
|
COLOR GAMUT – The range of colors a system can display or record.

When a specific color cannot be accurately displayed in a system, the color
is considered “out of gamut” for that system. The standard HD video color
space, defined by ITU-R BT.709 (aka. Rec. 709), is an example of a device-
dependent color space. It uses a YCrCb space, which separates luminance
(Y) and chrominance (color – described by Cr and Cb), and allows for
color subsampling, another means of compression.
One method to compare media and devices is to plot their color
gamuts on the chromaticity diagram. The CIE XYZ color space is the
standard device-independent reference. Although the color space uses a
three-dimensional model to represent its colors, it plots the visible colors
on a flat chromaticity diagram, actually a 2-D “slice” taken of the 3-D
model, usually represented as a colored horseshoe-shaped area, leaning
left on a 2-dimensional x-y axis. All colors visible to human perception are
plotted on the graph. The colors of the spectrum lie along the horseshoe
curve, left to right, blue to red. Red, green and blue primaries are specified.
The white point is nominally located where the three primaries are equal
in contribution, but to accommodate different color temperatures, their
respective white points are plotted along a curve across the center area
of the model. Specific colors are identified in the 2-D space by a set of
two numbers, called chromaticity coordinates (x, y), which specify their
position on the diagram.
COLOR MANAGEMENT – Maintaining the accuracy and consistency of
color throughout the digital workflow from capture to final release.
COLOR SAMPLING – Describes the precision of the measurement of
light and color by a camera system. It is represented by three numbers,
separated by colons, and refers to the relative frequency of measurement
(sampling) of color values. The first number represents luminance (Y,
commonly associated with green) and the second two numbers represent
chroma or color as red and blue difference values (Cr, Cb).
4:4:4 captures the most information, sampling color at the same
frequency as luminance. 4:2:2 is the current standard on HD production
cameras. The color information is sampled at half the frequency of the
luminance information. The color precision is lower, but is adequate in
most production situations. (Human vision is more sensitive to brightness
changes than it is to color variation.) Problems can arise, however, in post-
processing, such as in the compositing of greenscreen scenes, where color
precision is important. It is recommended that any visual effects shots that
require image processing be recorded in a 4:4:4 format.
Sampling can also be understood as a type of compression of image
data: 4:4:4 sampling is uncompressed, 4:2:2 sampling compresses color
values by a factor of 2, 4:1:1 compresses color 4:1.
| 407
Figure 2. 4:2:2 color sampling Figure 3. 4:4:4 color sampling
4:1:1 4:2:0 4:2:2 4:4:4

COLOR SPACE – A color space is a structure that defines colors and the re-
lationships between colors. Most color-space systems use a three-dimen-
sional model to describe these relationships. Each color space designates
unique primary colors—usually three (a defined red, green and blue)—on
which the system is based. The system can be device-dependent or device-
independent. A device-dependent system is limited to the color represen-
tation of a particular device or process, such as film. The RGB color space
that describes film is representative of the range of colors created by the
color dyes used in film. A device-independent color space defines colors
universally, which through a conversion process using a profile, can be
used to define colors in any medium on any device.
The color space designated as the standard for digital-cinema distri-
bution is X'Y'Z,' a gamma-encoded version of the CIE XYZ space. A de-
vice-independent space, X'Y'Z' has a larger color gamut than other color
spaces, going beyond the limits of human perception. All other physically
realizable gamuts fit into this space.
COMPONENT VIDEO SIGNAL – A video format in which luminance and
chrominance remain as separate entities and have separate values.
COMPOSITE IMAGE – An image created by the combination of two or
more source images. Compositing tools include matte-creation software,
and chroma-key hardware.
COMPOSITE VIDEO SIGNAL – A video format in which luminance and
chrominance are combined.
COMPRESSION – Compression is the process used to reduce the size of
an image file with the least sacrifice in quality or usable information. One
408
|
method used by compression schemes is to analyze image material, and

find and remove any redundancy, such as blue sky with a consistent color.
Compression reduces the size of the image file and as a result lowers the
data rate (and hardware) required to display the file in real time. Com-
pression can be interframe or intraframe.
In some compression schemes, the original image can be fully recon-
stituted. Such a scheme uses “lossless” compression. However, a danger
exists with certain compression codecs that claim to be “visually loss-
less.” The claim may be true for the direct display of an original image
that requires no manipulation in postproduction, but if it later needs to
be processed for significant color grading or visual effects, disturbing
artifacts may appear that significantly reduce the quality of the image.
Other codecs discard image information, which cannot be subsequently
retrieved. This is considered “lossey” compression.
COMPRESSION RATIO (uncompressed to compressed) – The ratio that
compares the image file size of the original noncompressed image to the
compressed version. The lower the ratio, the lower the compression, and
usually the higher the quality of the resulting image. A format that uses
a 2:1 compression ratio preserves more image information than one that
uses a 4:1 compression ratio.
CRT (cathode-ray tube) – The traditional TV monitor. Cathode rays are
streams of electrons produced by the heating of cathode inside a vacuum
tube. The electrons form a beam within the cathode-ray tube due to the
voltage difference created by two electrodes. The image on the screen is
formed by the manipulation of the direction of this beam by an electro-
magnetic field that sweeps across the surface of the phosphorescent
screen. Light is emitted as the electrons strike the surface of the screen.
CRUSHING THE BLACKS – Raises the slope of the toe, which reduces the
number of steps or values in the lower part of the scale. Darker shadow
details are lost in black.
DATA – Information stored in a digital file format. Digital image data is
assumed to be raw, unprocessed and uncompressed, retaining all original
information, as opposed to video, which has been processed and com-
pressed, thereby losing some information.
DCDM (digital cinema distribution master) – The distribution standard
proposed by DCI: uncompressed and unencrypted files that represent
moving image content optimized for direct electronic playback on a
projector and sound system in a theater. The DCDM’s files contain the
master set of raw files that produce high-resolution images, audio, timed
text (subtitles or captions) and metadata—auxiliary data that can include
data to control room lights and curtains.
DCI (Digital Cinema Initiative) – DCI was founded in March 2002 as a joint
| 409
venture of Disney, Fox, MGM, Paramount, Sony, Universal and Warner

Bros. DCI’s purpose was to create uniform specifications for digital cin-
ema projection, and establish voluntary specifications for an open archi-
tecture for digital cinema to insure a uniform and high level of technical
performance, reliability and quality control.
Some of the specs in the standard: 12-bit, 4K or 2K, JPEG 2000. The
DCI system specification defines a life cycle in which content exists in a
succession of states:
• DSM – Content originates as a digital source master; format not specified.
• DCDM – Content is encoded into a digital cinema distribution master,
according to the specification (12 bit, 4K or 2K, X'Y'Z' color space).
• DCP – Content is compressed and encrypted for transport to the theater
as a digital cinema package, covered by the specification (JPEG 2000).
• DCDM (again) – Content is unpackaged, decrypted and decompressed
at the theater for exhibition.
DCP (digital cinema package) – The compressed, encrypted DCDM (digital
cinema distribution master) When the DCP arrives at the theater it is un-
packaged or decoded—decrypted and decompressed—to make a DCDM
ready for display or projection.
DEAD (dark or bright) PIXEL – A pixel that is either permanently bright
(either white, red, blue or green) or dark (black) on the screen. It can occur
in either recording devices (as cameras) or display devices (as monitors or
projectors). Some cameras have systems to correct for dead pixels, filling
in the space with values from adjacent sensors.
DEVICE-DEPENDENT COLOR SPACE – A color space that is limited
by the performance of a particular device or process, such as film. Film
reproduces a particular range of color (color gamut). The RGB color space
that describes film is representative of the gamut of the color dyes used in
film. It is only accurate for describing colors in the film medium.
DEVICE INDEPENDENT COLOR SPACE – A color space that defines
colors universally. Through a process of conversion or translation specific
to the application, medium or device being used, its values can be used to
specify colors in any medium on any display device.
DIGITAL CONFORM – The process of editing the full-resolution digital
image material to match the EDL (edit decision list) which was created in
the editorial process.
DIGITAL INTERMEDIATE – The conversion of a film image into digital
media for all postproduction processes, including the final color grading,
as an intermediate step for distribution, whether by film or digital media.
Common usages of the term:
• The entire digital postproduction process that comes between principal
photography and the production of the final distribution master.
410
|
• The digital version (image files) of the production, used during post
production.
• The final colorgrading process (which may take weeks for a feature)
that integrates all visual elements and determines the final look of the
production.
• The product or result of the final colorgrading process (digital master).
DIGITAL MASTER – The completed output or result of the final color-
grading process, subsequently used to create the distribution masters. Or,
the full digital version of the camera-original production.
DIGITAL SIGNAL – A signal represented by discrete units (code values),
measured by a sampling process. A digital signal is defined by a finite code
Figure 5a. Camera Original Figure 5b. Graded Master
value in accordance with a predetermined scale.

DISCRETE – Discrete values use a finite scale of whole integers.
DLP (Digital Light Processing) – Texas Instruments’ micromirror technol-
ogy is a display solution that manipulates light digitally. It uses rectangu-
lar arrays of up to two million hinge-mounted microscopic mirrors that
reflect light onto the screen. In a three-chip DLP system, light generated
by the lamp passes through a prism, which divides it into red, green and
blue streams, each aimed at a DLP mirror chip assigned to that color. The
digital signal operates the hinges of the mirrors, controlling the degree
of reflection of each mirror. The colored light reflects selectively off the
micromirrors, combines into a single stream, and passes through the pro-
jection lens to project an image.
DPX FILE – A 10-bit log image file format. The current standard format
for recording digital image data back to film, it is the digital input format
required by the ARRI laser recorder. It is a versatile file format used for
the processing and storage of resolution-independent, pixel-based (bit-
mapped) images.
DYNAMIC RANGE – The range of light values from black to white that a
medium or device can record or display.
EDIT DECISION LIST (EDL) – A frame-accurate description of every cut
in a production, including digital effects and titles.
| 411
ENCODING – The recording of information into a particular digital for-

mat, or conversion of digital data from one format to another.
FIELD – An image composed of either the even or odd scan lines of an inter-
laced video frame. Two fields displayed sequentially make up a video frame.
FILE FORMAT – The software structure that records digital data. The file
format determines the organization of data within the file. File formats
often determine the spacial resolution and color depth of an image.
FILE-FORMAT CONVERSION – The transformation of an image file from
one format to another. A conversion may be required at various steps in
the workflow, depending on the devices and processes used. Conversion
is usually required for distribution on different media. Depending upon
the resolution and color parameters of a particular format, resolution and
color information can be lost in a file-format conversion. Referred to as
downconverting, image information is lost when converting to a smaller,
more limited format.
FLASH MEMORY – A storage scheme that records data on a chip and re-
tains that data when the host system has been shut off. Secure digital and
compact flash memory cards use flash memory. The Panasonic P2 system
uses flash memory.
FLOATING POINT – A numbering system that uses “real” numbers, rather
than integers. The decimal point of a real number floats, or moves, ac-
cording to its exponent component. A floating-point number consists of
two parts: the mantissa, which is a decimal value between 0 and 1, and a
multiplier of 10n. For example, <.123456 x 103> is a floating point number
with a value of 123.456.
This system is capable of defining a much wider range of values than
a fixed-point integer with the same number of bits. When used to create
digital image files, the greater range afforded by floating-point numbers
allows for finer shadow detail (small values) while still accommodating
the many stops (large values) required to display highlight detail. It creates
the capability to store, process and display images closer to a continuous
tone scale, reduce artifacts, and produce a more faithful rendition of the
original photographed scene.
FRINGING (COLOR) – A digital artifact resulting from errors in compos-
iting one image into another. An imperfect matte can cause a foreground
element to acquire an exaggerated (usually bright) color outline.
FTP – (file transfer protocol) A language used for file transfer from com-
puter to computer on the Internet.
GAIN – The process of manipulating a digital image by holding the black
value constant and raising or lowering the value that creates white, af-
fecting the apparent brightness of an image. On a video camera, the gain
control adjusts the CCD sensor’s sensitivity to light.
412
|
GAMMA – The slope or contrast of an image. In color-correction terms

gamma refers to the contrast affecting only the midtones of an image.
Adjusting the gamma changes the brightness of an image without dra-
matically changing the shadows and highlights. Film dyes have a log-
arithmic response to exposure. This is why the traditional sensitometric
curve, used to describe the behavior of film stocks, plots density as the y
coordinate and log exposure as the x coordinate. The resulting S curve,
with a lower slope in both the toe and shoulder, provides more informa-
tion about shadow and highlight detail. A logarithmic representation of
light values is typically used when working with scanned film material for
digital intermediate or visual-effects work. Logarithmic code values are an
important characteristic of what is considered film color space.
GAMMA CORRECTION – The gamma adjustment applied to a video
camera’s signal to allow correct display of the video image on a CRT
monitor. This is necessary because of the nonlinear response, or power
function, of the CRT monitor. Cameras that can record raw data do not
apply this gamma correction, and avoid the loss of information caused by
a gamma correction.
GRADING – The process of color correction, enhancement and manipula-
tion used to alter a recorded image. A tool in look management, a key
objective of grading is to maintain a consistent look throughout the entire
workflow of a production.
HDRI (high dynamic range image) – An image that records pixel values
spanning the full dynamic tonal range of a real-world scene. An HDRI
can encode the high contrast ratios that occur in nature—a 100,000:1 ratio
is common in the natural world. The values of an HDRI image are linear
and remain proportional to actual light values. An HDR image must be
encoded in a format that allows a large range of values, such as one that
uses floating-point values stored with 32 bits per color channel.
HEADER – The location in a digital image file where metadata (any relevant
information, such as frame number and image size) can be recorded.
HIGH DEFINITION (HD) – A video format consisting of either 720 ac-
tive lines of progressive video or 1080 active lines of either progressive or
interlaced video.
The standard HD frame is 1920 x 1080 or 1280 x 720. The aspect ratio
is 16:9, or 1.78:1.
HISTOGRAM – A graph that displays the distribution of light values in a
scene. Basically a bar chart, a histogram indicates the proportion of image
area (y axis) occupied by each level of brightness from 0–100 IRE, black to
white (x axis). With a clear indicator of clipping at either end of the scale,
it is useful for determining whether the values in a scene fit within the
camera system’s dynamic range.
| 413
HUE – A specific color in the spectrum, independent of saturation or lu-

minance.
HYBRID FILM/DIGITAL WORKFLOW – A production workflow that in-
tegrates digital technology into the traditional film workflow. A system that
mixes traditional silver-halide technologies with digital imaging technologies.
IMAGE FILE – A recording of a single image frame in a particular file format.
IMAGE PROCESSING – The process of altering a digital image with soft-
ware and hardware. Examples include color grading, compression and
color-space conversion.
INTERFRAME CODEC – The interframe codec uses series of frames, or
groups of pictures (GOP), to compress the image information. Interframe
compression compares consecutive frames to remove redundancy or
common elements and arrive at “difference” information. Although it’s
much more efficient and can achieve greater compression, this scheme
creates significant challenges in editing and postprocessing due to the
interframe dependency.
INTERLACE SCANNING – Since the beginning of television broadcast-
ing, interlaced scanning has been used. Two vertical sweeps are made for
each frame—one to scan the odd lines and the other to scan the even lines.
Each individual sweep is called a field.
The downside of interlaced scanning is a loss of resolution and steadi-
ness when shooting movement. With subject or camera movement, the
image changes from one field to the next. Two consecutive fields must
be perceptually merged to create each complete frame, even though the
moving subject has changed position. Therefore, any movement will ap-
pear blurred, and detail will be substantially reduced in areas of the frame
that contain movement. This factor also creates difficulty for any image
processing that involves spatial manipulation of the image.
A film release of interlaced material requires the creation of progres-
sive frames from pairs of fields, so as to convert the interlaced video fields
back to progressive film frames. There are various methods of doing this
video-field-to-film-frame conversion. Some involve the interpolation of
pixel values between the different fields. This merging of field pairs into
frames reduces resolution wherever there is movement. For this reason,
the progressive format is preferable for recording back to film.
INTERPOLATION – Connecting the dots or estimating in-between val-
ues. A process of creating new data based on a set of existing data, such
as inserting new material between existing material; estimating missing
values by taking an average of known values at neighboring points; an
estimation of a value between two known values in a sequence of values.
Interpolation can be used by software to increase apparent resolution by
adding pixels based on the value of surrounding pixels.
414
|
INTRAFRAME CODEC – A compression scheme that processes each

frame individually. It is limited to removing redundant information
from within each frame. Generally the result is higher quality, but less-
compressed than interframe compression.
ITU (International Telecommunications Union) – An international broad-
cast-standards committee that replaced the CCIR.
ITU-R BT. 601.5 – Commonly referred to as Rec. 601. The standard for
standard-definition television. The international standard for component
digital television from which the SMPTE 125 M standard was derived. This
ITU recommendation defines the sampling systems, matrix values and filter
characteristics for both Y/ B-Y/R-Y and RGB component digital television.
LASER RECORDER – A device used to record digital image data onto
motion-picture film. It uses red, green and blue lasers to write directly to
each color layer of the film. The brightness of the lasers allow the use of
fine grain, low sensitivity intermediate stock.
LCD (liquid crystal display) – Used for the viewfinder of most digital cam-
eras.
LIFT – The process of altering an image by holding the white point constant
and raising or lowering the value that creates full black, affecting shadow
detail. The slope of the transfer function is changed, affecting the contrast
of the image.
LINEAR CODE VALUES – A linear representation of light assigns code
values directly proportionate to the intensity of the light. Double the light
intensity (as from 100 to 200 foot candles) and the linear code value dou-
bles. The ISO/ASA measurement scale is linear. The disadvantage of using
a linear scale in digital photography is that a disproportionate number
of code values end up assigned to the highlight values. In a 10-bit linear
scale of 0–1023, the top range of 512–1023 defines only one f-stop, leaving
0–512 for all the rest of the f-stops.
LINEARITY – The degree to which the input of a signal is proportional to
the output.
LOGARITHMIC CODE VALUES – The system that uses a logarithmic
formula (log2) to encode light values. The logarithmic representation of
light assigns an equal number of code values to each f-stop. This scale al-
lows more precision in defining values, particularly in the toe region, and
is consistent with the photographic f-stop scale, zone system and human
visual perception.
LOOK – The look expresses the visual tone of a project, supports the direc-
tor’s vision and integrates the contributions of the creative members of
the team.
The cinematographer develops and maintains the look throughout
the production.
| 415
LOOK MANAGEMENT – The process of creating, communicating, pro-

tecting and maintaining the agreed-upon look of the picture throughout
the digital workflow.
LOSSLESS COMPRESSION – A compression scheme in which all image
information can be reconstituted when the image is decompressed.
LOSSY COMPRESSION – A compression scheme in which discarded
image information is lost, and cannot be retrieved.
LUMINANCE – Lightness or brightness. Measured as the average of the
brightest and darkest of the three individual RGB channels that define the
value of a pixel.
LUT (look-up table) – A software tool for manipulating the color of an image;
a method of mapping input colors to output colors. Used in color man-
agement and look management to create a particular look or convert an
image to be read or display properly on a device.
MPEG (Motion Pictures Expert Group) – A digital compression standard.
METADATA – All data associated with a specific image, recorded in the
digital image file header (or alternatively, in an external database), to fa-
cilitate some aspect of the processing the image to realize the project.
MOIRÉ PATTERN – Usually seen as a wavy, vibrating pattern in an image,
a moiré pattern is an artifact that appears when one repeating pattern in-
terferes with another repeating pattern. It often occurs in scenes with fine
repetitive detail, such as on a herringbone fabric design or an old screen
door. The moiré pattern appears when the level of detail (frequency) in the
scene exceeds the recording capability (resolution) of the recording device
(camera or scanner).
MOTION-CONTROL DATA – Metadata documenting camera placement
and movement, including lens height, position, pan and tilt data, and all
pertinent camera information.
NOISE – Noise appears like randomly placed “snow” or film grain in an
electronic image. Caused by electronic error or signal interference in a
camera, noise can represent irregular level fluctuations of a low order of
magnitude. Some noise is present in all analog video signals. The level of
noise is a function of the efficiency of the sensors (such as CCD or CMOS)
and signal processing of the camera. Noise levels can vary by color chan-
nel when sensors are more sensitive to particular colors.
Digital signal compression is another common source of noise, where
it can appear as high-frequency signal fluctuation.
NONLINEAR EDITING – A system that allows random access to all images
in a database. An edited sequence can be rearranged without re-recording
the images from the beginning of the sequence, as a linear system requires.
NTSC (National Television Systems Committee) – The color television stan-
dard format that uses 525 lines, 30 interlaced frames (60 fields) per second.
416
|
OPEN EXR – An image file format capable of recording extremely high

dynamic range images (HDRI) far beyond the capability of human per-
ception. Developed by ILM, it stores data as 16- or 32-bit floating-point
values. It is also capable of recording metadata, such as camera position
and color balance information. It supports several lossless compres-
sion schemes and is extensible, capable of integrating new compression
schemes. Open EXR files record images as scene-referred data, document-
ing the actual light values in the physical scene that was photographed. It
is a good candidate for archival image storage.
OVERSAMPLING – The process of using multiple measurements of a
particular value to arrive at a more accurate, higher quality result. An
example is the use on a chip of multiple photoreceptors to define each
individual pixel.
PANALOG – A transfer curve used in software developed by Panavision that
extends the dynamic range of their Genesis camera by up to 3 stops. The
curve transforms the camera’s internal 14-bit linear digital signal (with a
range of 0–16,383) to a 10-bit “quasi-log” signal (with a range of 64–1,019).
Designed to capture and preserve the entire range of the camera’s sig-
nal, the Panalog scale creates a perceptually equal grayscale in which each
stop increment is expressed by the same number of code values, providing
more values in the darker regions of an image than the linear scale provides.
The white point is placed at 680 on the log scale (about 70% up the 10-bit
scale), allowing for more highlight detail, and helping to avoid clipping.
PERSISTENCE OF VISION – The perceptual process of the brain or the
retina that retains an image for a period of time. Motion blur in each
image helps create the illusion of motion which results when a series of
images are displayed in quick succession.
PEDESTAL – The light level at which a pixel turns black. The pedestal can
be raised or lowered to change the level at which an area of an image turns
black, thereby crushing to black or increasing shadow detail.
PHOTOSHOP – A software program that can be used to grade and ma-
nipulate digital images.
PIXEL – Short for “picture element,” a pixel is the building block of the
digital image. It represents one sample of picture information. Pixels are
grouped into fixed arrays of straight rows and columns. The pixels remain
in the same position frame to frame. All pixels in an array are the same
size, usually square, but in some cases, rectangular.
POWER FUNCTION – The nonlinear gamma of a CRT. In order to prop-
erly view video on a CRT, a nonlinear gamma correction has to be applied
to the video.
PRIMARIES – The specific colors that define a color space or are specified
to calibrate a display device. Most systems today use three primaries—a
| 417
specified red, green and blue. But some systems are now experimenting
with the use of more than three primaries in order to improve color display.
PRIMARY COLOR CORRECTION – Independently adjusting the lift, gain
and gamma of the three additive color channels (red, green and blue).
PROGRESSIVE – The method of sequentially scanning all lines of a pixel
array to create a frame in one vertical sweep. There are no fields as there
are in interlaced scanning—only complete frames. Computer screens use
progressive scans.
QUANTIZATION ARTIFACT – Where a region of an image is distorted as
a result of some form of image processing. Loss of image information can
result from reduction in color space, bit depth, color sampling or spatial
resolution. A common example: downconverting an image from 10-bit
color to 8-bit color.
RAID ARRAY (Redundant Array of Independent Disks) – A group of
hard-disk drives, linked by a system that enhances speed, increases storage
capacity and performs automatic backup of data. A RAID array writes to
multiple disks simultaneously and backs up data in the process. Often used
in film production, the benefits of the RAID system include increases in
data integrity, fault tolerance, throughput, speed and storage capacity. There
are different RAID levels, including 0 (no redundancy or backup capability)
and 5 (most common, with redundancy) through level 10 and “RAID S.”
RAW FILES – The digital recording of images without applying the gamma
correction of video. This unprocessed data retains more information and
can potentially produce a higher-quality image. Instead of applying the
video gamma correction, they convert the linear code values it captures
into logarithmic code values which approximate a film gamma and color
space, resulting in a wider dynamic range and greater shadow detail.
REDUNDANCY – Values that are repeated in an image.
Compression codecs find and compress redundant values in an image
in order to reduce the size of the file. An example of visual redundancy is
a blue sky with consistent color.
REGION-OF-INTEREST CORRECTION – Using “windows” or articu-
lated mattes to isolate specific areas of the frame to make alterations.
REGISTRATION (COLOR) – The alignment of the three-color sensors
or sources (red, green, blue) on a three-chip recording or display device.
Poor alignment can create color ringing artifacts. Registration charts can
be used to insure that the camera is in precise registration.
RENDER – The process of re-recording a digital image in accordance with
certain specifications or adjustments.
RESOLUTION – The measure of the finest detail visible in an image. The
resolution of an image is measured by the number of pixels in the image
display, not by the number of sensors in the camera chip.
418
|
Camera resolution is commonly referred to by the number of lines of

pixels (scan lines) it records. An increase in the pixel line count will pro-
duce a proportional increase in resolution and representation of fine detail.
Image resolution is expressed by two numbers: (pixels per line) x
(number of lines). A doubling of pixel lines and pixels per line (as the
change from 2K to 4K), increases the total pixel count by a factor of 4,
requiring four times the memory to store the image data. However, the
MTF (an optical measure of line pair resolution) only doubles.
Some of the most common resolution standards:
Standard Definition Analog NTSC = 640 x 480
(1.33:1)
High Definition = 1920x1080 or 1280x720
(1.78:1)
2K = 2048 pixels per line
(line count varies with aspect ratio)
4K = 4096 pixels per line
Pixel count is not the only element in the imaging chain that affects
picture resolution. Lens quality, registration of multiple chips in a three-
chip camera and image scaling conversions also affect image resolution.
RGB (Red, Green, Blue) – The additive primary colors. Combining all three
primary lights in equal intensities produces white. The RGB color space is
widely used to encode scanned film material.
RINGING – An artifact that appears like an outline or halo on the edge be-
tween bright and dark areas of an image. An exaggerated transition often
caused by excessive sharpening.
SCAN – The process of reading and converting an analog image into a digi-
tal file.
SCENE-REFERRED DATA – Device-independent data that documents the
light values in the physical scene that is photographed. Most image data
is output-referred data that depends on the limited dynamic range and
gamut of a particular output device or display. Scene-referred data does
not have these limits, but can be tone-mapped or converted for display on
any output system or device—including systems yet invented.
SECONDARY COLOR CORRECTION – Isolating specific colors in the
spectrum, then modifying their hue, saturation and brightness.
SENSOR – A photoreceptor site on a chip. The two most common types
are known as CCD (charge coupled device) and CMOS (complementary
metal oxide semiconductor).
SOFT CLIP – Software functions used to compress values as they approach
the top of the scale. This effect keeps some detail in the bright areas, or at
least softens the edge where the clipping begins.
SPACIAL CORRECTIONS – Examples include: frame repositioning, digital
| 419
filter blur, flaring, sharpening and adding grain or noise.

SPACIAL RESOLUTION – Actual detail recorded. Sharpness and contrast,
dynamic range of exposure, highlight detail and shadow detail; the accu-
racy, subtlety and character of color; the smooth portrayal of motion in-
cluding the capability to alter speed or frame rate and the degree to which
you will be able to manipulate the
image in post.
STANDARD DEFINITION – Refers
to an NTSC, PAL or SECAM video
format. NTSC uses of 480 active
lines of interlaced video.
SUBTRACTIVE COLOR – Primaries:
cyan, magenta, yellow—the comple-
ments of the additive primaries: red, Single-chip camera
green and blue. Used as filters that

absorb discrete color values, the Figure 6. Single-chip camera – A chip design
three subtractive colors will absorb in which an array of filtered photoreceptor
sites on a single chip record a full-color im-
all color, resulting in black. age. Proprietary chip designs have different
TC – Time code. designs of color-filtered sensors.
TIFF FILE (tagged image file format)
– An industry standard raster file format developed for universal transfer
between many digital imaging applications and devices.
TONAL MAPPING – Converting the tonal (brightness and color) values of
an image from one system to another with a different dynamic range and
color gamut. For example, converting a high dynamic range 12-bit image
to a much more limited 8-bit format that can be displayed on a standard
CRT monitor. The challenge of tonal mapping is to preserve the quality
and character of the original image on the more limited system.
VECTORSCOPE – A vectorscope is an oscilloscope that measures the ac-
curacy of color recorded by a camera and facilitates the adjustment of the
camera to record accurate color. It does not measure the brightness or lu-
minance of a signal. The vectorscope screen has a marked scale (graticule)
inscribed with a circle marked in degrees (0–360), a center crosshair, and
six individual boxes distributed around the circle (one for each primary
and secondary color).
The direction of the “vector” (measured by degrees) pointing from
the center to a point on the circle, indicates the hue (phase) of a color in
the scene. The length of the vector, starting at the center and extending
toward the circle, indicates the saturation (amplitude) of the color. The
center point is considered 0 saturation or neutral (as gray, black or white).
The color boxes correspond to the SMPTE/NTSC specifications for
the hue and saturation of that particular color in the standard color-bar
420
|
Bayer Pattern 2x2 Matrix
RGB Stripe Pattern 2x3 Matrix

Figure 7. Single-chip systems using a color filer array mosaic
test signal. 0 degrees is at the 9 o’clock position. The color boxes, clockwise
from 0 degrees, are labeled YL (yellow), R (red), MG (magenta), B (blue),
CY (cyan) and G (green).
VFX (Visual-effects material) – Plates, composites, CGI, etc.
VISUALLY LOSSLESS – A compression scheme that claims to be lossless
and may appear to be lossless without manipulation (or color correction).
However, when it must be manipulated for color grading or visual effects,
artifacts often appear.
WATERMARK – A security scheme in which bits are altered within an
image to create a pattern, which constitutes proof of ownership.
WAVEFORM MONITOR – An oscilloscope adapted to display luminance
or brightness light levels across a scene, from left to right, as recorded by a
camera, as well as other video signal levels. It measures voltage levels from
the camera and displays them on a screen in a scale of 0–100 IRE (black to
white). If the levels flatten out at either end of the scale (at 0 or 100, with
some exceptions), the image will clip and record no detail in those areas.
WHITE POINT – The white reference, expressed in chromaticity coordi-
nates, that determines the white balance, or color temperature, of a system.
An important parameter in color management and device calibration.
Devices such as monitors and projectors can be calibrated to a
| 421
given white point, usually designated by color temperature, such as D65

(6500°K).
WYSIWYG (Wizee-wig: “What You See Is What You Get”) – An acronym
describing the situation in which the image displayed on a monitor closely
resembles the final image output for display.
XYZ COLOR SPACE (and X'Y'Z') – A color space using a linear representa-
tion of light values in a scene. The values can be output-referred (based
on the capability of a particular display system) or scene-referred (based
on the values measured in the physical scene). The <'> indicates that the
values are encoded with a 1/2.6 gamma function. The X'Y'Z' color space
has been designated by DCI as the standard for digital cinema.
YCrCb – the standard HD video color space that converts RGB by encod-
ing luminance (Y) and color-difference signals (Cr, Cb) separately. One
reason for this separation is that the human eye is less sensitive to chro-
minance than luminance. Compression algorithms take advantage of this
phenomenon.
YCM (yellow, cyan, magenta: the subtractive colors) – Refers to the black-
and-white separation film negatives that are created to archive a color
motion picture. Three monochrome negatives are made from the original
color negative, separating out the contribution of each color layer in the
negative. This process was developed to take advantage of the much longer
lifespan of black-and-white film.
422
|
|Ref-1
Further References
3-D
Spottiswood, Raymond, Theory of Stereoscopic Transmission, Berkeley, CA;
University of California Press , 1953.
Wagtendonk, W.J., Principles of Helicopter Flight, Aviation Supplies & Academics,
Newcastle, WA, 1996.
Crane, Dale, Dictionary of Aeronatical Terms, Aviation Supplies & Academics,
Newcastle, WA, 1997.
Spence, Charles, Aeronautical Information Manual and Federal Aviation Regulations,
McGraw-Hill, New York, NY, 2000.
Padfield, R., Learning to Fly Helicopters, McGraw-Hill, New York, NY, 1992.
Industry-Wide Labor-Management Safety Bulletins at: http://www.csatf.org/bulletintro.shtml
Arctic Cinematography
Eastman Kodak Publication: Photography Under Artic Conditions.
Fisher, Bob, “Cliffhanger’s Effects were a Mountainous Task,”
American Cinematographer, Vol. 74, No. 6, pp. 66-74, 1993.
Miles, Hugh, “Filming in Extreme Climactic Conditions,” BKSTS Journal Image Technology,
February 1988.
Moritsugu, Louise, “Crew’s Peak Performance Enhanced Alive,” American Cinematographer,
Vol. 74, No. 6, pp. 78-84, 1993.
Biographies and Interviews

Almendros, ASC, Nestor, A Man With a Camera, New York, Farrar, Straus, Giroux, 1984.
Bitzer, ASC, Billy, Billy Bitzer – His Story: The Autobioagraphy of D.W. Griffith’s Master Cameraman,
New York, Farrar, Straus, Giroux, 1973.
Brown, ASC, Karl, with Brownlow, Kevin, Adventures With D.W. Griffith,
New York, Farrar, Straus, Giroux, 1973.
Cardiff, BSC, Jack, Magic Hour, London, Faber and Faber, 1996.
Challis, BSC, Christopher, Are They Really So Awful?: A Cameraman’s Chronicle,
Paul & Co. Publishing Consortium, 1995.
Clarke, ASC, Charles G., Highlights and Shadows: The Memoirs of a Hollywood Cameraman,
Metuchen, NJ; Scarecrow Press, 1989.
Eyman, Scott, Five American Cinematographers: Interviews with Karl Struss, Joseph Ruttenberg,
James Wong Howe, Linwood Dunn and William H. Clothier, Metuchen, NJ; Scarecrow Press, 1987.
Higham, Charles, Hollywood Cameramen: Sources of Light, Indiana University Press, 1970.
Kalmus, Herbert T. and Eleanor King, Mr. Technicolor, Abescon, NJ; Magic Image Filmbooks, 1993.
Lassally, BSC, Walter, Itinerant Cameraman, London, Murray, 1987.
Ref-2
|
Further References
Laszlo, ASC, Andrew, Every Frame a Rembrandt: Art and Practice of Cinematography,
Boston, MA; Focal Press, 2000.
Laszlo, ASC, Andrew, It’s A Wrap!, Hollywood, CA; ASC Press, 2004.
LoBrutto, Vincent, Principal Photography: Interviews with Feature Film Cinematographers,
Westport, CT; Praeger, 1999.
Maltin, Leonard, The Art of the Cinematographer: A Survey and Interviews with Five Masters,
New York, Dover Publications, 1978.
McCandless, Barbara, New York to Hollywood: The Photography of Karl Struss, ASC,
Albuquerque, NM; University of New Mexico, 1995.
Miller, ASC, Virgil E., Splinters From Hollywood Tripods: Memoirs of a Cameraman, New York,
Exposition Press, 1964.
Rainsberger, Todd, James Wong Howe Cinematographer, San Diego, CA; A.S. Barnes, 1981.
Rogers, Pauline B., More Contemporary Cinematographers on Their Art, Boston, MA;
Focal Press, 2000.
Schaefer, Dennis and Salvato, Larry, Masters of Light: Conversations with Contemporary
Cinematographers, Berkeley, CA; University of California Press, 1985.
Sterling, Anna Kate, Cinematographers on the Art and Craft of Cinematography,
Metuchen, NJ; Scarecrow Press, 1987.
Walker, ASC, Joseph, The Light On Her Face, Hollywood, CA; ASC Press, 1984.
Young, BSC, Freddie, Seventy Light Years: An Autobiography as told to Peter Busby,
London, Faber and Faber, 1999.
Camera
Adams, Ansel, The Camera, New York, Morgan and Morgan, Inc., 1975.
Fauer, ASC, Jon, Arricam Book, Hollywood, CA; ASC Press, 2002.
Fauer, ASC, Jon, Arriflex 16 SR Book, Boston, MA; Focal Press, 1999.
Fauer, ASC, Jon, Arriflex 16 SR3 the Book, Arriflex Corp., 1996.
Fauer, ASC, Jon, Arriflex 35 Book, Boston, MA; Focal Press, 1999.
Fauer, ASC, Jon, Arriflex 435 Book, Arriflex Corp., 2000.
Samuelson, David W., Panaflex Users’ Manual, Boston, MA; Focal Press, 1990
Camera Manufacturers
Aaton, +33 47642 9550, www.aaton.com
ARRI, (818) 841-7070, www.arri.com
Fries Engineering, (818) 252-7700, www.frieseng.com
Ikonoskop AB, +46 8673 6288, info@ikonoskop.com
Panavision, (818) 316-1000, www.panavision.com
Photo-Sonics, (818) 842-2141, www.photosonics.com
Pro8mm, (818) 848-5522, www.pro8mm.com
|Ref-3
Further References
Camera Supports
A + C Ltd., +44 (0) 208-427 5168, www.powerpod.co.uk
Aerocrane, (818) 785-5681, www.aerocrane.com
Akela: Shotmaker, (818) 623-1700, www.shotmaker.com
Aquapod, (818) 999-1411
Chapman/Leonard Studio Equipment, (888) 883-6559, www.chapman-leonard.com
Egripment B.V., +31 (0)2944-253.988, Egripment USA, (818) 787-4295, www.egripment.com
Fx-Motion, +32 (0)24.12.10.12, www.fx-motion.com
Grip Factory Munich (GFM), +49 (0)89 31901 29-0, www.g-f-m.net
Hot Gears, (818) 780-2708, www.hotgears.com
Hydroflex, (310) 301-8187, www.hydroflex.com
Isaia & Company, (818) 752-3104, www.isaia.com
J.L. Fisher, Inc., (818) 846-8366, www.jlfisher.com
Jimmy Fisher Co., (818) 769-2631
Libra, (310) 966-9089
Louma, +33 (0)1 48 13 25 60, www.loumasystems.biz
Megamount, +44 (0)1 932 592 348, www.mega3.tv
Movie Tech A.G., +49 0 89-43 68 913, Movie Tech L.P., (678) 417-6352, www.movietech.de
Nettman Systems International, (818) 623-1661, www.camerasystems.com
Orion Technocrane, +49 171-710-1834, www.technocrane.de
Pace Technologies, (818) 759-7322, www.pacetech.com
Panavision Remote Systems, (818) 316-1080, www.panavision.com
Panther, +49 89 61 39 00 01, www.panther-gmbh.de
Spacecam, (818) 889-6060, www.spacecam.com
Strada, (541) 549-4229, www.stradacranes.com
Straight Shoot’r, (818) 340-9376, www.straightshootr.com
Technovision, (818) 782-9051, www.technovision-global.com
Wescam, (818) 785-9282, www.wescam.com
Cinematography
Brown, Blain, Cinematography, Boston, MA; Focal Press, 2002.
Campbell, Russell, Photographic Theory for the Motion Picture Cameraman,
London, Tantivy Press, 1970.
Campbell, Russell, Practical Motion Picture Photography, London, Tantivy Press, 1970.
Ref-4
|
Further References
Carlson, Verne and Sylvia, Professional Cameraman’s Handbook, 4th edition,

Clarke, ASC, Charles G., Professional Cinematography, Hollywood, CA; ASC Press, 2002.
Cornwell-Clyne, Major Adrian, Color Cinematography, 3rd edition, Chapman Hall LTD 1951.
Malkiewicz, Kris J. and Mullen, M. David, Cinematography: A Guide for Filmmakers and Film
Teachers, New York, Fireside, 2005.
Mascelli, ASC, Joseph V., The 5 C’s of Cinematography, Beverly Hills, CA,
Silman-James Press, 1998 (c1965).
Wilson, Anton, Anton Wilson’s Cinema Workshop, Hollywood, CA; ASC Press, 1983, 1994.
Color
Albers, J., Interaction of Color, New Haven and London; Yale University Press, 1963.
Eastman Kodak Publication H-12, An Introduction to Color, Rochester, 1972.
Eastman Kodak Publication E-74, Color As Seen and Photographed, Rochester, 1972.
Eastman Kodak Publication H-188, Exploring the Color Image, Rochester.
Evans, R. M., An Introduction to Color, New York, NY; John Wiley & Sons, 1948.
Evans, R. M., Eye, Film, and Camera Color Photography, New York, NY; John Wiley & Sons, 1959.
Evans, R. M., The Perception of Color, New York, NY; John Wiley & Sons, 1974.
Friedman, J. S., History of Color Photography, Boston, MA;
American Photographic Publishing Company, 1944.
Hardy, A. C., Handbook of Colorimetry, MIT, Cambridge, MA; Technology Press, 1936.
Hunt, R. W. G., The Reproduction of Colour, Surrey, UK, Fountain Press, 1995.
Itten, J., The Art of Color, New York, Van Nostrand Reinhold, 1973.
National Bureau of Standards Circular 553, The ISCC-NBS Method of Designating
Colors and A Dictionary of Color Names, Washington D. C., 1955.
Optical Society of America, The Science of Color, New York, NY; Thomas Y.
Crowell Company, 1953.
Society of Motion Picture and Teclevision Engineers, Elements of Color in Professional
Motion Pictures, New York, NY, 1957.
Wall, E. J., History of Three-Color Photography, New York and London, Boston, MA;
American Photographic Publishing Company, 1925.
Film
Adams, Ansel, The Negative, New York, Little Brown, 1989.
Adams, Ansel, The Print, New York, Little Brown,1989.
Eastman Kodak Publication H-1: Eastman Professional Motion Picture Films.
Eastman Kodak Publication H-23: The Book of Film Care.
Eastman Kodak Publication H-188: Exploring the Color Image.
|Ref-5
Further References
Eastman Kodak Publication N-17: Infrared Films.

Eastman Kodak Publication: ISO vs EI Speed Ratings.
Eastman Kodak Publication: Ultraviolet and Fluorescence Photography.
Hayball, Laurie White, Advanced Infrared Photography Handbook, Amherst Media, 2001.
Hayball, Laurie White, Infrared Photography Handbook, Amherst Media, 1997.
Film Design
Affron, Charles and Affron, Mirella Jona, Sets in Motion, Rutgers University Press, 1995.
Carrick, Edward, Designing for Films, The Studio LTD and the Studio Publications Inc, 1941, 1947.
Carter, Paul, Backstage Handbook, 3rd edition., Broadway Press, 1994.
Cruickshank, Dan, Sir Banister Fletcher’s A History of Architecture, 20th edition,
New York, NY, Architectural Press, 1996.
Edwards, Betty, Drawing on the Right Side of the Brain, revised edition, Jeremy P. Tarcher, 1989.
de Vries, Jan Vredeman, Perspective, Dover Publications, 1968.
Heisner, Beverly, Studios, McFarland and Co., 1990.
Katz, Stephen D., Shot by Shot – Visualizing from Concept to Screen, Boston, MA;
Focal Press, 1991, pp. 337-356.
Preston, Ward, What an Art Director Does, Silman-James Press, 1994.
Raoul, Bill, Stock Scenery Construction Handbook, 2nd edition, Broadway Press, 1999.
St John Marner, Terrance, Film Design, The Tantivy Press, 1974.
Film History
The American Film Institute Catalog: Feature Films 1911–1920, Berkeley and Los Angeles,
University of California Press, 1989.
The American Film Institute Catalog: Within Our Gates: Ethnicity in American Feature Films
1911–1960, Berkeley and Los Angeles, University of California Press, 1989.
Belton, John, Widescreen Cinema, Cambridge, MA; Harvard University Press, 1992.
Brownlow, Kevin, Hollywood the Pioneers, New York, NY; Alfred A. Knopf, 1979.
Brownlow, Kevin, The Parade’s Gone By, New York, Knopf, 1968.
Coe, Brian, The History of Movie Photography, New York, Zoetrope, 1982.
Ref-6
|
Further References
Fielding, Raymond, A Technological History of Motion Pictures and Television,

Finler, Joel W., The Hollywood Story, New York, Crown, 1988.
Ryan, R.T., A History of Motion Picture Color Technology, London, Focal Press, 1977.
MacGowan, Kenneth, Behind the Screen: the History and Techniques of the Motion Picture,
New York, Delacorte Press, 1965.
Rotha, Paul and Griffith, Richard, The Film Till Now: A Survey of World Cinema,
London, Spring Books, 1967. (New York, Funk & Wagnalls, 1951.)
Schatz, Thomas, The Genius of the System: Hollywood Filmmaking in the Studio Era,
New York, Pantheon, 1988.
Turner, George E., The Cinema of Adventure, Romance and Terror, Hollywood, CA; ASC Press, 1989
Film Processing
ACVL Handbook, Association of Cinema and Video Laboratories.
Case, Dominic, Motion Picture Film Processing, London, Butterworth and Co. Ltd. (Focal Press), 1985.
Eastman Kodak publications: H-1, H-2, H-7, H-17, H-21, H-23, H-24.07, H-26, H-36, H-37, H-37A,
H-44, H-61, H-61A, H-61B, H-61C, H-61D, H-61E, H-61F, H-807 and H-822.
Happe, L. Bernard, Your Film and the Lab, London, Focal Press, 1974.
Kisner, W.I., Control Techniques in Film Processing, New York, SMPTE, 1960.
Ryan, R.T., Principles of Color Sensitometry, New York, SMPTE, 1974.
Filters
Eastman Kodak Publication B-3: Filters.
Harrison, H.K., Mystery of Filters-II, Porterville, CA; Harrison & Harrison, 1981.
Hirschfeld, ASC, Gerald, Image Control, Boston, MA; Focal Press, 1993.
Hypia, Jorma, The Complete Tiffen Filter Manual, AmPhoto, New York, 1981.
Smith, Robb, Tiffen Practical Filter Manual.
Tiffen Manufacturing Corporation Publication T179: Tiffen Photar Filter Glass
Journals, Magazines and Associations

ANSI Standards, American National Standards Institute, www.ansi.org.
American Cinematographer, ASC Holding Corp.,www.cinematographer.com.
BKSTS Journal, “Image Technology,” British Kinematograph, Sound and Television Society,
www.bksts.com.
SMPTE Journal, Society Of Motion Picture and Television Engineers, www.smpte.org.
Lenses
Angenieux, P., “Variable focal length objectives,” U.S. Patent No. 2,847,907, 1958.
Bergstein, L., “General theory of optically compensated varifocal systems,”
JOSA Vol. 48, No. 9, pp. 154-171, 1958.
|Ref-7
Further References
Cook, G.H.,”Recent developments in television optics,” Royal Television Society Journal,

pp. 158-167, 1973.
Cox, Arthur, Photographic Optics, A Modern Approach to the Technique of Definition,
expanded edition, London, Focal Press, 1971.
Kingslake, R. “The development of the zoom lens,” SMPTE Vol. 69, pp. 534-544, 1960.
Mann, A., Ed., “Zoom lenses,” SPIE Milestone Series Vol. MS 85, 1993.
Neil, I.A. and Betensky, E.I, “High performence, wide angle, macro focus, zoom lens for
35mm cinematography,” SPIE Vol. 3482, pp. 213-228, Kona, Hawaii, U.S.A., 1998.
Neil, I.A., “First order principles of zoom optics explained via macro focus conditions of
fixed focal length lenses,” SPIE Vol. 2539, San Diego, California, U.S.A., 1995.
Neil, I.A., “Liquid optics create high performance zoom lens,” Laser Focus World,
Vol. 31, No. 11, 1995.
Neil, I.A., “Uses of special glasses in visual objective lenses,” SPIE Vol. 766, pp. 69-74,
Los Angeles, California, U.S. A., 1987.
Zuegge, H. and Moellr, B., “A complete set of cinematographic zoom lenses and their
fundamental design considerations,” Proceedings of the 22nd Optical Symposium,
pp. 13-16, Tokyo, Japan, 1997.
Lighting
Adams, Ansel, Artificial Light Photography, New York, Morgan and Morgan, Inc., 1956.
Alton, John, Painting With Light, Berkeley and Los Angeles, University of California Press, 1995.
Bergery, Benjamin, Reflections – 21 Cinematographers at Work, Hollywood, CA; ASC Press, 2002.
Box, Harry, Set Lighting Technician’s Handbook, Boston, MA, Focal Press, 2003.
Malkiewicz, Kris J., Film Lighting: Talk with Hollywood’s Cinematographers and Gaffers,
New York, Touchstone, a Division of Simon & Schuster, 2012.
Millerson, Gerald, The Technique of Lighting for Television and Film, Boston, Focal Press, 1991
Miscellaneous
Arnheim, Rudolf, Art and Visual Perception, Berkley, CA, University of California Press, 1974.
Darby, William, Masters of Lens and Light: A Checklist of Major Cinematographers and
Their Feature Films, Metuchen, NJ, Scarecrow Press, 1991.
Houghton, Buck, What a Producer Does, Silman-James Press, 1991.
Kehoe, Vincent J. R., The Technique of the Professional Makeup Artist, Boston, MA,
Focal Press, 1995.
Kepes, Gyorgy, Language of Vision, New York, MA, Dover Publications, 1995.
Moholy-Nagy, L., Vision in Motion, Wisconsin; Cuneo Press, 1997.
Nilsen, Vladimir, The Cinema as a Graphic Art, New York; Garland Pub., 1985.
Waner, John, Hollywood’s Conversion of All Production to Color Using Eastman Color
Professional Motion Picture Films, Newcastle, ME; Tobey Publishing, 2000.
Ref-8
|
Further References
Photography
Evans, R.M., W.T. Hanson Jr., and W.L. Brewer, Principles of Color Photography,
New York, John Wiley & Sons Inc., 1953.
Mees, C.E.K., The Theory of the Photographic Process, New York, Macmillan, 1977.
Thomas Jr., Woodlief, SPSE Handbook of Photographic Science and Engineering,

New York, John Wiley & Sons, 1973.
Woodbury, Walter E., The Encyclopaedic Dictionary of Photography,

New York, The Scovill and Adams Company, 1898.
Traveling Matte Composites

Composite Components Corp. (323) 257-1163, www.digitalgreenscreen.com
Curious Software (gFx roto)

UK: Tel: +44 (0)20 7428 0288 Fax: +44 (0)20 7428 5811
US: Tel: +1 505 988 7243, Fax: +1 505 988 1654
Email: info@curious-software.com
Web: www.curious-software.com
Dazian Theatrical Fabrics: East Coast (877) 232-9426 or East Coast Design Studio (212) 206-3515,
West Coast (877) 432-9426 or West Coast Design Studio (818) 841-6500.
Flo Co (818) 780-0039 or (661) 269-2065, www.flo-co.com
Keylight (650) 326-2656, www.thefoundry.com
Kino Flo (818) 767-6528, www.kinoflo.com
Pinnacle Systems, (Commotion, Primatte) www.pinnaclesys.com
Primatte: Phototron USA, Inc. (530) 677 9980, www.primatte.com or www.phototron.com
Red*D*Mix [Ray McMillan, Flo Co Distributor] (416) 879-3761

email: mcmillan20@cogeco.ca
RFX (compositing software and hardware) (323) 962-7400, www.rfx.com
The Science and Technology Council of the Motion Picture Acadamy

(310) 247 3000, www.oscars.org/council/index.html
Stewart Filmscreen Corp. (310) 784-5300, www.stewartfilm.com
Ultimatte Corp. (818) 993-8007, www.ultimatte.com
Mertens, Lawrence, In Water Photography: Theory and Practice, Wiley Interscience,
New York, John Wiley & Sons, 1970.
Ryan, R.T., Underwater Photographic Applications – Introduction, SMPTE Journal,

Vol. 82, No. 12, December 1973.
Industry-Wide Labor-Management Safety Bulletins at: http://www.csatf.org/bulletintro.shtml

|Ref-9
Further References
Visual Effects
Abbott, ASC, L.B., Special Effects with Wire, Tape and Rubber Bands,
Hollywood, CA; ASC Press, 1984.
Bulleid, H.A.V. (Henry Anthony Vaughan), Special Effects in Cinematography,
London, Fountain Press, 1960.
Clark, Frank P., Special Effects in Motion Pictures Some Methods for Producing Mechanical Effects,
New York, SMPTE, 1966.
Dunn, ASC, Linwood, and Turner, George E., ASC Treasury of Visual Effects,
Hollywood, CA; ASC Press,1983.
Fielding, Raymond, The Technique of Special Effects Cinematography, Boston, MA;
Focal Press, 1985.
Glover, Thomas J., Pocket Ref, Littleton, CO, Sequoia Publishing, 1997.
Harryhausen, Ray, Ray Harryhausen: An Animated Life, New York, NY, Billboards Books, 2004.
Rogers, Pauline B., The Art of Visual Effects: Interviews on the Tools of the Trade,
The Nautical Almanac, commercial edition, Arcata, CA, Paradise Cay Publications (yearly).
Vaz, Matt Cotta and Barron, Craig, The Invisible Art: The Legends of Movie Matte Painting,
San Francisco, CA; Chronicle Books, 2002.
Ref-10
|
Further References
| In-1
Index
INDEX
Page numbers followed by an “f” refer to a figure or illustration
Nos. aperture specifications, 7

Arri accessories, 525
1.85 aspect ratio advantages and disadvantages,
Arri matte boxes, 526
65–69, 66f
ArriCam Lite, 504–506
2 x 2 matrix, 11f
ArriCam Studio, 506–509
2 x 3 matrix, 11f
Arriflex 35-2C, 509–510
2K
Arriflex 35II models, 511
definition, 31n1
Arriflex 35-3, 513–516
digital cinema mastering and distribution
Arriflex 35-3C, 512–513
question, 262–264
Arriflex 35BL, 523–525
2.40 aspect ratio advantages and disadvantages,
Arriflex 235, 516–517
69–73
Arriflex 435 Advanced, 435 Extreme, 517–519
3-D
Arriflex 519–521
polarization filters and, 154–155
Arriflex 535B, 521–523
stereoscopic cinematography, 193–205
Arriflex/SL-35 Mark II, 513
4K
Arriflex/SL-35 Mark III, 511–512
definition, 31n1
Bell & Howell Eyemo Model 71, 529–530
digital cinema mastering and distribution Bell & Howell 35mm Steve’s Cine Service
question, 262–264 conversion, 530-531
16mm cameras, 559–577 Eclair CM-3 16/35mm, 531–532
Aaton A-Minima (Super 16 only), 559–560 Fries 435, 533
Aaton XTRprod, 560–562 Moviecam Comp0act, 533–536
aperture specifications, 8 Moviecam SL, 536–538
Arri fuses for, 571 Moviecam Super America MK2, 538–540
Arriflex 16BL, 563–565 Panavision Gil Golden Panaflex (PFX-GII),
Arriflex 16S; 16M, 565–567 540–544
Arriflex 16SR-1; 16SR-2, 567–570 Panavision Millenium (PFX-M), 544–546
Arriflex 16SR3, 16HSR3, 16SR advanced, Panavision Millenium XL (PFXMXL), XL2,
16HSR3 advanced, 570–573 542–544
arctic cinematography and, 239 Panavision models list, notes, and accessories,
Bell & Howell 16mm Film, 550–552
Bolex H16mm, 573–575 Panavision Panaflex Panastar II High-Speed,
Glidecam V-20, 174–175 547–548
Ikonoskop A-Cam SP-16, 562–563 Panavision Panaflex-X (PFX-X), 546–547
Panaflex “Elaine,” 575–577 Panavision Platinum Panaflex (PFX-P), 548–550
16mm extreme close-up depth of field and panning speeds at various frame rates chart,
exposure factor versus magnification recommended, 891, 893
or field of view chart, 811 panning speeds in degrees per second chart,
16mm film apertures, 8 recommended, 892
16mm film format tables Photo-Sonics 4CR, 553–555
footage, 880–882 Photo-Sonics 4ER, 555–556
footage and frames obtained at various Photo-Sonics 4ML (reflex and nonreflex),
camera speeds, 887 556–557
frame totalizer, 890 Ultracam, 557–559
16mm/Super 16 field of view tables, 800–803 35mm extreme close-up depth of field and
35mm cameras, 502–559 exposure factor versus magnification or
Aaton 35-3P, 35III, 35II, 502–504 field of view chart, 810
In-2
|
Index
35mm film forma t Reference Input Capture Device (RICD),

footage and frames obtained at various 285, 286
camera speeds chart, 888 reference rendering transform (RRT),
footage table, 883–885 286, 287f, 288
frame totalizer chart, 890 reference signal chain, 286f
35mm film master creation, 28 scene-referred linear floating-point format,
65mm 5-Perf TODD-AO/Panavision 65mm 278
spherical imaged negative aperture technical description, 284–288
specifications, 8 Academy of Motion Picture Arts and Sciences
65mm-15 Perf IMAX/OMNIMAX aperture Academy Color Encoding System.
specifications, 8 See Academy Color Encoding Systems
65mm cameras, 581–590 (ACE)
Arriflex 765, 581–583 Science and Technology Council, 21
Fries Model 865 65mm/8-perf., 583–584 Acme-Dunn Optical printer, 313
Mitchell FC, BFC, 584–585 Active matrix, 253
MSM Model 8870 65mm/8-perf., 585–586 Advanced Camera Systems, Inc., 173–174
Panavision System-65, 586–589 Advanced video coding high definition
Panavision System 65mm handholdable, (AVCHD) format, 44–45
589–590 Aerial camera stabilizing systems, miniature,
65mm film footage obtained at various 182–184
camera speeds chart, 888 Aerial cinematography, 215–228
65/70mm film format footage table, 886 aesthetics, 216–217
70mm and aspect ratios of 1.85, anamorphic aircraft selection considerations, 215
2.40, and Super formats, 68, 71, 78 aviation basics, 217–220
camera mounts and dedicated gyro systems,
A 226–228
Aaton 35-3P, 35-111, 35-11, 502-504 camera mounts types, 223–225
Aaton AMinima (Super 16) camera, 559–560 choice of sides for mounting, 220
Aaton XTRprod 16mm camera, 560–562 common fixed-wing aircraft, 222–223
Above Ground Level (AGL) altitude, 220, 221 communications: radio and pilot jargon, 219
Academy aperture customized Lear jets, 223
1.85 aspect ratio and, 65, 66f dynamic action creation, 217
definition, 63, 63n2 FAA regulations for camera mounts, 217
Academy Color Encoding Systems (ACE) film stocks/T-stops, choice of, 216–217
ASC CDL and, 277 flight envelope definition, 220
benefits, 283–284 fuel trucks, 219
block diagram, 287f glossary of flight physics, 220–221
chromaticity diagram, 285f ground effect definition, 220
components and features, key, 284 helicopters, common, 221–222
definition, 283 losing tail-rotor authority definition, 220–221
digital basic concepts and, 21, 28 max speed, 220
DPX and, 359 metering, 216
input device transforms (IDT), 286f, 287, 288 miniaturized radio-controlled helicopters, 216
linear definition, 283n1 mounts/heads, 468
log definition, 2813n1 onboard, 215
look modification transforms (LMT), 287 pilots, 217–218, 222
output device transform (ODT), protocol, planning, and nonessential payload,
286, 287, 287f, 287n2 218–219
overview, 283–288 radios and pilot jargon, 219
reference device transform, 287n2 rudder pedals definition, 221
| In-3
Index
safety, 220 Animation, 135

settling with power definition, 221 sequence, 189–190
skydiving, 216 Anamorphic
speeds, filming, 216 35mm film, 3
stabilizers in the air, alternate, 225–226 240 aspect ratio advantages and
time of day, 216 disadvantages, 69–73
“Velocity Not to Exceed” (VNE), 220 adjusting attachment, 88–89
Visual Flight Rules (VFR), 220 advantage, 87
weather basics, 219–220 CinemaScope aspect ratio, 62, 62n1
yoke/control stick definition, 221 CinemaScope introduction, 87, 88, 122–123
Aerocrane Strobe Tracking System, 354 cinematography, 87–92
Aerospatiale A-Star/Twin Star conversions, 316–317
(AS350 and AS355 twin) helicopter, 221 definitions, 87, 401
Aerospatiale Llama helicopter, 221 digital intermediate process and, 88
AG-DVX 100, Panasonic, 37 dome port and, 234
AG-HVX200, Panasonic, 37 flat port and, 234
Alpha (matte) channel, 332–333, 332, 346, 351 lenses, 31, 38, 41, 235
American Society of Cinematographers (ASC) lenses use and, 90–92, 120–122
clubhouse photograph, 1f optical system used for CinemaScope and
color decision list (CDL) 24, 25, 261, 405–406 Panavision, 88
digital assistant handheld app, 848 Panavision and anamorphic mumps, 89
magazine, 2 Panavision’s Panaflasher, 297
origins, 1–2 prisms and mirrors and, 89–90
Previsualization Society, 187 rear ana morphic unit, 89
purpose, 1
SMPTE 50-1998 and, 122
screen credit for member, first, 1
spherical versus, 124–129, 127f
Technology Committee’s Digital Intermediate
VistaVision negatives and, 87
Subcommittee, 24, 273
widescreen format, 38
American Society of Cinematographers color
Anamorphic mumps, 89, 125
decision list (ASC CDL), 24, 25, 273–282
Android handheld apps for production use,
behavior for different image encodings, 277
843–844, 851–852
benefits, 261, 273
Answer print, 294
corrections, 275
Aperture
interchange formats, 279–280
compensator for extreme close-ups chart, 812
log, 278–279, 280f
plate, 184
offset, 273, 276, 276f, 278, 278f, 280f
power, 273, 276–277, 276f, 278, 279, specifications summary, 7–8
279f, 281f Apple ProRes 4444, 358
production workflow usage, full, 274f Apps for production use, handheld, 843–854
saturation, 27, 276, 277, 278, 279, 279f, 280, Accessory, 843
280f, 281f Android, 843–844
slope, 273, 275–276, 276f, 278, 278f iOS, 845–851
transfer functions, 275–277 iOS and Android, 851–852
video-gamma and linear (gamma 1.0), Windows and Android, 854
277–279 Aquacolor and underwater filters, 166
XML files and, 280 Arc discharge and fluorescent lamps table,
Analog and digital basic concepts, 9–10 filters for, 832
Analog definition, 401 Arc lamps at 115–120 volts, 695
Analog-to-digital (A/D) conversion definitions, Archival elements, 270–271
10, 401 Archiving, digital basic concepts and, 28
In-4
|
Index
Arri tripods, 241–242

2-C, 178, 180, 182, 183 winterizing cameras, 240, 241, 242
3” x 3” matte box, 183 Artifact definition, 401
35-3, 181 Aspect ratio
435, 179, 225 1.85 advantages and disadvantages, 65–69,
accessories, 525, 525f, 526f 66f
Alexa, 11, 14, 329, 356, 357, 604–605 definition, 401
Alexa camera simulator handheld app, 853 issues, 38–41, 40f
Alexa M, 605–606 Audio issues for shooting HD video digital
Alexa Plus, 606–607 still camera, 52
Alexa Studio, 607–609 Aurasoft™, 388
D-21, 269, 609–611 Autodesk Lustre, 26, 263
fuses for 16mm cameras, 571 Automatic gain control (AGC), 52
matte boxes, 526–528 Avid and nonlinear editing system, 259
raw, 16, 357 AVC-Intra format, 45
ArriCam Lite 35mm camera, 504–506 Avid log exchange (ALE), 261, 275, 280
ArriCam Studio 35mm cameras, 506–509
Arriflex B
16BL 16mm camera, 563–565 Back scattering, 237
16S; 16M 16mm cameras, 565–567 Backgrounds, projection chart for process, 897
16SR-1, 16SR-2 16mm cameras, 567–570 Baked in definition, 402
16SR-3, 16HSR3, 16SR Advanced, 16HSR3 Banding definition, 402
Advanced 16mm cameras, 570–573 Bandwidth definition, 402
35II models, 511 Barndoors, 387, 391
35-3, 513–516 Bayer pattern, 11–12, 11f, 34, 35f, 357, 402
35-2C, 509–510 Bayer pattern photogate sensor, 85f, 86, 358
35-3C, 512–513 Beam projectors, 390
35BL, 523–525 Beaulieu Super 8mm cameras
235, 516–517 4008, 578
435 Advanced, 435 Extreme, 517–519 6008 Pro, 578
535, 519–520, 522f 7008 Pro, 579
535B, 521–523 Bell & Howell 35mm Eyemo cameras, 216,
765 65mm camera, 581–583 529–530
stabilizer, 224 Bell & Howell 35mm Steve’s Cine Service
Varicon, 295 conversion, 530–531
Arriflex/SL 35 Mark II, 513 Bell Huey UH-1 (civilian version 204/205)
Arriflex/SL 35 Mark III, 511–512 helicopter, 222
ArriLaser film recorder, 267 Bell Jet Ranger (206) helicopter, 221
Arriscan, 261 Bell Long Ranger (206L) helicopter, 222
Artemis Cine/HD Camera Stabilizing system, 177 Bellow-type lenses, 133, 667–668
Artemis Remote for iPad, 850 Bench test negative preparation, 299
Arctic cinematography, 239–250 Binary code definition, 402
broad-range lubricants, 240 Bit definitions, 9, 402
equipment and filming techniques, 242 Bit depth
equipment preparation, 240–242 definition, 402–403
equipment problems main reason, 239 digital basic concepts and, 21–22
film, 243 black-and-white film day-for-night
frostbite danger, 242 cinematography, 209
glove types, 242 Black body, 161
storage, 243 Black Pro Mist-type filters, 159
| In-5
Index
Black stretch, 14, 403 plastic materials as screen choice, 335

Black test, printed, 299–300 processed foreground, 332–333
Blackmagic projection composites, 334
cinema camera, 46, 49, 54 recording, 358
resolve, 26 rotoscoping, 334
Bleach bypass, 289 schematic H&D curve, 326f
Blow-ups and reduction and optical printer, 316 screen choices, 334–336
Blue- and greenscreen photography, 325–362 screen brightness setting, 328–329
alpha (or matte) channel, 332–333, 335 smoke plumes normal and underexposed, 327f
backing types and lighting, 343–345 software creation of composite, 331
backing uniformity and screen correction, spatial resolution, 356–357
336–338 spill light control, 351–352
balancing screen (backing) brightness to technical previs and, 188–189
shooting stop, 325–328 television monitors and on-set preview,
camera negative selection, 355–56 354–355
camera selection, 356–357 tracking markers, 353–354
choosing the backing color, 341–343 underwater cinematography and, 237–238
color ban dwidth and compression 358 underwater photography, 346–347
color correction, 359–360 visual effects supervisor, working with, 360
composite, 333 wardrobe colors and, 341
compositing software, 360–361 white card in set lit to shooting stop, 328f
costs and consequences, 355–359 white card method for screen balancing,
day-lit backings limits, 345 328–329
day-lit green and blue backings, 344 BodyCam, 173–174
distinctions from other compositing methods, Bodymount, 180
334 Body-Pan, 184
edge enhancement/sharpening/detail settings, Body-worn stabilizing systems, 173
358 Bolex H16mm cameras, 573–572
environmental bounce light, 347–352 Boresight, 138
eye protection, 341 Brightness and contrast for picture monitor
fabric screen choice, 334–336 set-up, 899–900
Brute Fresnel lens spotlight, 380
filtration, 359
Bungee or surgical tubing mounts, 225
fix it in post, 325
Byte definitions, 9, 403
floor shots, virtual sets, 345–346
foreground lighting and color bias, 348
foreground lighting on the stage, 347 C
front-lit backings, 344 Cable voltage drop, 697
illumination and reflections from the backing, Cablecam, 185–186
348–351, 350f CalColor filters, 383, 834
illuminators, 338–339 Calibration
Keylight package, 360 definition, 403
lenses, 135 of lens for underwater cinematography,
levels for digital original photography, 329–331 233–234
light sources, alternative, 339–341 Camera chart, telescopic, 499
lighting to eliminate shadow, 353–354 Camera, film, 501–601
lighting virtual sets, 352–353 16mm, 559–577; see also 16mm cameras
local color, 348, 349f 35mm, 502–559; see also 35mm cameras
negative scanning and digital conversion, 359 65mm, 581–590; see also 65mm cameras
on-set preview, 354–355 IMAX, 595–601; see also IMAX
overview, 325 Super 8mm, 578–581; see also Super 8mm
painted screen choice, 334, 336 cameras
In-6
|
Index
VistaVision 35mm, 591–595; remote handwheels chart, 489

see also VistaVision remote heads, 468
Camera filters. See Filters, camera remote heads charts, 490–494
Camera mounts for aerial cinematography, suspended camera system chart, 498
types of, 223–225, 226–227 telescopic camera chart, 499
Camera rental order handheld app, 849 track systems, remotely operated, 469
Camera selections for digital cinematography Camera-top encoder handheld app, 843
on a budget, 53–56 Cameras, digital. See Digital cameras and
Camera speed exposure compensator chart, 875 recorders
Camera stabilizing systems, 173–186 Canon cameras
Advanced Camera Systems, Inc., 173–174 814 XL-S, 1014 XL-S Super 8mm, 579
aerial, 225–226 C300, 53, 611–613
aerial systems, miniature, 182–184 EOS 1D C, 54
Artemis Cine/HD, 177 EOS 5D Mark III, 55
Body-cam, 172–173 EOS 7Dm 56
body-worn systems, 173 Vixia HF S21, 56
Cablecam, 185–186 XA10, 56
Coptervision, 182–184 XF format, 45
CVG-A, 182–183 XF105, 55
Flying-Cam, 183–184 XF305, 54
George Paddock Inc., 175–176 Cathode ray tube (CRT) definition, 408
Glidecam Gold Series, 175 Cathode ray tube (CRT) monitors, 15, 17, 19, 254,
Glidecam Industries, Inc., 174–175, 181 260
Glidecam V-16, 175 Cessna (many variants), 222, 227
Glidecam V-20, 174–175 Chapman charts for
handheld systems, 180 cranes, 471–474
Hydro Gyro, 182 dollies, 485–486
Libra III, 182 jibs, 495
MK-V Evolution Modular Sled system, 177 remote heads, 490
PRO-TM System’s Donkey Box II, 175–176 Charge coupled device (CCD), 31, 32, 34, 84,
Rollvision, 181 264, 403
Sachtler, 177 Charts and diagram creation handheld app, 850
Spydercam, 184–185 Chip definition, 403
suspended camera system, 184–186 Chips and pixels, digital basic concepts and,
Tiffen Steadicam, 178–180 10–12
Universal Model III, 178 Chromakey green fabric, 344
Camera-support systems, 467–499 Chromatic aberration, 233
airplane and helicopter mounts/heads, 468 Chromaticity coordinates definition, 404
arms remotely operated, 469 Chromaticity diagram, 20f, 404
body-worn systems chart, 498 Cine lens list, 653–672
cable rigged, 469 Cine zoom lens, 139
crane charts, 471–483 Cinedeck EX onboard recorder, 638–639
cranes, 467–468 CinemaScan process, 261
dollies, 467 CinemaScope. See Anamorphic cinematography
dolly charts, 485–488 Cinematographer and the laboratory. See
hand wheels, remote, 468 Laboratory and the cinematographer
jib charts, 495–497 Cineon digital film system, 258, 259, 263, 264,
jibs and sliders, small, 469 361
radio controlled vehicles, 469 Cineon file definition, 405
remote controlled camera chart, 498 Circle of confusion 140
| In-7
Index
Clinometer handheld app, 845 Color decision list (CDL), ASC. See ASC color
Clipping, 13, 14, 405 decision list (CDL)
Close Encounters of the Third Kind and streaking Color gamut definition, 406
of hot (bright) objects, 128 Color grading, 257, 257f, 265–266
Close-to-macro focusing, 130 Color sampling
Close-up and split-field diopter lenses, 169 as compression of image data, 131f
CMY (cyan, magenta, yellow) definition, 405 definition, 406–407
Code of Safe Practices, 421 digital basic concepts and, 22–23, 23f
Code value definition, 405 Color space
CODEC correlated, 375
avoiding recording, 47–48 definition, 407
definition, 405 digital basic concepts and, 18–21
interframe definition, 413 Color temperature
intraframe definition, 408 definition, 375
Codex ArriRaw onboard recorder, 639 meters, 164, 255, 380
Codex Onboard M recorder, 639–640 Colortran boosting, 380
Codex Onboard S recorder, 639–640 Commission Internationale de L’Eclairage (CIE)
Color, filters for, 161–167 XYZ color space, 19, 21, 405
color-conversion filters, 162 Common center, 75, 75f, 76
color-compensating (CC) filters, 162, 164 Common topline, 75, 75f, 76
coral filters, 165–166 Communications
Decamired filters, 162–163 radios and pilot jargon, 219
Didymium filers, 166 underwater, 236
differences between camera and lab/post Comparisons of 35mm, 1.85, anamorphic,
correction, 166 Super 35 film formats, 59–86
in digital world, 161–162 advantages of 1.85, 65–68
digital post processing options, 167 advantages of Super 35 composed for 2.40
fluorescent and other discontinuous spectra aspect ratio, 76–78
lighting correction, 163, 164 composition, 64, 67, 68, 69, 72
gel filters, 145, 164 digital motion picture cameras, 81–86
graduated color filters, 164–165 disadvantages of 1.85, 68–69
information, general, 161 disadvantages of anamorphic 2.40, 69–71
light-balancing filters, 162 disadvantages of Super 35 composed for
LL-D, 167 2.40 aspect ratio, 78–81
mixed lighting situations, 164 fill factor, 83–84
sepia filters, 166 history and definition of terms, 62–64
stripe filters, 165 introduction, 59–62
underwater color-correction filters, 166 magnification, 68
Color absorption and underwater cinematography, negative area, 68, 69, 78, 79
231 pixels, 85–86
Color and look management, digital basic pros and cons clarification, 64–69
concepts and, 23–26 resolution versus megapixels, 83
Color balancing scanner versus cameras, 84–85
for existing fluorescent lighting chart, 841 screen heights importance, 60–62
for existing nonphotographic lighting chart, video transfers, 67–68, 71–72, 78
839 Complementary high-density metal oxide
to match daylight or ambient light on location semiconductor (CMOS) definition, 405
interiors chart, 840 Component video signal definition, 407
Color correction collection (CCC), 281 Composite Components Co. (CCC), 335, 335n3,
Color correction for carbon arcs table, 833 339, 341, 346
In-8
|
Index
Composite image definition, 407 laboratory handling, 289

Composite video signal definition, 407 one-light, 293
Composition, 64 one-light print versus, 306
1.85 aspect ratio and, 67, 68 single printer light setting printing, 305
2.40 versus 1.85 aspect ratios and, 64 timed, 293
anamorphic 2.40 aspect ratio and, 69, 72 timer, 309
Compression Data definition, 408
definition, 408 Database application handheld app, 849
digital basic concepts and, 18 Day-for-night cinematography, 207–214
ratio definition, 408 black-and-white film, 209, 210
Computer generated effects contrast between players and background,
filtering for digital intermediate and, 171 208
Computer-graphics imaging (CGI), 135 filters and degree of underexposure, 207
data management and, 264–265 lighting, 207
Computer displays, filming, 251–255 negative color film, 209–210
color temperature, 254–255 neutral density filters and, 208
frame rates and exposure time, 251–252 Pola screens, 207, 208
LCD monitors and projectors, 253 reversal color film, 209
monitors, 252–253 skies and, 207–208
obstacles, principal, 251 Dazian and blue- and greenscreens, 335
over- and undercranking, 252, 253 Dead (dark or bright) pixel definition, 409
plasma displays, 253 De-Bayer process, 12, 16, 357
Computers and lenses, 139 De-mosaic process, 12
Computers and motion control, 323 Decamired filters, 162–163, 376, 834
Contrast control filters, 159 Decimal equivalents of fractions and equivalents
Contrast-viewing filters, 167, 168 of fractions of an inch in mm, 691
Convergent Design Gemini onboard recorder, Dedo light, 390, 390f
641–642 Delrama, 90
Conversion factors, measuring, 698 Densitometer, 291, 292, 299, 300
Conversion tables, 689 Depth of field
Cooke S4 Wide Prime lens, 182 65mm-format lenses and, 135
Copper cable type SC entertainment 90°C, 697 aspect ratio and, 65, 67
Copperwire voltages drop, 695 calculators, 844, 848, 852, 853, 854
Coptervision stabilizing system, 181 circles of confusion and, 140
Coral filters, 166 diffusion filters and, 156
Crane charts, key to reading, 470 effects and focal-length changes with filter
Cranes, 467–468, 471–480 considerations, 169–170
Crushing the blacks definition, 408 formulas, 673–687, 679f
Cucoloris, 392 hard lighting and, 124
Custom (homemade and field-ready) filters, HD video shooting on digital still camera, 50
170 lens, 136
Cyclorama luminaries, 388 lighting and, 124
neutral-density filters and, 149
D optical breathing and, 130
out-of-focus objects and, 120
Dailies
sensor size and 48–49
ASC CDL workflow and, 274, 280
spherical versus anamorphic and, 124–125,
color analyzers for, 305
128
digital postproduction for feature films and,
slant-focus lens and, 133
259–261
Super 35 formats and, 76
laboratory aim density and, 292–293
| In-9
Index
Depth of field tables, 699–751 lenslats, 158

5.9mm, 700–701 mesh, fine, 155
8mm, 702–703 moiré patterns, 156, 158
9.8mm, 704–705 multiple filter use and, 170
10mm, 706–707 optical diffusion digital camera, 158
12mm, 708–709 pattern handling, 156
14mm, 710–711 sliding, 158
16mm, 712–713 Digital assistant handheld app, 848
18mm, 714–715 Digital calculator handheld app, 847
20mm, 716–717 Digital camera equipment preparation.
24mm, 718–719 See Preparation of digital camera equipment
25mm, 720–721 Digital camera postproduction considerations,
28mm, 722–723 some, 271–272
32mm, 724–725 Digital cameras, 602–638
35mm, 726–727 Arri Alexa, 604–605
37.5mm, 728–729 Arri Alexa M, 606–607
40mm, 730–731 Arri Alexa Plus, 605–606
50mm, 732–733 Arri Alexa Studio, 607–609
75mm, 734–735 Arri D-21, 609–611
85mm, 736–737 Canon C300, 611–613
100mm, 738–739 list of, 602–603
105mm, 740–741 Panasonic AJ-HDC27H VariCam, 613–614
135mm, 742–743 Panasonic AJ-HPX3700 VariCam, 614–615
150mm, 744–745 Panavision Genesis, 615–617
200mm, 746–747 Pro I/O module, 618–619
300mm, 748–749 RED EPIC, 617–619
400mm, 750–751 RED ONE, 619–620
and exposure factor versus magnification or RED Scarlet-X, 620–621
field of view (16mm extreme close-up), 811 Silicon Imagining SI-2K, 621–622
and exposure factor versus magnification or Sony F23 CineAlta, 622–624
field of view (35mm extreme close-up), 810 Sony F35 CineAlta, 624–626
introduction, 699 Sony F65 CineAlta, 626–627
Super 8mm/6.5mm, 804 Sony HDC-F950 CineAlta, 627–628
Super 8mm/13mm, 805 Sony HDW-F900 CineAlta, 628–630
Super 8mm/38mm, 806 Sony HDW-F900R CineAlta, 630–631
Super 8mm/50mm, 807 Sony PMW-F3 CineAlta, 631–632
Depth of focus formulas, 676–677 Vision Research Phantom 65 Gold, 637–638
Designing the look responsibilities, 3 Vision Research Phantom Flex, 632–634
Device dependent color space definition, 409 Vision Research Phantom HD Gold, 634–635
Vision Research Phantom Miro M320S,
Device independent color space definition, 409
Didymium filters, 166 635–637
Diffusers, 390 Digital cinema, mastering basic concepts of, 28
Diffusion filters, 155–158, 157f Digital cinema distribution master (DCDM), 28,
center spot filter, 156, 157 268, 270, 271, 408–409
circular, 156 Digital cinema initiatives (DCI), 28, 262, 266,267,
depth of field and, 156 409
dot filters, 158 Digital cinema package (DCP), 28, 409
faces, diffusing, 156 Digital cinematography on a budget, 37–56
filtering for digital intermediate and, 171 aspect ratio issues, 38–41
green- and bluescreen and, 359 camera selection, 53–56
depth of field and sensor size, 48–49
In-10
|
Index
DSLR video, beyond, 53 Super 35 formats and, 73, 78, 80, 81

frame rate issues, 41–43 video masters creation and, 269
history, brief, 37–38 Digital light processing (DLP) definition, 410
recording CODEC, avoiding, 47–48 Digital master definition, 410
recording formats, 44–45 Digital micromirror devices (DMDs), 254
shooting HD video on digital still camera, 49–53 Digital motion picture cameras, 81–86
videotape, beyond, 45–47 capture format, 82
Digital concepts, basic, 9–29 fill factor, 83–84
ACES, 21 log images, 277
analog and digital, 9–10 pixels, 85–86
archiving, 28 resolution versus megapixels, 83
bit depth, 21–22 scanner versus cameras, 84–85
color sampling, 22–23 Super 35 formats and, 73, 78, 80, 81
color space, 18–21 Digital motion pictures cameras primer,
compression, 18 evaluating, 31–35, 32f
conclusion, 29 complicated question, 35
exposure, 13–14 fill factor, 33
gamma and log/lin, 14–15 megapixels, 34
intermediate, digital, 26–27 resolution versus megapixels, 32
look management and color management, scanner versus cameras, 32–33
23–26 sensors and amount of light they capture,
mastering, 27–28 digital, 33–34
pixels and chips, 10–12 Digital movie slate (clapper board) handheld
raw versus de-Bayer, 16 apps, 843, 849
resolution, 12–13 Digital onboard recorders, 638–647
video versus de-Bayer, 16 Cinedeck EX, 638–639
video scanning—interlace or progressive, Codex ARRIRAW, 639
16–18 Codex Onboard M, 639–640
Digital conform definition, 409 Codex Onboard S, 640–641
Digital Green© fabrics, 345 Convergent Design Gemini, 641–642
Digital ICE, 265 Panavision SSR-1, 642
Digital intermediate (DI) Sony SR-R1, 643
2K, 59 Sony SR-R4, 643–644
2K resolution and, 262 Sony SRW-1, 644–645
3-D post handling and, 202 S.Two OB-1, 645–646
4K resolution and, 263, 264 Vision Research Phantom CineMag-II,
4K workflows, 59 646–647
Anamorphic cinematography and, 88 Digital post processing options and filters for
basic concepts, 26–27 color, 169
color grading, 27f Digital postproduction for feature films, 257–272
data dailies and, 261 2K or 4K, 262–264
definitions, 26, 409–410 archival elements, 270–271
edit decision list and, 26 color gamut, 266, 267, 267f
filtering for, 171 color grading, 257, 257f, 265–266
impact on final image quality, 59 conforming, 265
Kodak and, 259 dailies, 259–261
lens and, 668, 669 data dailies, 261
log images, 279 data management, 264–265
postproduction for feature films, 257, 257f, digital camera considerations, some, 271–272
258 digital cinema, 268–269
|In-11
Index
digital intermediates and, 257, 257f, 258 Dynalens stabilizers, 226

dirt fix, 264 Dynamic range
dodge and burn, 257 definition, 411
film output, 267–268 testing for, 302–303
history, brief digital, 258–259
home video masters, 269–270 E
scanning, 261–262 Éclair CM-3 16/35mm camera, 531–532
Digital sensors and amount of light they capture, Editors decision list (EDL)
33–34 CMX EDL files, 275
Digital signal definitions, 9, 410 conforming shots to, 265
Digital source master (DSM), 27, 267, 270, 271, data dailies and, 261
409 definition, 411
Digital technology brief history, 258–259 digital intermediate and, 26
Digital terminology introduction, 401–421 scanning and, 261
Dimmers, electrical, 392 Egg crates to control spill light, 393
Diopter lenses, 131, 668–669, 687 Egripment cranes, 473–474
Diopter lenses focal length conversion tables, Electrical toolkit handheld app, 851
plus, 813, 818 Element Labs, 398–399
Direct to edit (DTE) format, 45 Empireflex VistaVision camera, 227
Director’s viewfinder handheld app, 844, 851 Emulsion test preparations to establish a
Discrete values definition, 410 single printer light (tungsten), 306–310
Disk space calculator for Final Cut Pro editors, 853 ASA Exposure Index, 307
Distribution systems, 696–697 conclusion, 309–310
DLP displays and projectors, 254, 265, 266 filters, 307
Dodge and burn shots digitally, 266 fleshtone tweaking, 310
Doggicam laboratory instructions, 308
dollies, 486 lighting, 307
handheld stabilization systems, 180 miscellaneous, 308
remote heads, 490–494 preliminaries, 306–307
Dogma ’95 movies, 37 results, 309
Dollies T-stop, 307
camera-stabilizing systems and, 173 test, actual, 308–309
camera-support systems and, 467 Emulsion testing, 299–304
charts, 485–488 bench test negative preparation, 299
charts, key to reading, 484 black test, printed, 299–300
with lifting arms, 467 density aim, laboratory, 302f
with lifting center posts, 467 dynamic range, testing for, 302–303
location/doorway, 467 exposure calibration, 300–302
optical zoom and, 315 judging, 299
with track for motion control, 322–323 object of, 299
Dome lens port and underwater cinematography, trust your eye more than your meter, 303–304
232, 233, 234, 235 Encoding definition, 411
DPX file definition, 410 ENR, 289, 300, 303, 307
DSLR video, beyond, 53 Equipment and underwater cinematography,
Dustbusting, 265 231–232
DV recording format, 44 European comparison of filter system names
DVCAM, 44 chart, 827
DVCPRO, 44 Evaluating digital motion pictures cameras
DVCPRO-HD format, 16, 44 primer, 31–35, 32f
Dynalens, 666–667 complicated question, 35
In-12
|
Index
fill factor, 33 Spectra Cinespot IE Spot Meter, 108

megapixels, 34 Spectra Professional IV, 109
resolution versus megapixels, 32 Spectra Professional IV-A, 109–110
scanner versus cameras, 32–33 testing, 96
sensors and amount of light they capture, tropical cinematography and, 244
digital, 33–34 underwater cinematography and, 235–236
Exposure, digital basic concepts, 13–14 Exposure time and filming television and
Exposure calibration, 300–302 computer displays, 251–256
Expose factor and depth of field versus Exposure using incident light, 685–686
magnification or field of view (16 and 35mm Exposure/dynamic range test, 461–462
extreme close up) tables, 810 Extenders, 133–134
Exposure composure compensation for extreme Extension tubes or bellows, 667, 668, 686–687
close-up cinematography table, 814 Extreme close up (16mm or 35mm cameras)
Exposure Index (EI), 295, 300, 301, 302, 307, 310 tables, 810–815
Exposure index/ASA exposure index aperture compensator for, 812
reduction table, 869 depth of field and exposure factor versus
Exposure meters, 57, 93–110 magnification or field of view, 810–811
analog type, 94 exposure compensation for, 814
Gossen Color-Pro 3F, 97 focusing chart for 2, 3, 4, and 6 inch lenses,
Gossen Luna-Pro Digital F, 98 815
Gossen Luna-Pro F, 98 plus diopter lenses for focal length
Gossen Luna-Pro S, 98–99 conversation, 813
Gossen Luna-Star F2, 97 Eye protection, UV, 341
Gossen Mavolux 5032 C/B, 99
Gossen Starlite, 96–97
Gossen Ultra-Spot2, 99 F
Handheld app for production use, 843 FAA Motion Picture Manual, 218
incident-light meters, 93–95 Falling body, distance and velocity of free, 688
Minolta 1-Degree Spot meter, 102 Far and Away and slant-focus lenses, 124
Minolta Auto Meter IV F, 100–101 Field definition, 411
Minolta Auto Meter V, 101 Field of view
Minolta Auto Meter VF/Kenko KFM-1100, calculator handheld app, 846, 847, 848
101–102 f/2 aperture and, 123f
Minolta Cinemeter II, 99–100 optical breathing and, 130
Minolta Color Meter III-F/Kenko KCM-3100, Field of view tables, 752–799
102–103 8mm, 752–753
object of, 93 9.8mm, 754–755
Pentax Digital I Degree Spot Meter, 103 10mm, 756–757
Pentax Spotmeter V, 103 12mm, 758–759
reflected-light meters, 94, 95–96 14mm, 760–761
Sekonic L308BII, 105 16mm, 762–763
Sekonic L358M, 106 16mm extreme close up, 811
Sekonic L398M, 105 16mm/Super 16, 800–803
Sekonic L508C, 103–104 18mm, 764–765
Sekonic L558, 106–107 20mm, 766–767
Sekonic L608C, 104–105 24mm, 768–769
Sekonic L778 Dual Spot F, 107 25mm, 770–771
special effects, 95 28mm, 772–773
specific situations in outdoor photography, 95 32mm, 774–775
Spectra Cinespot Foot Lambert Meter, 108 35mm, 776–777
|In-13
Index
35mm extreme close up, 810 Kodak light balancing filters, 822
40mm, 778–779 Kodak ultraviolet and haze cutting filters, 823
50mm, 780–781 mired shift value effects, 835
75mm, 782–783 mired values of color temperatures from
85mm, 784–785 2000° to 10,000°K, 835
100mm, 786–787 neutral density filter selector chart, 824
105mm, 788–789 sizes for motion pictures, most common, 817
135mm, 790–791 Tiffen decamired filters, 834
150mm, 792–793 underwater photography color filter
200mm, 794–795 selection chart, 826
300mm, 796–797 Filters, camera, 143–172
400mm, 798–799 application filters, special, 167–168
Super 8mm/6.5mm, 804 barrier, 213
Super 8mm/13mm, 805 for black-and-white, 160–161
Super 8mm/38mm, 806 CalColor, 383
Super 8mm/50mm, 807 checking out, 434
File format conversion definition, 411 circular gradient and circular/elliptical
File format definition, 411 center spot ND filters, 152
File transfer protocol (FTP) definition, 411 close-up and split-field diopter lenses, 169
Fill Facture and digital motion picture cameras, for color, 161–167
33, 83–84 color and black-and-white, 145–155
Film calculator, 852 color in digital world, 161–162
Film camera equipment preparation. color-compensating filters, 162
See Preparation of motion picture film color-conversion filters, 162
camera equipment contrast control filters, 159–160
Film formats contrast-viewing filters, 167, 168
comparisons, 59–86 for control of natural daylight, 379
history and definition of terms, 62–64 coral filters, 165–166
Film rate calculator, 852 custom (homemade and field-ready) filters,
Film sizes in use, 120 170–171
Film stocks chart, Eastman Kodak, 864–865 Decamired® filters, 162–163
Film-strip projectors, 300 definition, 143
Filmlight Baselight, 26 Didymium filter, 166
Filmlight Northlight scanner, 265 diffusion, 155–158
Filter tables, 819–834 digital postprocessing options, 167
for altering black-and-white contrast of effect of depth of field and focal-length
colored subjects, color filters, 820 changes, 169–170
for arc discharge and fluorescent lamps, 832 exciter, 213
for black-and-white cinematography daylight factors, 144
exteriors, selected color filters, 819 filtering for digital intermediate, 171
color temperature adjust and lighting filters, fluorescent and other discontinuous spectra
830–831 lighting correction, 163–164
comparison of system names, 827 fog, double fog, and mist filters, 158–159, 170
compensator chart, 825 grades and nomenclature, 144–145
correlated color temperature of typical graduated color filters, 164–165
light sources, 821 graduated ND, 151–152, 151f
for daylight correction, 829 for incandescent lamps, 379–380
Kodak color compensating filters for infrared filters and digital imaging, 147
color films, 823 infrared filters and film imaging, 146
Kodak conversion filters for color films, 822 infrared-neutral density, 149–150
In-14
|
Index
lab/post correction versus camera, 166–167 Focusing digital still camera lens, 50–51
light-balancing filters, 120 Fog, double fog, graduated fog, and mist filters,
LL-D, 167 158–159, 170
mixed lighting situations, 164 Formats summary, 7–8
multiple use, 170 Formulas, 673–687
neutral-density, 148 conversion tables, 687
optical diffusion in digital camera, 158 distribution systems, 696–697
planning, 143 lens, 673–681
polarization filters and 3-D, 154–155 light and exposure, 685–687
polarizing, 152–153 measurement conversion factors, 698
rain spinners, 171 Ohm’s Law, 691
secondary reflections, 170 shooting practices, 681–685
sepia filters, 166 Frame rate issues, 41–43, 42f
sizes, shapes, and on-lens mounting Frame rates for slow-motion on HD video digital
techniques, 144 still camera, 51
sliding diffusion filters, 158 Fresnel lens spotlights, 386–387, 388, 390
special effects filters, 155–160 Fries 435 35mm camera, 533
star and flare effects, 160 Fries Model 865 65mm/8-pef camera, 583–584
technology changing, 143 Fringing (color) definition, 411
Tessive Time Filter, 150 fStop Wireless WiFi Receiver, 843
tilting filters, 170 Fuji comparison of filter system names chart,
tone control filters, 160–161 827
ultraviolet, 145–146
underwater cinematography and, 235
underwater color-correction filters, 166 G
variable neutral density, 148–149 Gain definition, 412
Wratten filters. See Wratten filters Gamma and Log/Lin, digital basic concepts, 14–15
Wratten numbers, 145 Gamma correction definition, 412
FireWire (IEEE 1394), 44 Gamma definition, 412
Fixed-wing aircraft, common, 222–223 Gekko Technology Ltd., 398
Flash memory definition, 411 Gel filters, 164, 170
Flashing, 296–297 Gel frames, 391
Flat lens port and underwater cinematography, Gel swatch library handheld app, 846
232–233, 234 Glasses for 3-D, 202–205
FLEx files, 275, 280 active (shuttered), 204f, 205
Floating point definition, 411 anaglyh, 204–205, 204f
Fluorescence (black-light) photography, 211–212 circular polarization, 203–204, 203f
Fluorescent lamps, filters for arch discharge and, Dolby, 204, 204f
832 first public use, 202
Fluorescent lighting, 381, 837 leakage of image information, 203
Fluorescent and other discontinuous spectra left and right eye images and, 197
lighting correction, 163–164 linear polarized, 202, 203, 203f
Flying-Cam miniature aerial system, 183–184 liquid crystal, 204f
Focal length matching exposure calculator passive, 203
handheld app, 844, 846, 847 Pulfrich, 204f, 205
Focal length of lens in mm charts, 891, 893 Gobos, 392
Focal length versus horizontal angle table, Glidecam Industries, Inc., 174–175, 175m 181
effective, 809 Gold Series, 175
Focal length versus vertical angle table, Gyro Pro, 181
effective, 808 V-16, 175
Focus conversion table, plus diopter lenses, 818 V-20, 174–175
|In-15
Index
GoodReader for iPhone app, 849 Helicopters

Gossen exposure meters ball mounts, 224–225
Color-Pro 3F, 97 belly mounts, 223, 226
Luna-Pro Digital, 97–98 common, 221–222
Luna-Pro Digital F, 98 door mounts, 226
Luna-Pro F, 98 door/side mounts, 224
Luna-Pro S, 98–99 exterior gyrostabilized mounts, 224–225
Luna-Star F2, 99 fixed-wing aircraft versus, 215
Mavolux 5032 C/B, 99 ground effect, 220
Ultra-Spot 2, 99 losing tail-rotor authority, 220–221
Grading definition, 412 mounts/heads, 468
GR-HD1 HDV camcorder, JVC, 37 nose/belly mounts, 223
Green and bluescreen. See Bluescreen and pilots, 218
greenscreen photography settling with power, 221
Greenscreen, 135 weather basics, 219–220
Ground fault circuit interrupted (GFCI), 238 High definition (HD) format definition, 412
Group of pictures (GOP) and compression, 18 High dynamic range image (HDRI) definition,
Gyro Pro camera stabilizer, 181 412
GyroMaster, 226 Histogram, 13–14, 412–413
Gyron systems, 227 HMI lights/lamps
Gyrosphere (dedicated mass-gyro system), 227 as AC discharge lamp, 382
beam projectors and, 390, 903
H blue- and greenscreen and, 341, 348
Handheld apps for production use. characteristics, typical, 837
See Apps for production use, handheld color rendering index and, 376–377
Handwheels chart, remote, 489 electrical and physical characteristics
Hazletine timer, 309 summary, 836
HDCAM format, 16 flicker and, 364
HDCAM-SR format, 16 flicker free calculator handheld app, 847
HD cameras, early, 10 Fresnel lens spotlights and, 386
HD underwater, 229–230 Fresnels chart, 904
HD video shooting on digital still camera, globe wattage and globe amperage, 694
49–53 helium filled balloons and, 388
audio, 52 open reflector variable-beam spotlights and,
bypassing internal recorder, 51–52 387
control for, handheld app, 851 PAR charts, 906–909
depth of field, 50 quartz beam projector table, 905
focusing lens, 50–51 tungsten lamps versus, 353
frame rates for slow-motion, 51 underwater cinematography and, 236, 237
issues to consider, 49–52 Hobbit 3-D 48 fps, The, 193
lens coverage, 50 Horizontal angle versus effective focal length
lens selection, 50 table, 809
monitoring, 51 Hot mirrors, 147, 150
picture quality, 52 Hue and chroma for quick picture monitor set-up,
postproduction, 52 899
recording, 51 Hue definition, 413
sensor sizes, 49 Hughes 500 series (A through E + NOTAR models)
storage, 51 helicopters, 222
HDV format, 44 Hybrid film/digital workflow definition, 413
Header definition, 412 Hydro Gyro digital-stabilization head, 182
In-16
|
Index
Hyperfocal distances, 140 -R Recommendation 709, 266, 269, 276, 277

Super 8mm/6.5mm table, 804 Internegatives, 292, 294
Super 8mm/13mm table, 805 Interpolation definition,
Super 8mm/38mm table, 806 Interpositives, 292, 294, 303
Super 8mm/50mm table, 807 Intra-frame CODEC definition, 414
iOS handheld apps for production use, 845–851
I ITU-R BT.709, 20, 283, 284
ITU-R Recommendation 709, 266, 269, 276, 277
IEEE 1394 (FireWire), 44
Ikelite digital exposure meter, 235, 236
Ikonoskop A-Cam dII camera, 54 J
Ikonoskop A-Cam SP-16 16mm camera, 562–563 Jet systems, customized, 223
Illuminators for blue- and greenscreen, 338–339 Jibs charts, 495–497
Image file definition, 413 Joysticks, 321–322, 323
Image flip feature cameras, 48 JVC cameras
Image on film, putting, 57–58 GR-HD1 HDV camcorder, 37
Image processing definition, 413 GY-HM100U, 55
IMAX GY-HM250U, 55
3-D camera, 595–597 GY-HM750U, 55
15-perforation film, 120
Cameras, 595–601 K
conversions, 79f Kenworthy snorkel camera systems, 666
IW5, IW5A cameras, 597–598 Kenyon stabilizer, 226
MKII 4W and H/S cameras, 598–600 Ki Pro recorder, 46
MSM 9801 camera, 600–601 Kino
In-camera compositing. Flo fixtures table, 917–919
See Miniatures with full-scale live-action Flo lamps, 348
actors, in-camera compositing of Flo lamps for underwater, 346
Incandescent lamps globe wattage and Flo white tubes, 341
globe amperage, 694 Flo-4-tube and 8-tube units lamps, 338
Incident key light/T-stop table, 866–868 fluorescent lighting, 381
Incident-light meters, 93–95 green fluorescents, 330
Infrared cinematography, 210–211 Kodak
Infrared filters color compensating filters for color films table,
digital imaging and, 147f, 147 823
film imaging and, 146f, 146–147 conversion filters for color films table, 822
Input Device Transforms (IDTs), 21, 286f, 287, 288 Gray Card, 96
Institute of Radio Engineers (IE) measurements, light balancing filters table, 822
13, 14 Neutral Test Card, 96
Inter-frame CODEC definition, 413 ultraviolet and haze cutting filters table, 823
Interlace scanning definition, 413 Vision 2 films, 356
Interlaced video scanning benefits and
downsides, 17 L
Internal recorder bypassing for shooting HD video Laboratory Aim Density (LAD), 292–293, 294
on digital still camera, 51–52 Laboratory and the cinematographer, 289–298
International Cinematography Summit Conference aim density, 302
(ICSC), 2 answer print, 294
International Telecommunications Union (ITU) black-and-white negative and reversal films,
definition, 414 294–295
-R BT.601.5 definition, 414 color reversal films, 294
-R BT.709, 20, 283, 285, 406 dailies. See Dailies
|In-17
Index
densitometer, 291 antireflection coatings, 115, 116

emulsion, 292 apertures T versus f-stops), 680
exposure reporting, 293–294 applications of, special, 135–136
flashing, 296–297 attachments, 133–134
forced development of color films, 295–296 attributes and format size, 134–135
Lightflex, 296, 297 bellow-type, 133, 667–668
monitoring the process, 291–293 calibration for underwater cinematography,
negatives, life of, 292 231–232
printer calibration, 291 catadioptic still photography lenses, 114,
printer lights matching, 289 114n3
printer points, 300–301 check back focus on, 464
processing, special 294 checking out, 435
pushing, 289, 295, 296 cine lens list, 653–664
release printing procedures, 294 cine versus still, 118
techniques by lab to modify look of film, 289 close-up cinematography, extreme,
Lagrange invariant, 113f, 122, 124 667–671
Lamps glove wattage and amperage, components, 114, 114–115n4, 115
incandescent, 694 computers and, 139
Laser recorder definition, 414 Cooke, 235
Latitudes chart, 855–856 data tables, 140
LED (liquid crystal display), definition, 414 definition, 112
LED displays, filming, 254 depth of field. See Depth of field
LED lighting for motion picture production, depth of focus formula, 676–677
395–399 design and manufacture, 136–137
advantages over conventional lighting designs of early lens, 115–116, 117f
technology, 395 designs of some modern lens, 121f
challenges, 395–397 diopter, 669–671
color, 396–397 diopter and split diopter close-focusing
companies providing, 395, 397–399 lenses, 133
diffusion and light loss, 396 displacement for focusing, 681
fixture types available, 397–399 Dynalens, 666–667
glare, 396 ergonomics, 137–139
heat management, 397 extenders, 133–134
shadows, multiples, 396 extension tubes or bellows, 667–668
LEDs (light emitting diodes), 390 focal length conversion table, 690
LEDz, 399 focusing digital camera, 50–51
Lens, subject, distance, and image size format size and lens attributes, 134–135
relationships, 678–680 formats and, 120–124
Lens angle calculator, 852 formulas, 673–687
Lens coverage for shooting HD video on Frazier lens system and The Rock, 132
digital still camera, 50 future, 139–140
Lens focal length conversion table, 816–817 hyperfocal distance formulas, 673
Lens mounted filters, 144 introduction, general, 111–119
Lens selection for shooting HD video on Kenworthy snorkel camera systems, 666
digital still camera, 50 kinds and types, 119–120
Lenses, 111–142 LL-D, 167
anamorphic cinematography and use of, manufacture and design, 137
90–92, 119–120 medium-angle, 117f, 119, 121f
anamorphic formats, 120 mesmerizers, 133, 134
anamorphic versus spherical 124–129, 127f objective, 112
In-18
|
Index
Panavision 24mm T3.5 or 45mm T.28 filters for control of natural daylight, 379
slant focus lens, 664–665 filters for incandescent lamps, 379–380
Panavision/Frazier lens system and camera fluorescent lighting, 381
and Titanic, 132 grip accessories for light control, 392–393
pitching lens, 132, 665 high pressure DC short arc xenon
ports, flat or dome, 232–233 light sources, 383
prime lens. See Prime lenses HMI lamps, 382, 836, 837
reflective lens layout, 114f incandescent lamp operational characteristics,
selection for underwater cinematography, 379–383
233–235 incandescent light sources, 379
semi-telephoto, 667 intensity and lighting quantities, 377
single lens definition and illustration, 113f lighting and other flashing-type lighting
slant-focus lenses, 122, 132–133 effects, 384–385
special purpose, 664–665 low-pressure AC/DC xenon arc lamps, 384
spherical versus anamorphic, 124–129 MIRED (Micro Reciprocal Degrees) system,
specialty lenses and systems, 130–133 375–376
Super 35 and size of, 78 national carbons for studio lighting chart, 838
Swing Shift, 664 photographic light sources, 378, 828
Viewing angles formulas, 678 programmable handheld app, 845
wide-angle, 119, 121f special purpose photographic lighting
Zeiss, 235 equipment and systems, 384–386
zoom lens. See Zoom lenses synchronized stroboscopic lighting, 385–386
Letterbox format voltage operation, boosted, 380–381
anamorphic 2.40 aspect ratio and, 72 Lighting and underwater cinematography,
digital cameras and, 82 236–237, 238
Super 35 composed for 2.40 aspect ratio and, Lighting fixture charts, 901–919
81 factors as supplied by manufacturer,
Libra III digitally-stabilized camera mount, 182 multiplying, 916
Lift definition, 414 HMI and quartz beam projectors, 905
Light and exposure formulas, 684–686 HMI fresnels, 904
Light-balancing filters, 162 HMI PAR, 906–909
Light emitting diodes. See LEDs illumination data for source four
Lightflex, 296, 297 ellipsoidal spotlights, 916
Light meters. See Exposure meters Kino Flo fixtures, 917–918
Light scattering and underwater cinematography, open face, 901
237 PAR 64 fixtures, 910–915
Light sources, luminaries, and lighting filters, softlights, 901
375–393; see also Luminaries SoftSun 5400°K, 919
AC arc lamp flicker problem, 381–382 tungsten fresnels, 902–903
AC arc lamps, enclosed, 381 Lighting situations and filters, mixed, 164–165
accessories, light-control, 391–395 Line of sight, 138
calculations general comments, 378 Linear code values definition, 414
characteristics of typical photographic Linearity definition, 414
light sources table, 837 Liquid crystal display. See LED
color rendering index, 376 Litepanels Inc., 395, 397–398
color temperature, 375 Logarithmic color values definition, 414
commercial/industrial light sources, 386, 828 Look and color management, digital basic
coverage and lighting quantities, 377–378 concepts and, 23–26
daylight, natural, 378–379 Look management
DC carbon arc sources, 380 definition, 415
|In-19
Index
preparation, 447–448 considerations, important, 367–371

systems, 23, 413 mixed scales, 371
test, 460 nodal pans and tilts, 371
Look-Up Table (LUT), 24, 25, 260, 265, 267, perspective, forced, 371
268, 270, 415 speed of moving objects, adjusting, 371–373
Lossless compression definition 415 Miniaturized radio-controlled helicopters, 216
Lossy compression definition, 415 Minolta
Luminaires, 386–390 1-Degree Spot meter exposure meter, 102, 236
beam projectors, 390 Auto Meter IV F exposure meter, 100–101
cyclorama, 388 Auto Meter V F exposure meter, 101
Dedo light, 390 Auto Meter V F/Kenko KFM-1100 exposure
ellipsoidal spotlight, 390, 390f meter, 101–102
Fresnel lens spotlights, 386–387 Cinemeter II exposure meter, 99–100
light emitting diodes (LEDs), 390 Color Meter II-F/Kenko KCM-3100 exposure
open reflector variable-beam spotlights, meter, 102–103
387–388 Color Temperature, 236
optical lens system lights, 390 IV, 236
Par fixtures and lamps, 389 Mired comparison of filter system names chart,
soft lights, 388–389 827
tungsten-halogen floodlights, 388, 391 Mired-coordinated filters, 163
Luminance Mired shift value (filter) effects examples table,
definition, 415 835
masks creation, 331 Mired value, 163
Mired values of color temperatures from 2000° to
M 10,000°K table, 835
Macbeth Color Checker color chart, 302, 342 Mist filters, 159
Macro-focus lens, 139 Mitchell-based platform-mounted stabilization
Mastering, digital basic concepts and, 27–28 system, 181
MatchLens handheld app, 845 Mitchell FC, BFC 65mm camera, 584–585
Mattes, traveling, 316, 325 Mitchell Mk-2 camera modified, 227
Measurement conversion factors, 688 MKII 4W and H/S IMAX camera, 598–600
Medium-angle lens, 117f, 119, 121f MK-V Evolution Modular Sled system, 177
Megapixels and evaluating digital motion picture Modulation transfer function (MTF), 111, 137,
cameras, 32, 34 263–264
Mercury-vapor lamps, 213 Moiré pattern definition, 415
Mesmerizers, 133, 134 Monitor set-up, quick picture, 899–900
Metadata definition, 415 Monitoring issues for HD video on digital
Micro Reciprocal Degrees (MIRED) system, 375–376 still camera, 51
Microwave transmitters and safety, 424 Moon and sun calculator handheld app, 846
Miniatures: camera speed, model speed, Motion-control cinematography, 319–323
exposure factors versus miniature scale chart, capabilities, cinematic, 320
896 definition, 319–320
Miniatures, photographing, 363–365 equipment and field, 322–323
camera speed calculation, 365 equipment and visual effects studio, 321–322
information needed, 363 technique, 323
model size, 364 tracking extends cinematic capabilities,
shooting speeds, 364–365 320–321
Miniatures with full-scale live-action actors Motion control data definition, 415
in-camera compositing, 367–373 Motion Pictures Expert Group (MPEG) definition,
advantages of hanging miniatures, some, 367 415
In-20
|
Index
Movie*Slate handheld app, 849 motion adding, 315

Moviecam compact 35mm camera, 533–536 quality manipulation, 315
Moviecam SL 35mm camera, 536–538 size or position change, 314
Moviecam Super America MK 2, 538–540 split screen, 315
MSM 9801 IMAX camera, 600–601 superimpose, 315
MSM Model 8870 65mm/8-perf camera, 585–586 systems, new, 317–318
Multiple filter use, 168 transitional effects, 314
traveling mattes, 316
N uses, general, 316–317
zoom, optical, 315–315
National Television Systems Committee (NTSC)
definition, 416 O-rings, cleaning and checking, 231–232
Nautica Super 8mm camera, 580 Out of gamut definition, 19
Negative color film and day-for-night Output device transform (ODT), 21, 286, 287,
cinematography, 209–210 287f, 288
Neutral-density (ND) filters, 148–152 Oversampling definition, 416
circular gradient and circular/elliptical
center spot, 152 P
daylight correction table, 829 Pan and scan
infrared, 149–150 anamorphic, 75
graduated, 151f, 151–152 common topline format versus, 76
selection chart, 824 definition, 71
types, 148 HDTV 16 x 9, 72
variable, 148–149 home video masters and, 269
Nikon Super 35 composed for 2.40 aspect ratio and,
comparison of filter system names chart, 827 81
D800 camera, 55 video transfer and, 71, 72
R8, R10 Super 8mm cameras, 578 Panaflex 16mm camera (Elaine), 575–577
Nila JNH series, 397 Panalog definition, 416
Noise definition, 415 Panasonic cameras
Nonlinear editing definition, 416 AG-AC160, 55
NTSC AG-AF100, 55
aspect ratio, 38 AG-DVX100, 37
definition of 7.15 IRE, 13 AG-HMC150P, 55
frame rate, 41 AG-HPX250PJ, 55
television standard 30 frames per second, 16 AG-HVX200, 37
Nuke IBK (Image-based Keyer), 361 AG-HX370, 54
AJ-HDC27H VariCam, 613–614
O AJ-HPX3700 VariCam, 614–615
Ohm’s Law and formulas, 692 AVCCAM AG-HMC40, 56
Onboard cinematography, 215 Genesis, 615–617
Open EXR definition, 416 Lumix DMC-GH2, 56
Optical breathing, 129–130 Panasonic external recorder, 46
Optical disk system, 45 Panavision. See Anamorphic cinematography
Optical low pass filter (OLPF), 158 Panavision 35mm cameras
Optical printer, cinemagic of, 313–318 Gil Golden Panaflex (PFX-GII), 540–542
anamorphic conversions, 316–317 Millenium (PFX-M), 544–546
blow-ups and reductions, 316 Millenium XL (PFXMXL), XL2, 542–544
doctoring, modifying, and salvaging, 317 models list, 552
frame sequence modification, 314 notes and accessories for, 550–553
history, brief, 313–314 Panaflex Panastar II High-Speed, 547–548
|In-21
Index
Panaflex-X (PFX-X), 546–547 Portable DTE recorders, 46

Platinum Panaflex 35mm (PFX-P), 548–550 Postproduction issues for shooting HD video on
Panavision 65mm cameras digital still camera, 52
System-65, 586–589 Postproduction responsibilities, 5–6
System-65mm Handholdable, 589–590 Potentiometers, 321
Panavision Genesis camera, 12 Power calculator handheld app, 848
Panavision Panascourt handheld app, 845 Power function definition, 417
Panavision SSR-1 onboard recorder, 642 Power load calculator, 852
Panavision Solid State Recorder SSR-1, 271 Precision fader, 384
Panning speeds tables, 891–893 Preparation of digital camera equipment,
Par fixtures and lamps, 389 447–466
PDF Expert for iPad handheld app, 850 back focus adjustment steps, 464–465
Pedestal definition, 416 checkout, 450–460
Pentax Digital 1 Degree Spot Meter choices and working style, 448–450
exposure meter, 103 color space, bit depth, and sampling
Pentax Spotmeter V exposure meter, 103 ratio selection, 457
Permanent Gray Card, 301 crew preparation, 451–456
Persistence of vision definition, 416 daily prep and maintenance, 463–465
Petro Vlahos Technique, 352, 353 delivery of elements to post dailies, editorial,
Photo fx handheld app, 847 VFX, and DI, 465–466
Photodiodes sensors, 34, 85f, 86 digital camera preparation steps, 452–456
Photogates sensors, 34
exposure/dynamic range test, 462
Photometric calculation tool handheld app, 846
factors to consider, 448–449
Photoshop definition, 416
field calibration setup steps with SMPTE bars,
Photo-Sonics 35mm 4CR camera, 553–555
463
Photo-Sonics 35mm 4ER camera, 555–556
format settings, 456–457
Photo-Sonics 35mm 4ML (reflex and nonreflex),
frame rate, scan type, and resolution setting,
556–557
457
Photo tools handheld app, 844
image and color settings, 459–460
Picture monitor set-up, quick, 899–900
in-house preparation, 451
Picture quality issues for shooting HD video n
digital still camera, 52 inventory and master production
Pixels (picture element) equipment list, 450–451
check steps, 451 look-management preparation, 449–450
chips and, digital basic concepts and, 10–12 menus, setting up, 456
definitions, 34, 85, 416–417 monitor check and calibration, 455–4564
digital motion picture cameras and, 85–86 objective, 447
Pixels (picture element) and chips, digital pixel check stops, 453
basic concepts, 10–12 responsibilities of prep crew, 447–450
Pocket DIT handheld app, 847 sensor cleaning steps, 453–454
PocketLD handheld app, 847 targets for testing, useful, 460–461
Pogocam, 180 testing, 460–462
Pola filter, 819 time code, 458–459, 462
Pola screens, 207, 208, 210 universal preparation procedures, 451–452
Polarized light definition, 153 viewfinder functions, 455
Polarizing filers, 152–155 viewing environment guidelines, 464
3-D and, 154–155 workflow chronological list, 447–448
circular, 154 workflow test, 461
digital intermediate and filtering for, 171 Preparation of motion picture film camera
multiple filter use and, 170 equipment, 427–445
variable-density filter and, 154 accessories, 438
In-22
|
Index
batteries and cables, 433–434 pitchvis, 188

blimp housing, 436 postvis, 189, 191
camera assistant, 444–445 problem solving, additional, 190–191
camera body, 431–432 process step-by-step, 189–191
checkout of equipment, 429–438 sequence animation, 189–190
film tests, 441–442 sequence cutting, 190
filters, 436 study and review, 190
flange focal depth (flange to aperture-plane technical previs, 188
depth), 432 virtual shoot, 190
fluids, useful, 442 Previsualization Society, 188
format (movement, aperture plate, lens Primaries definition, 417
position, assist monitor), 432–433 Primary color correction definition, 417
ground glass, 433 Primatte, 361
inventory, 427–428 Prime lenses
invoice check, 428 components, 129
lamps and LEDs, 434 depth of field and, 136
lens focus calibration, 438–439 ergonomics and, 138
lenses, 435 future of, 139
magazine, 437 zoom lenses versus, 129–130, 139
matte box, 437 Principal photography responsibilities, 4–5
ramping, 438 Printer light, finding your own, 305–310;
scratch test, 439–440 see also Emulsion test preparations
shooting, daily preparation for, 440–441 to establish a single printer light (tungsten)
spreader, 429 Printer lights adjustment to match a sample clip,
steadiness test, 440 311–312
test procedure, 439 Prism system, three-chip, 10f
Pro8mm Pro II Super 8mm camera, 580–581
time code, 438
ProCyc’s Pro Matte plastic material, 336, 345–346
tools, 442–444
Progressive definition, 417
tools list, optional, 444
Projection chart for process backgrounds, 897
tools list, standard, 442–444
Propeller aircraft, customized, 223
tripod head, 429–430
Prosumer cameras
tripods, 429
aspect ratio, 39
variable shutter, 434
definition, 38
video assist (video camera, monitor,
depth of field and, 48
recording device), 437
HD-SD1 output, 48
viewfinder, 434
interlace-scan, 42
zoom lens, 435–436
sensor size and, 49
zoom motor and control, 436
Pro-TM camera stabilizing systems, 175–176
Preproduction planning responsibilities, 3–4
Pull processing, 289, 299, 302
Previsualization (previs), 187–191 Push processing, 289, 295, 296, 299, 302
assets assembly, 189
benchmarks for successful previs, some, 191
cinematographers and, 191 Q
computers, before, 187 Quantization artifact definition, 417
crew communication and outreach, 190
definitions and types, 187, 188–189 R
digital dailies and, 260 Radial distortion, 233
digital development, 187 Raid array definition, 417
introduction, 187–188 Rain spinners, 171
on-set previs, 188–189, 191 Raw files definition, 417
|In-23
Index
Raw versus de-Bayered, digital basic concepts, diffusion materials, 347

16 lighting filers, 383
REC. 709, 20 Rotoscoping (roto), 334, 338, 343
Recorders, external, 46–47
Recording formats, digital, 44–45 S
Recording HD video on digital still camera, 51 Safety on the set, 423–425
Red Epic camera, 358 Salvaging, doctoring, and modifying with
Red Epic digital camera, 617–618 optical printer, 317
Red One digital camera, 619–620 Scan definition, 418
Red Scarlet camera, 46, 54 Scanner versus cameras and digit motion picture
Red Scarlet-X digital camera, 618–619 cameras, 32–33, 84–85
Redcode raw format, 16, 45 Scanning and digital postproduction for
Redundancy definition, 417 feature film, 261–262, 262f
Reflected-ultraviolet photography, 211 Scanning video—interlace or progressive, 16–18
Refraction and underwater cinematography, 231, Scattering and underwater cinematography, 231
232, 233, 234 Scene-referred data definition, 418
Reflection(s) Screen heights importance, 60–62
polarizing filters and, 171 Screen ratios, most common, 8
secondary, 170 Scrim, 391, 392
underwater cinematography and, 230 Secondary color correction definition, 419
Reflector boards, 393 Sekonic exposure meters
Region of interest correction definition, 417 L 164 Marine meter, 235–236
Registration (color) definition, 418 L 308 BII, 105
Releases handheld apps, 848, 849 L 358, 106
Remotevision, 181 L 398M, 105
Render definition, 418 L 508C, 103–104
Rental orders handheld app, 848 L 558, 106–107
Resolution, digital basic concepts, 12–13 L 608C, 104–105
common image standards, 13 L 778 Dual Spot F exposure measure, 107
definitions, 12, 34, 86 Sensor definition, 419
Resolution definition, 418 Sensor size
Resolution versus megapixels and digital depth of field and, 48–49
motion picture cameras, 32 HD video shooting on digital still camera, 49
Responsibilities of cinematographers, 3–6 Sensors and amount of light they capture, digital,
designing look, 3 33–34
postproduction, 5–6 Sepia filters, 166
preproduction planning, 3–4 Shadows
principal photography, 4–5 multiple shadows and LED lighting, 396
Reversal color film and day-for-night Shooting practices formulas, 681–684
cinematography, 209 computing exposure times, 682
RGB (red, green, blue) footage in feet and frames, 682
definition, 418 footage versus time relationship, 681
stripe pattern, 11f, 12 frames rates for miniatures, 682–683
Rigid/hard mounts (fixed wing or helicopter), 226 high speed formulas, 683
Ringing definition, 418 image blur, 684–685
Robodog, 180 running times, feet, and frames, 681
Rock and Frazier lens system, The, 132 Shooting to screen time calculator
Rollvision stabilization system, 181 handheld app, 847
Rosco Shutter angle compensator for constant
CalColor line of primaries, 341 exposure chart, 873
In-24
|
Index
Shutter angle exposure compensation chart, XDCAM PMW-EX3 camera, 54

reduced, 875 XDCAM MW-F3 camera, 54
Shutter angle/f.p.s./T-stop change chart, 872 Spacecam (dedicated mass-gyro system), 227
Shutter speeds chart, 876–877 Spacial corrections definition, 419
Silicon Imaging SI-2K digital camera, 621–622 Spacial resolution definition, 419
Sin City and digital cameras, 31 Special effect filters, 155–160
Skydiving cinematography, 216 contrast control, 159–160
Slant-focus lenses, 122, 132–133 diffusion, 155–158
Snoots, 391 general information, 155
Society of Motion Pictures and Television optical diffusion in digital camera, 158–159
Engineers (SMPTE) sliding diffusion, 158
59-1998, 122 star and flare effects, 160
125 M standard, 414 Specialty lenses and systems, 130, 132–133
CinemaScope aspect ratio, 62n1 Spectra exposure meters
DC28, 266 Spectra Cinespot Foot Lambert Meter, 108
DCI/SMPTE Reference Projector Spectra Cinespot IE Spot Meter, 108
specifications, 268 Spectra Professional IV, 109
Recommended Practice for Critical Viewing Spectra Professional IV-A, 109-110
Conditions for the Evaluation of Color Speed and streaking of hot (bright) objects, 128
Television Pictures (RP 166), 269, 462
Spherical lenses
RP431-2-2007, 287f
anamorphic lenses versus, 124–129, 127f
safe image area for digital projection, 269
image produced by, 140
ST2065-1:2012, 284
Spirit Datacine, 260, 261
ST2065-2:2012, 284
Split diopter close-focusing lenses, 133
ST2065-3:2012, 284
Split screen technique, 315
standard for motion picture projection, 384
Spot meters, 301
standards for digital cinema release, 28
Spydercam, 184–185
Soft clip definition, 419
Stabilizers in the air, alternate, 225–226
Soft contrast filter, 159
SoftSun 5400°K fixtures chart, 919 Standard definition defined, 419
Solid-state flash memory cards and drives, 45–47 Star and flare effects filters, 160
Sony Star Wars Episode 3 and digital cameras, 31
external recorder, 46 Steadicam. See Tiffen Steadicam
F23 CineAlta digital camera, 357, 622–624 Steady tester, 428
F35 CineAlta digital camera, 357, 624–626 Stopwatch handheld app, 851
F65, 12, 14, 331, 358 Storage and HD video on digital still cameras, 51
F65 CineAlta digital camera, 626–627 Storage Area Network (SAN), 264
HDC-F950 CineAlta digital camera, 627–628 Storage requirement calculations
HDR-AX2000, 55–56 handheld app, 847, 850
HDW-F900 CineAlta digital camera, 628–630 Storyboard composer handheld app, 845
HDW-F900 HDCAM pro camcorder, 37 Stripe filters, 165
HDW-F900R CineAlta, 630–631 S.Two OB-1 onboard recorder, 645–646
NEX-FS100U camera, 56 Subtractive color definition, 419
NEX-FS700U camera, 54 Sun and moon calculator handheld app, 846
NEX-VG10 camera, 56 Sun position calculator handheld apps, 844, 846
NXCAM HXR-NX5U camera, 56 Sunpath computer software charts, 855–863
SR-R1 onboard recorder, 643 daylight hours, 858–859
SR-R4 onboard recorder, 643–644 height of sun, maximum, 857–858
SRW-1 onboard recorder, 644–645 latitudes of world cities, 855–856
XDCAM EX cameras, 46 sea level sunrise/sunset, 860–863
XDCAM PMW-EX1R camera, 56 Sunrise and sunset calculator, 853
|In-25
Index
Super 8mm cameras, 578–581 object appearance in front of screen, 197–198,

6.5mm field of view table, 804 198f, 200
13mm field of view table, 805 plane of convergence definition, 194–195
38mm field of view table, 806 proscenium arch, 195, 200, 201
50mm field of view table, 807 resurgence reason, 193
footage table, 878 right eye (master eye or dominant eye)
footage/time table, 879 definition, 193
Super 35 formats, 73–81 rigs weight and, 193
advantages of, composed for 2.40 screen plane definition, 194
aspect ratio, 76–78 stereoscopic (binocular vision) definition, 193
camera to screen, from, 74f TIFF file definition, 419
common center, 75, 75f, 76 Tiffen Steadicam, 178–180
common topline, 75, 75f, 76 aerial cinematography and, 215
digital intermediate and, 73 AQUACOLOR, 235
disadvantages of, composed for 2.40 buoyancy control and underwater
aspect ratio, 78–81 cinematography, 229
other names for, 73 camera-support systems and, 465
scanning and, 262 decamired filters table, 834
Superman Returns and digital cameras, 31 Master Series, 178–179
Superimpose, 315 Master Series Elite, 179–180
Support systems for camera. mount for aerial cinematography, 225
See Camera-support systems Ultra Elite, Ultra Cine, 179
Sync box, 252 Universal Model III, 178
Vector 12, Vector EL, Vector CN, 179
Tilting filters, 170
T Time codes preparation, 458–459, 462
Technovision. See Anamorphic cinematography Time lapse chart, 895
Television displays, filming, 251–255 Time/speed effects tables, 894–896
color temperature, 254–255 Titanic
frames rates and exposure time, 251–252 lens requirements and, 135
obstacles, principal, 251 Panavision/Frazier lens system and camera,
over- and undercranking, 252 132
plasma displays, 253 Tonal mapping definition, 419
Tessive Time filter, 150 Tone control filters for black and white, 160–161
Thomson Spirit Datacine, 261 TouchDraw handheld app, 850
3-D Traveling mattes, 316, 319
definition, 31n1 Tropical cinematography, 244–249
polarization filters and, 154–155 biological factors, 244, 246, 247
3-D stereoscopic cinematography, 193–205 black-and-white film, 247–248
audience position, 199–200, 200f color film, 248–249
convergence definition, 194, 194f dehumidifiers, 246
editing constraints, 201–202 dew point, 244
floating objects, 201–202 equipment preparation and protection, 246–247
glasses, 196, 202–205 relative humidity and, 244, 245, 246, 247
higher frame rates usage, 193 storage of photographic materials, 245
interocular (interaxial) definition, 195f, 195 Truelight, 23
left eye definition, 193 T-stops (transmission-stops), 57–58
monocular depth perception, 193–194, calibration of, 57–58
196, 201 compensation for camera speed table, 870–871
object appearance behind screen, 197, 197f, definition, 57
198f, 199, 200, 200f, 202 scale, 138
In-26
|
Index
Tungsten fresnels chart, 902–903 Video scanning—interlace or progressive,

Tungsten-halogen floodlights, 388, 289f, 837 digital basic concepts, 16–18
Twin fixed wing aircraft, 222–223 Video storage calculator, 852
Video versus data, digital basic concepts, 15–16
U Video walls, filming, 254
Ultimatte Preview Viewfinder handheld app, 844, 851
compositing software, 360–361 Vision Research Phantom digital cameras,
digital, 355 632–638
video, 343 Vision Research Phantom CineMag-II onboard
Ultra contrast filter, 160 recorder, 646–647
Ultracam 35mm camera, 557–559 Vistacam, 227
Ultraviolet cinematography, 211–214 VistaVision, 63
considerations, special, 214 8-Perf Horz. Pull Across aperture specs, 8
exposure determination, 213 8-perforation film, 120
eye protection, 212 35mm cameras, 591–595
films, 213 high quality anamorphic release prints and, 87
fluorescence (or black-light) photography, Visually lossless definition, 420
211–212 Voltage drop of copperwire, 693
mercury-vapor lamps, 213
precautions, 212
reflected-ultraviolet photography, 211
W
Watermark definition, 420
Ultraviolet filters, 145, 146
Waveform monitor definition, 421
Underwater cinematography, 229–238
Weather conditions display handheld apps,
blue- and greenscreen, 237, 238, 346–347
851, 853
buoyancy control, 229, 231
color absorption and, 231 White point definition, 421
color correcting for, 230 Wide-angle lens, 119, 121f
communications underwater, 236 Wilcam 35mm cameras
distance calculator handheld app, 846, 847 W-7 VistaVision High Speed (pic, one diagram),
equipment, 231 590–591
equipment care, 231–232 W-9 VistaVision Lightweight, 592–593
exposure meters, 235–236 W-11 VistaVision Sound, 593–594
filters, 166–167, 235, 826 Wrap Time Lite handheld app, 848
flat or dome lens ports, 232–233 Wratten filters
HD underwater, 229–230 # 1A, 823
lens calibration underwater, 233–235 # 2A, 213, 823
lens selection, 234–235 # 2B, 212, 213, 823
light is changed by water, how, 230–231 # 2C, 823
lighting, 236–237 # 2E, 212, 823
lighting safety, 238 # 3, 212, 819, 820
reflection, 28 # 8, 160, 819, 820
refraction, 231, 232, 234 # 11, 161, 819, 820
scattering, 231 # 12, 211, 819, 820
service kit, 232 # 15, 160, 819, 820
Underwater color-correction filters, 166–167, 826 # 16, 160
# 18A, 211, 213
V # 21, 819, 820
Vectorscope definition, 419–420 # 23A, 209, 819, 820
Vertical angle versus effective focal length table, # 25, 160, 210, 819, 820
808 # 29, 160, 210, 819, 820
Vertical blanking, 252 # 47, 819, 820
|In-27
Index
# 56, 209, 819, 820 checking out, 433–434

# 58, 819, 820 cine zoom lens, 139
# 70, 210 color sampling and, 131f
# 80, 162 computers and design of, 139
# 80A, 827 depth of field and, 136
# 80B, 827 ergonomics and, 138, 139
# 80C, 147, 827 extension tubes and, 668
# 80D, 147, 827 fixed focal-length lens (or prime) and, 116, 118
# 81, 162 focus and, 127
# 81A, 827 optical, 314–315
# 81B, 827 optical breathing and, 130
# 81C, 827 popularity reasons, 129
# 81D, 827 prime lenses versus, 129–130, 139
# 81EF, 827 scales, 138
# 82, 162, 827 spherical zoom-lens cam technology, 128
# 82A, 827 split-field diopter lenses and, 670–671
# 82B, 827 telephoto, 121f
# 82C, 827 Zylight, 393
# 85, 148, 151, 162, 167, 209, 827
# 85B, 162, 827
# 85C, 827
# 87, 146
# 89, 146, 208
# 98 blue filter, 329
# 99 green filter, 329
comparison of filter system names chart, 827
HF-3, 823
HF-4, 823
HF-5, 823
Wratten numbers, 145
WYSIWYG definition, 421
X
XDCAM format, 45, 46
Xenon light sources, 383
AC/DC low pressure arc lamps, 384
cautions, 385
characteristics, typical, 837
flash tubes, 384, 385
XYZ definition, 421
Y
YCrCb definition, 421
YMC (yellow, cyan, magenta) definition, 421
Z
Zoom lenses, 120
anamorphic lenses and, 119, 120
aspherical surfaces and, 118
In-28
|
Index
HOLLYWOOD

Ac Manual10 Vol1

Uploaded by

Copyright:

Available Formats

Ac Manual10 Vol1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ac Manual10 Vol1

Uploaded by

Copyright:

Available Formats

AMERICAN

The ASC Press

Hollywood, California, USA

Cover design by Martha Winterhalter

Michael Goi, ASC

Arctic and Tropical Cinematography Vol. I 239

Digital Postproduction for Feature Films Vol. I 257

ASC Color Decision List Vol. I 273

The Academy of Motion Picture Arts and Sciences

The Cinematographer and the Laboratory Vol. I 289

Emulsion Testing Vol. I 299

Finding Your Own Printer Light Vol. I 305

Adjusting Printer Lights to Match Sample Clips Vol. I 311

Cinemagic of the Optical Printer Vol. I 313

Motion-Control Cinematography Vol. I 319

Greenscreen and Bluescreen Photography Vol. I 325

Photographing Miniatures Vol. I 363

In-Camera Compositing of Miniatures with Full-Scale

Light Sources, Luminaires and Lighting Filters Vol. I 375

An Introduction to Digital Terminology Vol. I 401

Safety On The Set Vol. II 423

Shutter Compensation Vol. II 872

Providing you with the industry’s widest selection of

Sony TRIMASTER™ EL monitors.

Visit sony.com/TrimasterEL to arrange an on-site demo.

Rental Depar t ment

On Location Dailies Deliverables including Complete 3D capabilities including

EC3, EFILM and Company 3 are Deluxe Companies

GO WHEREVER THE STORY TAKES YOU.

GET STARTED. CONTACT US: 855.CINE.EOS - CINEMAEOS.USA.CANON.COM

VFX Digital Cinema

3780_LF_CineAd_AW_5.5x8.5_V2.indd 1 16/11/2012 10:58

Toland is an application that provides solutions to the technical

‘Flicker Free’ warning

POCKET ALCHEMY CAN BE FOUND AT:

1212_ACManual2012_F.indd 1 12/4/12 3:00 PM

Profile. Calibrate. Measure.

ECO PUNCH PLUS

JAB HURRICANE SPACE LED LIGHT

Germany & 20 Euro Countries Switzerland Canada

Film & Digital

As the largest camera rental house north of LA, we

Purveyors of Fine Cine Cameras

seattle 206-467-8666 toll free 877-467-8666 marty @oppcam.com www.oppenheimercinerental.com

CineMonitorHD, Creatively Connected

Eclipse XR Series Wescam System Cineflex

A lot easier and much more accurate than

POCKET ALCHEMY CAN BE FOUND AT:

Francis Kenny, ASC

“When shooting a multi-camera

Nancy Schreiber, ASC ...but it’s nothing

Bill Bennett, ASC

“Schneider came through for us

Lowell Peterson, ASC

www.schneideroptics.com Phone: 818-766-3715 • 800-228-1254 Classic Filmmakers Choose Schneider

American Society of Cinematographers’ clubhouse.

DESIGNING THE LOOK

F ilmmaking is a uniquely collaborative form of artistic expression with

w 16mm Film Apertures 1.33:1 (4:3) Television Safe Area

w 65mm - 15-Perf IMAX/OMNIMAX