You are on page 1of 52

Playing games for a living . . .

Its good to be evil.


Alex Careld, Evil Cenius, puI Ihe lnIel' Ccre` i7 Prccesscr
ExIreme EdiIicn Ic Ihe ulIimaIe IesI. The Challenge. Cculd
iI per!crm righI cuI c! Ihe bcx without turning anything off
. . . and ncI drcp a !rame while playing CryIek's Crysis
*
?
The Result: Oh yes, it could.
Read Ihe !ull sIcry Icday.
www.intel.com/go/visualadrenaline
Z00B, lnIel CcrpcraIicn. All righIs reserved. lnIel and Ihe lnIel lcgc are Irademarks c! lnIel CcrpcraIicn in Ihe U.S. and cIher ccunIries.
*0Iher names and brands are Ihe prcperIy c! Iheir respecIive cwners.
If I were going to give advice to a gamer today
I would just say, Dont compromise. Go for a
processor like the Intel Core i7, which is easily
the best processor Ive ever used.
A|ex are|d, ExecuI|ve 0|recIcr, Team Ev|| en|uses
And nd cuI mcre abcuI cIher game inncvaIcrs, Iechnclcgies,
and Irends, by subscribing Ic lnIel' Sc!Iware 0ispaIch !cr
Visual Adrenaline. lI's FREEl www.intelsoftwaregraphics.com
CGW :109_p 12/23/08 10:14 AM Page 1
ON THE COVER
SEE IT I N
For more than a year and a half, the team at Digital Domain has
had countless senior moments, breaking new ground to create
the realistic digital human starring in the rst 50-plus minutes
of the lm The Curious Case of Benjamin Button, pg. 8.
High denition and audio for
indie lms.
Benjamin Buttons post workow.
Director Steven Soderbergh discusses
the lm Che.
Visual effects in commercials.
Features
Whats Old Is New Again
8
Digital Domain reaches a milestone in computer graphics, creating an emotional
performance by a realistic digital human in a lead feature-lm role for The
Curious Case of Benjamin Button.
By Barbara Robertson
Bold Camera, Subtle Animation,
Tactile World
18
What happens when the typical animation process is turned upside down?
You get Despereaux, a computer-generated fairy tale with stunning epic
environments and detailed characters both big and small.
By Barbara Robertson
Dancing with the Ghosts in the Machine
26
Advances in computer graphics technology has helped shape and expand
3D virtual worlds, where people assume unique identities and explore
unusual cyber lands while they socialize with others.
By Kathleen Maher
Alter Ego
32
Filmmakers nd ways to capture CG in-game sequences and integrate them
into scenes within a live-action documentary.
By Karen Moltenbrey
Autodesk Builds Out and In
36
At this years annual Autodesk University, users of the companys growing
product lines, particularly in the CAD and manufacturing sectors, keep up to
date with the latest industry trends and technologies.
By Kathleen Maher
Dancing with the
Ghosts in the Ma-
chine
26
Advances in computer
graphics technology has
helped shape and expand 3D
virtual worlds, where people
assume unique identities and
Bold Camera, Sub-
tle Animation,
Tactile World
18
What happens when the typical
animation process is turned upside
down? You get Despereaux, a computer-
generated fairy tale with stunning epic
environments and detailed characters both
big and small.
By Barbara Robertson
COVER STORY
January 2009 1
January 2009 Volume 32 Number 1 I n n o v a t i o n s i n v i s u a l c o m p u t i n g f o r t h e g l o b a l D C C c o m m u n i t y
Departments
Editors Note Living a Virtual Existence
2
Massively multiplayer online role-playing games, or MMORPGs, are gaining in
popularity, leading players of these online games to live double livesreal and
virtual. Why are they so popular, and where do people nd the time?
Spotlight
4
Products Nvidias Quadro FX 5800, 4800, and Quadro Plex 2200 S4.
SGIs VUE. Vicons T10 camera. Organic Motions LitePod. Xsenss Moven
successor. AMDs Shanghai.
Portfolio
40
The Art of the Game: Artwork from Robert Ludlums The Bourne
Conspiracy game.
Knowledge & Career
42
The strengths, weaknesses, opportunities, threats, and outlook that
recruiters face during the year ahead.
Back Products
48
Recent software and hardware releases.
Karen Moltenbrey
CHIEF EDITOR
karen@CGW.com
August 2008
EditorsNote
T
hey come from all walks of lifeteachers, police officers, accountants, lawyers,
doctors, waitresses, students, retail clerksthe list is endless. And they have one
purpose: to assume an alternate identity in the cyber worlds of massively mul-
tiplayer online role-playing games (MMORPGs) or in online communities. They
know the difference between reality and fantasy, but sometimes the line between
the two gets a bit blurred as these players storm medieval
castles, discuss the latest fashion trends with alien life
forms, locate untold treasures, and engage rival clans.
While the concept of MMORPGs and virtual communi-
ties is not new, their popularity continues to soar. In fact,
there is a site for just about everyone, young and old alike.
Last year my son informed me that he had to meet his friends
online in Club Penguin, a social networking site for young-
sters. They can accumulate money to purchase upgrades
for their igloos, participate in contests, and, yes, socialize.
In the world of MMORPGs, players assume a certain
look and role that reflects the particular theme of the game. For instance, in Star
Wars Galaxies, a player can choose among various species from the Star Wars universe
(Wookiee, Rodian, human) and professions (Jedi, bounty hunter, smuggler). During
gameplay, the person joins other players to accomplish certain tasks and goals.
Then there are virtual communities, which are similar to MMORPGs in that play-
ers assume identities inside cyber worlds. However, they are not confined to a theme,
unless, of course, they choose to put down roots within a special interest area. In
summary, the main point of an MMORPG is gameplay; socializing is secondary.
In an online virtual world, socialization is king.
In this issue, we offer a look at both types of cyber living. In Dancing with the
Ghosts in the Machine (pg. 26), the writer discusses the growing phenomenon
known as virtual worlds and the role that 3D computer graphics technologies play in
those universes. In Alter Ego (pg. 32), indie filmmakers look at the addictive nature
of MMORPGs and what happens when real life interferes with virtual life. In this
project, avatars are introduced along with their human counterpartsa process that
required two methods of capturing and integrating game footage into the live-action
documentary.
The bottom line is that the cyber worlds of MMORPGs and online communi-
ties hold a big attraction for players. Many have an alternate life inside these games,
complete with a dwelling, a job, a bank account of sorts, and a social network. They
participate in activities. They accumulate wealth and goods. They make friendsand
enemies. They develop skills. Perhaps one of the more inviting aspects is the ability to
re-create oneself. For instance, a conservative-looking businessman may choose to be
a punk rocker with neon pink hair and multiple piercings. Tired of being a brunette?
No problem: go blonde. Tired of being a human? Be a dog instead.
Personally, I do not have the time to invest in a virtual me. Its hard enough keeping
up with the real me. However, devotees will find the time, just as other folks find the
time for their favorite hobbies. Perhaps one day I will create a virtual me and show up
in a cyber destination, perhaps somewhere warm and sunny. n
Living a Virtual Existence
The Magazine for Digital Content Professionals
EDITORIAL
KAREN MOLTENBREY
Chief Editor
karen@cgw.com (603) 432-7568
36 East Nashua Road
Windham, NH 03087
CONTRIBUTING EDITORS
Courtney Howard, Jenny Donelan,
Audrey Doyle, Evan Marc Hirsch, George Maestri,
Kathleen Maher, Martin McEachern,
Stephen Porter, Barbara Robertson
WILLIAM R. RITTWAGE
Publisher, President and CEO,
COP Communications
SALES
LISA BLACK
National Sales Manager
Classifieds Education Recruitment
lisab@copcomm.com (877) CGW-POST [249-7678]
fax: (214) 260-1127
Editorial Office / LA Sales Office:
620 West Elk Avenue, Glendale, CA 91204
(800) 280-6446
PRODUCTI ON
KATH CUNNINGHAM
Production Director
kcunningham@cgw.com (818) 291-1113
MICHAEL VIGGIANO
Art Director
mviggiano@copcomm.com
CHRIS SALCIDO
Account Representative
csalcido@copprints.com (818) 291-1144
Computer Graphics World Magazine
is published by Computer Graphics World,
a COP Communications company.
Computer Graphics World does not verify any claims or
other information appearing in any of the advertisements
contained in the publication, and cannot take any
responsibility for any losses or other damages incurred
by readers in reliance on such content.
Computer Graphics World cannot be held responsible for
the safekeeping or return of unsolicited articles,
manuscripts, photographs, illustrations or other materials.
Address all subscription correspondence to: Computer
Graphics World, 620 West Elk Ave, Glendale, CA 91204.
Subscriptions are available free to qualified individuals
within the United States. Non-qualified subscription rates:
USA$72 for 1 year, $98 for 2 years; Canadian
subscriptions $98 for 1 year and $136 for 2 years;
all other countries$150 for 1 year and $208 for 2 years.
Digital subscriptions are available for $27 per year.
Subscribers can also contact customer service by calling
(800) 280 6446, opt 2 (publishing), opt 1 (subscriptions) or
sending an email to csr@cgw.com. Change of address can
be made online at http://www.omeda.com/cgw/ and click
on customer service assistance.
Postmaster: Send Address Changes to
Computer Graphics World, P.O. Box 3551,
Northbrook, IL 60065-3551
Please send customer service inquiries to
620 W. Elk Ave., Glendale, CA 91204
2 January 2009
www.aja.com
800.251.4224
Imagine true broadcast-level HD editing in places you never thought possible.
Io HD, paired up with Apples MacBook Pro and Final Cut Studio 2,
gets you a no-compromises HD editing suite wherever you want to work.
One FireWire cable to the Mac is all you need to work with full-resolution 10-bit
4:2:2 HD and SD video, using the powerful new ProRes 422 codec from Apple,
uniquely integrated directly into Io HDs hardware.
Add in Io HDs 10-bit up/down/cross conversion and unmatched video and audio
connectivity, and youre seamlessly working in any format you want,
anywhere you need it... in a portable, professional package.
Check out Io HD at our website, or give us a call to fnd an authorized
AJA Desktop Dealer near you.

A view from your edit suite.


CGW_ :1207_p 12/5/07 10:36 AM Page 1
4 January 2009
PRODUCT: GRAPHICS CARDS
Nvidia has made available its most powerful professional graph-
ics card yet, the Quadro FX 5800. This new card is tuned
for rapidly visualizing and interpreting massive data-
sets, which, until now, were unattainable on a worksta-
tion graphics board.
The card offers as many as 240 CUDA program-
mable parallel cores and the industrys rst 4GB graph-
ics memory. Other advanced features include: interactive 4D
modeling with time-lapse capabilities, massive memory band-
width of up to 102GB/sec, and ll rates that exceed 52 billion
texels per second and geometry performance of 300 million
triangles per second. Furthermore, the card supports next-
generation OpenGL and Microsoft DirectX 10 applications.
The Quadro FX 5800 GPU features true 10-bit color, enabling
billions, rather than millions, of color variations for rich, vivid
image quality with broad dynamic range for greater realism. The
Quadro FX 5800 graphics board, which is priced at $3499, is
available now.
The company also introduced the Quadro FX 4800 ultra-
high-end professional graphics card with advances in GPU
architecture, added graphics memory, and 30-bit color.
Equipped with 192 CUDAT parallel computing cores, the
Quadro FX 4800 delivers a 2x performance increase over previ-
ous generations of ultra-high-end graphics cards. Equipped with
1.5GB graphics memory, the Quadro FX 4800 has the capac-
ity to process large
textures and frames in
real time while providing ll
rates of 38 billion texels per second
and geometry performance of 300 million trian-
gles per second. The Quadro FX 4800 is avail-
able today for $1999.
In other news, Nvidia delivered the Quadro Plex 2200
S4 Visual Computing System (VCS), a 3D accelerator for
remote graphics visualization and rendering. Designed to
address demanding remote graphics and off-line rendering
challenges, the Quadro Plex 2200 S4 VCS is available in a
standard 1U form factor for large-scale server deployment.
The 2200 S4 contains four Quadro FX 5800 GPUs, with a total
frame buffer of 16GB (4GB per GPU), for real-time processing
of large textures and frames, and full-scene anti-aliasing. It also
sports a CUDA parallel computing processor (supporting as
many as 960 CUDA cores). The Quadro Plex 2200 S4 VCS is
available now for $14,995.
SGI Offers Brand-New VUE
PRODUCT: VISUALIZATION
Silicon Graphics launched VUE, a suite of software visualiza-
tion solutions that changes the way customers create, distrib-
ute, and use visual information.
With VUE software, SGI brings the power of its Reality
Center immersive environments to laptop, desktop, or mobile
devices. VUE software solutions combine visual information
from any source application or platform, fuse it into an intuitive,
three-dimensional viewing experience, and securely deliver that
experience to anyone, independent of the device being used.
VUE features ve components. FusionVUE integrates any
kind of information in any combination from any source, and
presents information via an intuitive, contextual, and three-
dimensional VUEspace environment. SoftVUE transforms any
standard computing system into an integrated compute and
visualization platform, without specialized graphics hardware.
PowerVUE, meanwhile, renders large-scale visualization inter-
actively and in real time. Additionally, RemoteVUEdelivers visual
information to any user, on any device, anywhere in the world.
And, EventVUE delivers high-level visibility into complex situa-
tions in real time, giving decision-makers a 360-degree view of
complex events while avoiding information overload.
SGIs VUE solutions are optimized for the companys Virtu
family of server-based visualization solutions and can run on
any industry-standard x86 computing platform.
Nvidia Deals New Cards
Introducing the worlds largest affordable
SD, HD and 3 Gb/s SDI router!
Eliminate complicated manual video patching
forever! Broadcast Videohub is a powerful
broadcast grade routing switcher featuring a
massive 72 inputs, 144 outputs, 72 deck control
ports, auto switching SD, HD, and 3 Gb/s SDI, in a compact rack
mount chassis only a few inches thick.
Industrial Strength Routing Switcher
With 72 inputs and 144 outputs, Broadcast Videohub has enough SDI
connections for 72 edit systems or decks. With twice as many SDI
outputs, each user can have a completely independent SDI output for
monitoring. This means you can monitor direct from any equipment
in your building without effecting routing! 72 deck control ports are
also included for a complete routing solution.
Simultaneous SD, HD or 3 Gb/s Video
Broadcast Videohub handles mixed SD, HD and
3 Gb/s SDI connections all on the same router at
the same time. Broadcast Videohub detects when
an input changes, and automatically sets all the outputs connected,
to match the changed input. SDI re-clocking and output SDI slew
rates will also change automatically.
Worlds Highest Quality
With exciting new 3 Gb/s SDI connections built in, Broadcast
Videohub allows twice the SDI data rate than normal HD-SDI.
Use 3 Gb/s SDI for high resolution real time 2048 x 1556 feature
lm editing. 3 Gb/s SDI has full compatibility with SD, HD and 2K
in 4:4:4 or 4:2:2 all with a single BNC cable.
Network Router Control
Broadcast Videohub uses software control panels that run on
Windows

and Mac OS X

for routing control right from your


editing system desktop! Broadcast Videohub connects to any
computer via USB and is then shared over your local network.
Router labels can be changed from any computer and are fully
unicode compatible for foreign language support.
Broadcast Videohub
$14,995
Learn more today at www.blackmagic-design.com
































































































































































sw auto ports,
mount chassis








D a , H D S
manual complicated Eliminate
ideohub V oadcast Br ever! for
switc outing r grade oadcast br
outputs, 144 inputs, 72 massive
in SDI, Gb/s 3 and HD, SD, witching
only a few inches thick
g







D s S //s d 3 Gb n a
patching video
powerful a is
a featuring cher
ol contr deck 72
rack compact a
s Highes orld W
n exciting ith W
allow ideohub V
SD Gb/s 3 Use
G 3 editing. lm
in 4:4:4 or 4:2:2
g







! r e t u o I r D
st Quality
built connections SDI Gb/s 3 new
n than rate data SDI the twice ws
2048 time eal r esolution r high for I
S with compatibility full has SDI Gb/s
all with a single BNC cable
g







oadcast Br in, t
HD-SDI. normal
e featur 1556 x
2K and HD SD,








mount chassis
e Industrial Str
inputs 72 ith W
fo connections
u each outputs,
Th monitoring.
buildin your in
also included f








only a few inches thick.
ength Routing Switcher
and 1 ideohub V oadcast Br outputs, 44
tw ith W decks. or systems edit 72 or
independe completely a have can user
om fr ect dir monitor can you means his
deck 72 outing! r fecting ef ff without g
outing solution. for a complete r
Simultaneous SD, HD or 3 Gb/s
mix handles ideohub V oadcast Br








SDI enough has b
SDI many as wice
for output SDI ent
equipment any m
e ar ports ol contr
ideo s V
and HD SD, xed
in 4:4:4 or 4:2:2
Network Route
ide V oadcast Br
indows W

and
system editing
U via computer
c labels Router
unicode compat








all with a single BNC cable.
ol er Contr
pane ol contr e softwar uses eohub
X OS Mac

r ol contr outing r for


co ideohub V oadcast Br desktop!
your over ed shar then is and USB
compute any om fr changed be can
eign language support. tible for for
ideoh oadcast V Br








on run that ls
your om fr right
any to onnects
network. local
fully e ar and er
hub








chang input an
c the match to
rates will also c








th on all connections SDI Gb/s 3
ideoh V oadcast Br time. same the
out the all sets automatically and ges,
and e-clocking r SDI input. changed
. change automatically y.








at outer r same e
when detects ub
connected, tputs
slew SDI output
e mor n Lear








9 9 , 14 $
.blackmagic- www w. at today








5
design.com
CGW :109_p 12/23/08 10:15 AM Page 1
6 January 2009
PRODUCT: MOCAP PRODUCT: PROCESSING
Xsens Releases the
Successor to Moven
Xsens Technologies has unveiled the next generation of
its Moven inertial motion-capture system. The new release
features higher-quality motion tracking, multi-actor record-
ing, an optimized suit design, more intuitive interfaces, and an
improved plug-in to Autodesks MotionBuilder for one-click
setup and improved data transfer.
The offering enables higher-quality tracking through more
accurate gyroscopes and an enhanced accelerometer range
(up to 18G). A new suit design features improved sensor-posi-
tion stability, while an additional motion tracker on the ster-
num renes the granularity of the motion-capture data of the
torso, shoulders, and spine. And now, users can easily transfer
motion trackers from one suit to another; they also can capture
two actors simultaneously, and because the recordings are
synchronized, they can be viewed and edited in one viewport.
This new version of Moven is now available for approxi-
mately $55,000.
AMD Shows Shanghai
AMD recently announced the widespread availability of its
45nm quad-core AMD Opteron processor, which is said to
deliver up to 35 percent more performance with up to a 35
percent decrease in power consumption at idle. As a result of
the release, global OEMs have begun offering systems based
on the new processor, code-named Shanghai, including HP in
its xw9400 workstation.
This processor represents the most dramatic performance
and performance-per-watt increases for AMD products since
the introduction of the rst x86 dual-core processors by AMD
nearly four years ago. The processor touts a 200 percent
increase in Level 3 cache size to 6MB, helping to speed
memory-intensive applications such as virtualization, database,
and Java. With DDR2-800 memory support, it delivers greater
memory bandwidth than current AMD Opteron processors.
The latest quad-core AMD Opteron processors at the 75-
watt ACP ranging from 2.3 to 2.7 GHz are available immediately,
with enhanced quad-core AMD Opteron HE (55-watt) and SE
(105-watt) processors to follow in the rst quarter.
PRODUCT: MOCAP PRODUCT: VR
Vicon Unveils the T10
Entry-Level Mocap Camera
Vicon introduced the T10 camera, which extends the benets
of the companys T-Series at a reduce price point. The newly
designed T10 replaces the MX3+ as Vicons entry-level camera,
offering three times the resolution along with faster perfor-
mance, 3D onboard tracking, backward compatibility, and new
high-powered strobes that increase camera range and detail.
With a new camera design, the T10 camera can be tted with a
range of lensesincluding 8.5 mm, 9 mm, and 12.5 mm sizes.
A one-megapixel camera, the T10 captures 250 frames per
second and enables users to capture ner details in larger
volumes. The camera, like all those in the T-Series, use a newly
designed LED matrix available as either visible red or near-
infrared LEDs. Using super sampling, the offering reduces the
work needed to x misidentied markers resulting from fast
movement, while onboard tracking enables the camera hard-
ware to process 2D tracking from marker trajectories, ensur-
ing better reconstruction data in real time. And with GigE
Ethernet, data streaming is 10 times faster, at 1000Mb/sec,
than with the previous MX3 system.
The T10 works with previous model MX cameras. Pricing
was unavailable.
Organic Motion Launches
LitePod Simulation Platform
Organic Motion debuted LitePod, a next-generation 3D
immersive simulation platform. Optimized for training and
simulation education technology, LitePod allows trainees to
interact directly with virtual environments, without the use
of body suits or tracking devices. As such, LitePod makes it
possible for people to interact, evaluate, and learn how to act
and react in new immersive environments in a natural way.
The stand-alone systems are capable of capturing and digitiz-
ing a subjects complete range of motion, and can be upgrad-
ed to include movements in the hand and ngers, multiple
subjects, the positioning of props or inanimate objects used
by the trainees, and facial movements. Additionally, LitePod
seamlessly integrates with the latest 3D engines to display
live motion data in any virtual environment. With those features
and functions, the platform is geared for any type of realistic
training, particularly for armed services, industry, academia,
and various government agencies. Additionally, the advanced
real-time display allows for multiple units to be simultaneously
connected, for instance, giving an entire regiment the ability to
train together, even from different locations. LitePod is avail-
able now; pricing for a single unit starts at $125,000.
Multibridge Pro is the most sophisticated
editing solution available. With a huge range
of video and audio connections and the
worlds rst 3 Gb/s SDI. Advanced editing
systems for Microsoft Windows

and Apple
Mac OS X

are now affordable.


Connect to any Deck, Camera or Monitor
Multibridge Pro is the only solution that features SDI, HDMI,
component analog, NTSC, PAL and S-Video for capture and
playback in SD, HD or 2K. Also included is 8 channels of XLR
AES/EBU audio, 2 channels of balanced XLR analog audio and
2 channel HiFi monitoring outputs. Connect to HDCAM, Digital
Betacam, Betacam SP, HDV cameras, big-screen TVs and more.
Advanced 3 Gb/s SDI Technology
With exciting new 3 Gb/s SDI connections,
Multibridge Pro allows twice the SDI data
rate of normal HD-SDI, while also connecting
to all your HD-SDI and SD-SDI equipment.
Use 3 Gb/s SDI for 4:4:4 HD or edit your latest feature lm using
real time 2048 x 1556 2K resolution capture and playback.
Worlds Highest Quality
Multibridge Pro includes 3 Gb/s SDI and Dual
Link 4:4:4 SDI for connecting to decks such as
HDCAM SR. Unlike FireWire, Multibridge Pro
has a 10 Gb/s PCI Express connection for powerful HD real time
effects in compressed or uncompressed video le formats.
Microsoft Windows or Apple Mac OS X
Multibridge Pro is fully compatible with Apple Final Cut
Pro

, Adobe Premiere Pro

, Adobe After Effects

, Adobe
Photoshop

, Fusion

and any DirectShow

or QuickTime


based software. Multibridge Pro instantly switches between
feature lm resolution 2K, 1080HD, 720HD, NTSC and PAL
for worldwide compatibility.
Multibridge Pro
$1,595
New Multibridge Pro has SDI, HDMI and Analog
editing with multi channel audio for only $1,595
Learn more today at www.blackmagic-design.com
The Drawn Together images are courtesy of Comedy Partners.
CGW :109_p 12/23/08 10:16 AM Page 1
8 January 2009

CG Character
In computer graphics, the Holy Grail has long been
the creation of a digital human. In this quest, Digital Domains
artists discovered that both god and the devil are in the details as
they delivered an emotional performance by a digital actor for
Paramount Pictures Te Curious Case of Benjamin Button. Te
result is a milestone in computer graphics and in lmmaking.
Based loosely on a short story by F. Scott Fitzgerald and di-
rected by David Fincher, Te Curious Case of Benjamin Button
is a fantastic tale of a man who is born old and ages backwards.
Brad Pitt stars in the title role. Cate Blanchett is Daisy, the love
of his life; Tilda Swinton is Elizabeth Abbott, his rst lover; and
Taraji P. Henson is Queenie, the woman who raises Benjamin
in her old-folks home.
To create the newborn but elderly Benjamin, Digital Do-
main immersed Brad Pitts performance in an aged version of
a digital double. Creating a digital human is di cult enough,
but two things made this particular task especially thorny. First,
Pitt has one of the most recognizable faces in the world. Sec-
ond, Digital Domains reenactment of Pitts performance stars
in the rst 52 minutes of the lm, from the time the character
is born until he is full grown. Te audience must believe in
Digital Domains CG character right from the beginning to
become absorbed in the story, and must remain convinced.
When Benjamin is young, actors of dierent ages and sizes
always perform his body, but his wrinkled face and head, from
his clavicle and shoulders up, is always computer generated.
Benjamins face is digital during his bath, when he crawls into
a tent with young Daisy and she touches his face, when he
struggles to walk during a revival, when he meets his father,
when he gets drunk in a bar, when he works on the tugboat,
and all the coming-of-age moments in between.
Digital Domains work ends on the tugboat. You see him
at the side of the boat reading letters from Daisy, says Ed
Ulbrich, executive VFX producer at Digital Domain. [His
face is] CG then. In the next shot, we see the real Brad Pitt in
makeup. Tats the hando.
I
m
a
g
e
s

2
0
0
8

P
a
r
a
m
o
u
n
t

P
i
c
t
u
r
e
s

C
o
r
p
.

a
n
d

W
a
r
n
e
r

B
r
o
s
.
C
o
u
r
t
e
s
y

D
i
g
i
t
a
l

D
o
m
a
i
n
.
Whats Old is
New AgainI
Whats Old is
New Again
Digital Domain creates a state-of-the-art digital
human for The Curious Case of Benjamin Button
By Barbara Robertson
Whats Old Is
New Again
January 2009 9
CG Character

In computer graphics, the Holy Grail has long been


the creation of a digital human. In this quest, Digital Domains
artists discovered that both god and the devil are in the details as
they delivered an emotional performance by a digital actor for
Paramount Pictures Te Curious Case of Benjamin Button. Te
result is a milestone in computer graphics and in lmmaking.
Based loosely on a short story by F. Scott Fitzgerald and di-
rected by David Fincher, Te Curious Case of Benjamin Button
is a fantastic tale of a man who is born old and ages backwards.
Brad Pitt stars in the title role. Cate Blanchett is Daisy, the love
of his life; Tilda Swinton is Elizabeth Abbott, his rst lover; and
Taraji P. Henson is Queenie, the woman who raises Benjamin
in her old-folks home.
To create the newborn but elderly Benjamin, Digital Do-
main immersed Brad Pitts performance in an aged version of
a digital double. Creating a digital human is di cult enough,
but two things made this particular task especially thorny. First,
Pitt has one of the most recognizable faces in the world. Sec-
ond, Digital Domains reenactment of Pitts performance stars
in the rst 52 minutes of the lm, from the time the character
is born until he is full grown. Te audience must believe in
Digital Domains CG character right from the beginning to
become absorbed in the story, and must remain convinced.
When Benjamin is young, actors of dierent ages and sizes
always perform his body, but his wrinkled face and head, from
his clavicle and shoulders up, is always computer generated.
Benjamins face is digital during his bath, when he crawls into
a tent with young Daisy and she touches his face, when he
struggles to walk during a revival, when he meets his father,
when he gets drunk in a bar, when he works on the tugboat,
and all the coming-of-age moments in between.
Digital Domains work ends on the tugboat. You see him
at the side of the boat reading letters from Daisy, says Ed
Ulbrich, executive VFX producer at Digital Domain. [His
face is] CG then. In the next shot, we see the real Brad Pitt in
makeup. Tats the hando.
I
m
a
g
e
s

2
0
0
8

P
a
r
a
m
o
u
n
t

P
i
c
t
u
r
e
s

C
o
r
p
.

a
n
d

W
a
r
n
e
r

B
r
o
s
.
C
o
u
r
t
e
s
y

D
i
g
i
t
a
l

D
o
m
a
i
n
.
Benjamin Buttons faces at various
ages in the images at left and
below are 3D models created with
computer graphics by artists at
Digital Domain who imbued the
digital faces with actor Brad Pitts
expressions and performance.
January 2009
Whats Old Is
New Again
January 2009 10

CG Character
10
Creating the System
In addition to tools and techniques devel-
oped specically to create Benjamin Button,
which Digital Domain calls the emotion
system, the studio leveraged recently devel-
oped oerings from Mova, Image Metrics,
and the Institute for Creative Technologies,
and relied on such standard commercial
software programs as Autodesks Maya and
Nvidias Mental Ray, as well as its own 3D
compositing software, Nuke (now sold by
the Foundry), and its Academy Award-win-
ning tracking software, Track. But, it took
the crews painstaking artistry to bring the
digital human to life.
[Te project] scared people at rst,
says visual eects supervisor Eric Barba. It
was di cult to sta. People were afraid we
were setting ourselves up to fail.
Te supervisors even had a hard time
hiring from outside the studio. I inter-
viewed people at SIGGRAPH, says char-
acter super visor Steve Preeg. Every person
I hired except for one opted to work on
other shows. And, that person left after
one day.
Eventually, Barba and Preeg structured a
team that was passionate about the project.
Once we started putting out renders, peo-
ple were knocking down our doors, Barba
says. But, by then, we had a smooth-run-
ning show. Tat ship had sailed.
At the peak of the project, 155 artists
worked on the lm, with most spend-
ing a year and a half creating and rening
shots, some of which took more than 100
iterations to nail down. It was incredibly
daunting, Barba says. We had to break
it into bite-size chunks so the artists heads
wouldnt explode. Its a testament to the
team, that they continually knocked down
challenges and were undeterred from the
task at hand.
Preeg jumps in with a laugh, saying, I
was deterred.
Expressions
Fincher commissioned the rst proof-of-
concept tests in 2004, but research had
begun at Digital Domain in 2002. Once
preproduction on this lm began, the stu-
dio started the process of replicating Pitts
expressions by painting his face green using
the special makeup devised by Mova for its
Contour system. For two days, one before
principal photography and one after, Pitt
bit his lips, pued his cheeks, looked up,
down, right, left, raised an eyebrow, tight-
ened his neck muscles, showed surprise,
fear, anger, and so forth, while Mova cap-
tured these expressions at 24 frames per
second using 28 camera.
We walked Brad [Pitt] through a FACS
session, where we recorded his face doing as
many of the individual motions as he could,
says Preeg. FACS, developed by Paul Eck-
man and Wallace Friesen (and later rened
by Joseph Hager), categorizes the anatomical
form of emotions that play across a persons
face; that is, which muscles move to create
particular expressions. Mova translated Pitts
expressions into useful data that the com-
pany sent to Digital Domain in the form of
digital meshes. Tere, the artists applied data
from the meshes to the animation rigs.
First, though, they stabilized the meshes.
Te crew wanted to encode the individual
shapes within a keyframable rig so that ani-
mators could tweak the expressions. But,
when Pitt raised an eyebrow, he might also
have tilted or turned his head, so they had
to remove the head tilts and turns before
they could apply the eyebrow motion to
the rig. To do that, they needed to align the
captured motion in all the Contour mesh
frames with a neutral pose and lock the mo-
tion to that pose. We had to know from
a neutral pose what raising his left brow
changed on his face, Preeg says.
Figuring out how to lock the individual
frames to a reference pose fell to Digital
Domains award-winning creative director
of software, Doug Roble. Te head motion
had to be removed from the motion-cap-
tured data, but all we had to work with was
the data captured on his face, says Roble.
Roble found technical papers that de-
scribed ways of aligning models, but they
didnt assume that the models changed
when the subject lifted an eyebrow. He had
to align models that changed over time.
Te face was always moving and deform-
(Left to right) Digital
Domain videotaped
Brad Pitt perform-
ing a scene on a
soundstage. Later, the
studio applied Pitts
performance to a CG
model of Benjamin.
(Top) A body double
wearing tracking
dots plays Benjamin
Button during principal
photography for that
scene. (Bottom) The
nal shot in the movie.
January 2009 11
CG Character

ing, Roble says. How do you tell what is


facial motion and what is the head moving?
Tere were no hard points, not even the
ears. It was a full-blown research project.
Roble solved the problem by devis-
ing software that searched on a per-frame
basis for the parts of the face that didnt
movethe forehead, perhaps, or the jaw.
You can watch [the program] highlight
the areas that are most static, Roble points
out. You can give the program hints, but,
for the most part, we didnt have to. It just
worked, which was a big surprise to me. It
locked the face down so that all the motion
was based only on the facial performance.
Once the artists had a stable face, they
began extrapolating Pitts individual mo-
tions from the data. It wasnt an automatic
process. Tey often had to make such deci-
sions as, for example, whether Pitt always
needed to move his mouth to raise his up-
per left eyebrow. And, they needed to add
data where the Contour system stopped
the outer rim of his eye, the edges of his
January 2009 January 2009
This was a nonlinear show, says Steve Preeg,
character supervisor. One of the rst things that
happened was a collaboration between anima-
tion, compositing, and lighting. Our animation
reviews were done using [Nvidia] Mental Ray
renders and the HDRs. To light the head, the art-
ists had to match lighting conditions in such dis-
parate environments as the interior of an old-folks
home, a front lawn in broad daylight, and the deck
of a tugboat on a stormy sea.
The director and DP did a meticulous job in
these locations, and our job was to be true to that, says
lighting supervisor Jonathan Litt. His team worked with the
360-degree HDR photographs taken on the sets, video
footage of the small-body actors, and the survey data.
We completely reconstructed the scene, Litt says, in-
cluding all the lighting apparatus, the bounce cards, the
scrims, and other diffusers. Then, rather than using illumina-
tion in the spherical HDR images to light the CG face direct-
ly, they placed the HDR data on the reconstructed lighting
elements. Every shot had its own HDR setup.
Even though reconstructing the set is one step more than
we usually take, we went for it because the head was often
moving and, in some cases, the lighting was subtle, Litt says.
On one hand, we exercised artistic control. On the other, we
wanted to start with the realistic falloff in the local environment.
The compositing team prepared the HDR images for light-
ing by projecting them in Digital Domains Nuke onto the
3D set geometry and then re-photographing the environ-
ment from the position of the head in the scene. Once we
were happy with the overall feeling of the lighting in previs,
we exported the 3D Nuke scene into a form [Autodesks]
Maya could read, and put each light in the HDR onto a card,
explains compositing supervisor Paul Lambert. The lighters
could tweak the cards and put objects in front of them.
The lighting artists would see a primitive version of the
real set around Benjamin, with his digital head and rough
hair rendered in Maya. They had everything of signicance
affecting lighting on the head, including inert objects that
blocked or broke up the light, Litt says. The kind of things
that would be held by C-stands are oating around the head
in the CG version.
The most difcult sequence for the lighters was the church
tent revival sequence. On set, the DP had used hundreds of
lightbulbs hanging from cords and two beauty light sources.
Director David Fincher had lmed the sequence on a windy
day. The whole tent was shaking, and those cords were
swinging, Litt says. We re-created the entire thing. The
cords. Every single light source. There was something about
the subtlety of the lights when some cords were swinging and
others werent, and the changes in the camera angles. With-
out re-creating it, the shots werent up to snuff.
Often, the artists compared the result to lighting on the
body actor in the plate, and having footage from all four cam-
eras on set helped them understand how light acted within
this shot and others. [The CG Benjamins] face is a different
shape, and his skin is different from the actors skin, but we
became so familiar with the tonality of each actors skin, it
was still useful, Litt says. Barbara Robertson
Light the Match
January 2009 12

CG Character
lips, his ears. At the end, they had a collec-
tion of thousands of little motions.
Tis is where we move into emotion
capture, not motion capture, Ulbrich says.
We built a library of micro-expressions con-
structed from bits and pieces of Brad so we
could assemble a performance of what Brad
really does. We arent interpreting his perfor-
mance. We have [encoded] the subtleties of
how Brads brain drives his [facial] muscles.
To use those bits and pieces, the team ap-
plied the library to a facial animation rig that
gave the animators control using blendshapes
and other techniques. Some things, like
corneal bulge and sticking lips, cant be done
with blendshapes, so we used a combination
of stu, says Barba. Stu? He hedges: We
have to tiptoe around the secret sauce.
After all this, though, the crew had a facial
animation rig with little bits of Pittalbeit
at his current age, not, as they needed, for the
60-, 70-, and 80-year-old faces. For each of
those ages, modelers had created heads and
faces from scans of wrinkle-faced maquettes.
Te maquettes, built by Rich Baker, were
the foundation for our work, says Barba.
One problem. Baker had based the elderly
To create Benjamin Buttons skin, the lighting and shading
crew combined characteristics from the three maquettes of
Button at age 60, 70, and 80, and photographs of Brad Pitt.
The maquettes, created by Rick Baker, provided details,
such as age spots, that the artists used for texture maps.
The silicon models also provided a starting point for shaders
that would gave Benjamins surface its realistic translucen-
cy through the scattering of light in the subsurface, sheen
through specular lighting on the surface, and shape through
such displacement detail as pores and wrinkles.
Theres interplay between the characteristics of the sur-
face, says Jonathan Litt, lighting supervisor. The displace-
ment affects the subsurface and specular lighting.
During the look-development phase, the lighting team
captured the surface characteristics of the maquettes using
the Light Stage system at the Institute for Creative Technolo-
gies Graphics Lab. This system surrounds a subjectin this
case, the maquettewith multiple cameras that re simulta-
neously. Usually this happens in changing light, but this time,
that wasnt important. We didnt want to invest too much
into matching the maquettes exactly, Litt says. We wanted
to use them as a visual reference.
Working from a set of photographs taken on the Light
Stage with all 164 lights turned on, the crew at Digital Do-
main processed the images based on lighting data from the
HDR photographs taken on location. Then, compositors
projected them onto a basic model in Nuke. The Light Stage
images provided the falloff and the intensity change caused
by lights at various angles. The HDRs provided the light from
the environments the crew needed to match.
Paul [Lambert] applied the lighting from the HDR envi-
ronment map in a way that correlated to each of the lighting
angles from Light Stage, and basically mixed hundreds of
those images until the CG head looked like it was lit by the
environment map, Litt explains. The compositors would run
this out in Nuke while calibrating the lighting setup for each
sequence. That way, they could quickly see a lit preview head
based on the Light Stage photos to know if they were in the
right ballpark to calibrate the color for each sequence.
Because the maquette was plastic, though, the CG mod-
el looked plastic, too. [Light Stage] was a neat tool for de-
velopment, Litt says. We wrote shaders to emulate human
skin based on combining photographs of Pitt and the Light
Stage data from the maquettes, but the Light Stage data
wasnt used directly in any nal frames.
Instead, the CG head used procedural shaders driven by
texture maps for the skin and head. We could light the pro-
cedural skin by any light, as if it were the real thing, Litt says.
You could plop the head down anywhere.
Although Digital Domains crew wrote specic shaders
for this show, Litt is quick to deect any special recogni-
tion. I want to give full credit to the amazing work thats
happened with skin shading in the last ve years, he says.
It was a lot of work nailing down the skin shading, though.
We pulled from industry-standard approaches to implement
our own avor in [Nvidias] Mental Ray and tune how we
used our texture maps to drive the shaders. But, this is not
a young character with perfectly smooth skin. We had many
an iteration tuned to what was happening in wrinkles and
creases. Barbara Robertson
Skin Deep
January 2009 13
CG Character

heads on Pitts head, but the small actors


playing the young Benjamin Button had
smaller heads, and Digital Domains heads
would need to t on the small actors shoul-
ders. Te riggers wanted to give the anima-
tors a specic facial animation rig for each
age. First, though, they did create a scalable
rig for the heads to show Fincher various
sizes for the dierent ages.
Once he settled on the scales he wanted,
Preeg explains, we retargeted the motion
data, making reasonable guesses for how
Benjamins jaw skin would move when he
is 80 and so on. Ten, we created a rig for
each age.
It took nine months to put the captured
shapes into deformation rigs that anima-
tors could manipulate for each age. Get-
ting those shapes into the rig was the magic
voodoo, Barba says.
And that was only the beginning. Creat-
ing a digital human is not easy.
Timing is Everything
We discovered that Benjamin not only
had to move exactly like Brad does, but
without Brads timing, he wouldnt come
across as Brad, Barba says. Tere was no
turnkey solution to capture both. Tat is,
to simultaneously capture his expressions
and his timing.
Te plan was to have Fincher direct
Pitt reading Benjamin Buttons lines on
a soundstage after Fincher had nished
shooting the small-body actors and the
rest of the cast on location. Te timing
Digital Domain needed would come from
that reading. In order to use Mova for
that, we would have had to put Brad in
green makeup for seven days, Barba says.
Tat wasnt an option. With Brad in green
makeup, it was hard for David [Fincher] to
discern his performance. Also, the Mova
system wouldnt mix well with the audio
recording. Tere are many hard drives,
Barba says, and fans are whirring.
Instead, as Pitt read the lines on the
soundstage, Digital Domain captured his
performance on video. Later, Image Met-
rics derived animation curves from that
video. It worked beautifully, Barba says.
Brad was relaxed and gave us a great per-
formance. We had him scheduled for 10
days, and he nished in seven. He was
happy. David was happy.
Heres how it worked: Pitt reclined in
an ab (exercise) machine on a soundstage,
with his knees raised up to align his body.
Around him, the crew positioned four
Tompson Viper digital cameras that were
constantly in sync and trained on his face.
His face was always in the center of the
frame, Barba says. And cameras set at
angles always gave us his prole.
Although Image Metrics typically uses
just one camera, Barba insisted on four.
With one camera, you cant see the jaw
from the prole or subtleties around the eyes
and ears, Barba says. Also, Barba wanted to
give the animators all four views later for
reference. On a nearby monitor, Pitt could
see the edited footage from shots lmed on
location. He wore an earbud to listen to the
recorded scenes. Hed look down, watch
the footage, then say, Great. David would
call, Action, and hed do the line, Barba
says. Ten wed show him the next clip.
Preeg was impressed: [Pitt] could watch
a 40-shot sequence twice and then perform
his lines strictly on the audio cues. I was con-
cerned about his matching the performance
months after the shoot to a persons motion
that was done 3000 miles away on set. But,
Ive got to say that Brads a lot more than a
narrow waist and broad shoulders.
After the performance capture and dia-
log session, Fincher inserted his selections of
Brads performance into the edited footage,
putting the selects on the side of that footage.
To turn a performance into motion data,
Image Metrics custom software usually
translates key points from a video onto a
Instead, as Pitt read the lines on the
soundstage, Digital Domain captured his
Instead, as Pitt read the lines on the
soundstage, Digital Domain captured his
Instead, as Pitt read the lines on the
soundstage, Digital Domain captured his
Instead, as Pitt read the lines on the
soundstage, Digital Domain captured his
January 2009
Hair He Is
When Benjamin Button is
rst born and very old, he
has little hair, but even that
sparse hair needed to be
modeled, moved dynami-
cally, and rendered. Mat-
tais Bergborn led the team
at Digital Domain that created a new hair system based on Mayas NURBS
curves. With the maquettes for reference, the team paid particular attention
to creating various ways of using the guide hairs to clump the hair. Even his
hair at the scraggliest has clumping behaviors and friction and electromagnetic
effects, Bergborn says. It might look good modeled, but it gets fuzzy and indis-
tinct when rendered. So we had to give it structure and nd ways to make the
light rays take more interesting paths to create shadows and highlights.
The hair stylists could set global parameters, control areas with texture maps,
and create clumps with guide hairs that attracted other hair; a virtual brush
gave the stylists control over CVs within a radius as a stylist stroked it along
the guide hair curves. Bergborn worked on Benjamin Buttons hair for a year.
David [Fincher] was meticulous, Bergborn says. Hed want minute changes.
We had to create the hair on a shot-by-shot basis. Barbara Robertson
January 2009 14

CG Character
standard rig. Tat wasnt appropriate for this
performance capture, though. Clearly, their
rig wouldnt do, Barba says. So, we built
our rig in a way that we could give it to them
and they could apply the curves to it.
And, that was the last ingredient in the
performance-capture soup. Image Metrics
translated the movement of Pitts face into
animation curves and applied them to a rig
that Digital Domain built using shapes de-
rived from data that Mova had captured dur-
ing the FACS session Pitt had done earlier.
When the animators got the footage,
they could see, attached to the small actors
bodies, the old faces and heads performing
Pitts dialog. It wasnt a turnkey solution,
Barba says of the performance capture. It
was a starting point for timing. Tere are
shots in the video where Brad Pitt is sitting
and staring, and you know hes thinking.
Ten you load up the rig. Hes not doing
anything, and he looks like a dead guy.
Tats where our animators came in.
On Location
To t the old heads onto the bodies of
the small actors lmed during principal
photography, the tracking team needed to
locate the camera and the actors head pre-
cisely in 3D space within each shot. Tey
needed to do this for 325 head replacement
shots. Each shot averaged 100 frames.
During lming, data integration lead Jesse
James Chisholm conducted an on-set sur-
vey, measuring the main points, taking still
images with a Canon 1DS, and collecting
high-dynamic range (HDR) images360-
degree photographs of the set taken at mul-
tiple exposures for each lighting setupfor
lighting. Te body actors wore custom-t-
ted blue hoodies onto which tracking su-
pervisor Marco Maldonado attached yellow
markers in specic patterns.
In addition to Finchers A- and B-roll
cameras, Maldonado added two Viper cam-
eras as witness cameras on either side of the
main camera. All the cameras were synced
with time code and were gen-locked so the
shutters were in sync as well. Te sync was
a new process, but we felt doing that would
save us time later, Maldonado says.
Te time code helped the team more easi-
ly nd matching data streams from the cam-
eras so they could do a visual check to see
if the data was in sync. We relied on that
before we went into the calibration process,
Maldonado says, referring to the method for
deriving the motion for the cameras.
Once the cameras were calibrated, we
oriented the dataset to the on-set survey
point cloud, Maldonado continues. Tat
gave us all four cameras calibrated to the
set in orientation, scale, and placement.
We could see where they were in relation
to, say, the Nolan house and front lawn;
we could see in Maya that they were in the
right place. Ten we did pattern tracking.
By tracking patterns in the 2D foot-
age, the software helped the crew locate
in 3D space the small actor wearing the
blue hoodie with the yellow dots. And,
that made it possible to replace the persons
hoodie head with the CG head.
We had 20 or 30 patterns on the actors
hoodie and face, Maldonado says. If you
track the same pattern of dots with four
Eyes Right
One artist on the crew worked full time on the eyes to get the peeled grape
quality director David Fincher asked for on the surface. There were some close-
up shots in the plate that denitely scared us, says lighting supervisor Jonathan
Litt. So we asked the modelers for an extra touch. Our eye artist, who took high-
denition footage of his own eyes, noticed the conjunctive layer, a netting that
connects the skin in the corner to the eyeball. So we rigged this layer into the
eye. When the eye turns to look in an extreme direction, this layer stretches and
extends with the eye.
For eye reections, the lighting artists used a hybrid of rendering and compos-
iting. We tried to give the compositors as much exibility as possible, Litt says.
Some eyes have straight 3D-rendered reections. Often, we needed serious
adjustment in compositing for the eyes and the skin. The compositors didnt ad-
just the lighting, but they balanced the skin tone and oily layer, and other optical
effects. Barbara Robertson
Lighting artists duplicated the lights on set to t the digital head seamlessly into the background.
hoodie and face, Maldonado says. If you
track the same pattern of dots with four
hoodie and face, Maldonado says. If you
track the same pattern of dots with four
pervisor Marco Maldonado attached yellow pervisor Marco Maldonado attached yellow
markers in specic patterns.
pervisor Marco Maldonado attached yellow
markers in specic patterns.
photography, the tracking team needed to
locate the camera and the actors head pre-
photography, the tracking team needed to
locate the camera and the actors head pre-
photography, the tracking team needed to
locate the camera and the actors head pre-
hoodie and face, Maldonado says. If you
track the same pattern of dots with four
hoodie and face, Maldonado says. If you
track the same pattern of dots with four
pervisor Marco Maldonado attached yellow
markers in specic patterns.
photography, the tracking team needed to
January 2009 15
CG Character

cameras, the system triangulates that pat-


tern into 3D space. We could see the 3D lo-
cator moving around in the environment.
At the end of that process, a tracking art-
ist had a point cloud moving in 3D space
in the virtual set. Tat was the foundation
from which everything started building,
Maldonado says. We attached the CG
head to the point cloud.
Benjamin Buttons hat complicated the
problem of setting the CG head on the
actors shoulders, especially because the CG
head wasnt exactly the same size as the actors
head. When you have a collar, you have a
little wiggle room, but the hat has a specic
line, says Roble. If theres any motion be-
tween the digital head and the hat, you see it.
You see the hat moving on top of the head,
or the head moving under the hat.
To help the 21-person tracking team,
Roble tweaked the studios custom soft-
ware, Track. I made it faster, he says,
denitely faster. And, more accurate.
Once they had tracked the head onto the
clavicle and sometimes under the hat, the
tracking artists could look at it from dierent
camera views to assure that the CG object
was in the correct place. Tey also did track-
ing checks at 2 resolution on large screens.
We used the alpha channels from the
roto team to do a quick composite through
Maya or Nuke to help verify that the track-
ing was good, says Maldonado. When
they were satised, the head moved on to
animation and lighting.
Tiny Tweaks
[Fincher] put a lot of eort into making
performance selects that timed correctly
with the body, Preeg says, but in many
cases, we had to change Brads perfor-
mance. In general, the animators tried to
match the 3D heads to the motion of the
small actors heads. Often Pitt had matched
that motion during his performance cap-
ture, but sometimes his head moved too
muchfor example, when the small actor
was walking. Or, his head tilt wouldnt cor-
rectly match the eye lines. Other times, the animators needed to change a small actors
A Blur and a
Final Touch
At the end of the pipeline, Benjamin Button arrived at compositing in 60 layers,
which the compositors could tweak in Digital Domains Nuke. They had indi-
vidual layers for lips, teeth, eye matter, hair, six different skin passes, layers for
specular and diffuse lighting, and much more.
Because the layers arrived without motion blur, the compositors added it us-
ing Vector Blur Z, a motion-blur technique developed by creative director of
software Doug Roble and implemented in Nuke. The technique helped reduce
rendering time.
Motion blur is interesting because parts of a scene are blurred and parts
arent, Roble says. It depends on where the geometry is in the scene. Its great
if you can calculate it in rendering, but its expensive because the renderer has
to essentially render a whole bunch of sub-frames.
To add motion blur in Nuke, Roble had Nvidias Mental Ray output depth
and velocity, as well as color, for each pixel. The renderer doesnt compute the
streak, but it tells us how its moving, if its moving, Roble says. So with that
information, we can streak the pixel in Nuke just as the renderer would, but more
quickly. And, the Z-depth tells us when part of Benjamins face goes behind
another part so we know not to blur there.
Vector Blur Z has made a huge difference at Digital Domain, cutting the time
to compute motion blur on a previous lm from 16 hours to four on some shots.
Theres no comparison data for Benjamin Button because the crew always
used Nuke on the project.
If we had had to motion-blur Benjamin with the renderer, the times would have
gone through the roof, says compositing supervisor Paul Lambert. Tweaking
the amount of motion blur added a slight bit of work in compositing, but it was
part of our template, so we just took it in stride.
Of all the shots in Benjamin Button, one of the most difcult for the composi-
tors was of Benjamin getting drunk in a bar. For that, in addition to the usual ad-
justments to t the somewhat drunken digital head into the scene, they painted
Benjamins face ushed using tools in Nuke that would stick the paint to the
face. The shot took 150 iterations before they nished. Usually, though, the work
owed remarkably smoothly.
The lighting renders were very close to the plate, so we rarely had to do over-
all color corrections, Lambert says. Daylight took more work than dark, moody
lighting because we had to increase the amount of subsurface scattering at cer-
tain angles. And, we did have to change things like skin blemishes. But, we knew
what to tweak based on previous shots and the environment Benjamin was in.
Lambert worked for 18 months on the show and yet, like other members of
the crew, claims he never found the work tedious. Because of the pipeline, it
was a really good experience, he says. It was awesome to see the character
come to life. Barbara Robertson
was a really good experience, he says. It was awesome to see the character
muchfor example, when the small actor
was walking. Or, his head tilt wouldnt cor-
January 2009
muchfor example, when the small actor
was walking. Or, his head tilt wouldnt cor-
muchfor example, when the small actor
was walking. Or, his head tilt wouldnt cor-
muchfor example, when the small actor
was walking. Or, his head tilt wouldnt cor-
come to life. Barbara Robertson
January 2009 16

CG Character
rectly match the eye lines. Other times, the
animators needed to change a small actors
head motion.
For example, Benjamin learns how to
walk during a sequence that takes place
in a revival tent. Hes seven years old, and
thats how Pitt plays him, but he looks 80.
Hes got his head down, and he looks up
shyly, Preeg says, mostly with his eyes at
the top of his eyelids, which is childlike be-
havior. You want to get that shy behavior,
but when you put that motion onto an 80-
year-old face, the overhang of his eyelids
means he cant see anything.
So, the animators started inching up
Benjamins 3D head and slightly moving
his eyelids. Wed gently nudge the lid one-
half millimeter, rotate the head a half de-
gree, and lower the eye a half degree, Preeg
says. It took iteration after iteration.
Another example: Tere were shots
where Benjamin is smiling and David
wanted a bigger smile, Preeg says. But
when he smiles, his eyes narrow, and his
eyes were already narrow. So wed try to
open them, and hed look like a crazy ax
murderer.
As they worked, the animators could
quickly see any of the four views of Pitt shot
during the performance-capture session on
that frame or in a separate window, and could
step through the mesh in Maya. Tey could
lock to the head or the neck, and see the facial
motion without any body motion. And, they
could query the system to see what shapes
were active in the rig at one time.
All told, only eight animators worked on
the performance. We werent sculpting the
face, Preeg says. We
were turning on bits of
Brad, which allowed
us to quickly tweak the
emotion with a small
number of animators.
David didnt want the
animators version of
Brad. He wanted Brad.
We used a minimum
number of shapes and let the data work for
us. It was a lesson in restraint. Te uncanny
valley is very steep.
Barba quickly adds, Steve [Preeg] and
I bought property there. We spent a lot of
time there.
Even though most of the artists who
worked on the lm lived with Benjamin
Button for more than a year, they didnt
nd the work tedious. For this crew, god
was in the details, not the devil after all.
For example, lighting supervisor Jona-
than Litt, an 18-month Benjamin Button
veteran, explains: On the one hand, it was
nice because its a project limited in scope.
On the other hand, the scope of that one
thing, a digital human head, is so expan-
sive, theres almost an innite amount of
detail. Skin. Eyes. Hair. Teeth. Tongue.
Gums. Stubble. Eyeglasses. Getting it to sit
on the neck. Animation. Tracking. We had
to go down to the smallest detail.
But it was time well spent. Two weeks
before the release date, Te Curious Case
of Benjamin Button received ve Golden
Globe nominations, including Best Mo-
tion Picture, Best Director, and Best Actor.
Its likely to score more, including, perhaps,
among potential Oscar nominations, one
for visual eects. In any case, if Brad Pitt
wins an award for best actor from anyone,
you can bet that several hundred people at
Digital Domain will be sharing those ear-
to-ear grins, too.
Barbara Robertson is an award-winning writer and a
contributing editor for Computer Graphics World. She
can be reached at BarbaraRR@comcast.net.
All told, only eight animators worked on All told, only eight animators worked on All told, only eight animators worked on All told, only eight animators worked on
the performance. We werent sculpting the
Benjamins digital head had to t between the body doubles
moving hat and shoulders, making tracking especially difcult.
The Quest
The rst lm to seriously open peoples eyes to the possibility of digital actors
starring in lms was Final Fantasy in 2001, but the milestone 3D digital stars list-
ed below have pushed computer graphics pilgrims further down the road toward
their Holy Grail: A digital human starring in a feature lm without restrictions.
2001: The cast of Final Fantasy (Square)
2002: Gollum (Andy Serkis) in The Lord of the Rings: The Two Towers
(Weta Digital)
2003: Gollum (Andy Serkis) in The Lord of the Rings: The Return of
the King (Weta Digital)
2004: Doc Ock (Alfred Molino) in Spider-Man 2 (Imageworks)
2004: The baby in Lemony Snickets A Series of Unfortunate Events
(Industrial Light & Magic)
2006: Davy Jones (Bill Nighy) in Pirates of the Caribbean: Dead
Mans Chest (Industrial Light & Magic)
2007: Davy Jones (Bill Nighy) and Jack Sparrow (Johnny Depp) in Pirates of
the Caribbean: At Worlds End (Industrial Light & Magic)
2008: The aged but young Benjamin (Brad Pitt) in The Curious Case of
Benjamin Button (Digital Domain)
CGW :808_p 7/16/08 11:45 AM Page 1
January 2009 18

CGI
Tactile World
2008 Universal Pictures.
Framestore Animations CG
fairy tale breaks new ground
By Barbara Robertson
niversal Pictures rst animated feature, Te Tale of Despereaux, is a fairy
tale, not a toon. A real fairy tale, with action and emotion, sweetness and
horrornot a fractured, gag-lled, pop-culture satire of a fairy tale. Its
a drama about forgiveness and honor, and the consequences of mistakes
and mistaken intentions. And, every inch of the CG lm looks the part.
I feel lucky we were able to tell an animated story with cinematic aspirations, says
writer and producer Gary Ross. We were able to aspire to things visually that usually
arent seen in the genre because of the dramatic intentions and themes in Kates book.
Kate is Kate DiCamillo, the author of the books upon which Te Tale of Despereaux
is based: the stories of a rat and a beautiful princess who discover the power of forgiveness,
and a mouse and a maid who long for things that those with more experience know they
couldnt possibly achieve. Tus, they do.
As did the crew.
Ross knitted the separate stories into one screenplay that opens with the rat Roscuro
(Dustin Homan) making a serious mistake that causes the happy Kingdom of Dor to
become monotonously gray and gloomy. Soon after, we meet Pea, a beautiful princess
(Emma Watson) who longs for something else, even rain; Miggery Sow, a home-
ly maid (Tracey Ullman) who wants to be a princess; and Despereaux (Matthew
Broderick), a heroic little mouse with enormous ears who lives with his timid
brother and his fearful parents. Framestore debuted its new animation division
by bringing all these CG characters to life.
Evgeni Tomov, who had been production designer for the 2D animated
feature Te Triplets of Belleville and has received an Annie nomination for
Despereaux, worked initially with Triplets director Sylvain Chomet at Studio
Django on concept art for the fairy tales environments and on character
design. When Chomet left the project, the production moved to London.
Tomov moved, too.
Despereaux was the rst 3D lm Tomov had worked on, and he welcomed
the challenge. Tere are limitations in 2D animation, he says. If you push
the environment to a believable, immersive level, the characters look cutout
because theyre at and not aected by the lighting. But in 3D, especially
with contemporary tools, you can create the immersive atmosphere
that live-action lms have. I hadnt seen it in the animation world,
but that was something I tried to implement, to make the audi-
ence feel like theyre part of the action and that they can walk
between the characters.
Tomovs instinct blended well with Rosss expertise as the
Oscar-nominated writer, director, and producer of Seabiscuit
and other live-action lms. In many ways, Ross approached the
making of his rst animated feature as if it were a live-action
lm, and the result rippled through the production.
Shot List for an Animated Feature
When Despereaux moved to London, director Mike Johnson (Corpse Bride)
replaced Chomet for a short time, and contributed additional concept art and
early character development. In May 2006, animator Robert Steven hagen
came onboard as head of story. Ten, after Johnson returned to Los Angeles, di-
rector Sam Fell (Flushed Away) joined the project, Stevenhagen moved into position
January 2009 19
CGI

Despereaux, the
bold little mouse
with big ears (at
right), stands out
from the rest of
the pack in mouse
world (at left).
January 2009 20
nnnn
CGI
as co-director, and the new team pushed
the reset button. It was early 2007.
We stopped everything, Fell says. We
had a lot of visual stu done by various
people and some sequences. It didnt hang
together. We went back to Garys screen-
play and worked with him as if we were
starting afresh. We basically imagined the
lm together. Also joining Ross, Fell, and
Stevenhagen was Framestores cinematog-
rapher and head of layout Brad Black-
bourn, who had served in those roles at
DreamWorks for Flushed Away and as head
of previs for that studios Kung Fu Panda.
Working from Rosss screenplay, the group
met in London, Boston, and Los Angeles
as they decided on the camera coverage and
essentially edited the lm theoretically.
I shot-listed all of Seabiscuit, and I
shared that process with the guys, Ross
says. We shot-listed this entire movie from
scratch with lens lengths, camera angles,
shot movement, and cutting patterns.
Teres always a moment in any mov-
ie, Ross adds, when you have to commit
to a decision. Te question is, when? If
you can commit to some creative decisions
early, I think it creates certain clarity. Te
cinematic, artistic intentions become more
specic, and the production becomes
more ecient. Tat eciency resulted in
both time and cost savings.
For Blackbourn, it was an exciting, new
process. As they talked, all four of the
menRoss, Fell, Stevenhagen, and Black-
bournplus Len Morganti, an artist who
had worked with Ross in the past, would
sketch thumbnails.
We drove the sta of restaurants crazy
ordering Japanese barbecue, tapaswhat-
ever as we worked through the script,
Blackbourn says. Len would quickly sketch
out shots that Gary would describe with
words and with his hands. What if we put
the camera lower and rake the angle so we
can push the perspective? What if the leg of
the chair comes across here? What if were
shooting a small creature in the big world
with a 12mm lens?
At the same time, Stevenhagen drew
thumbnail sketches of the character perfor-
mances. Te combination gave the team
the acting performances and the eect the
camera had on the characters and the set
around them. And that turned the typical
animation process upside down.
Usually in major, traditional studios, the
lms are storyboard-driven, Blackbourn
says. But by the time we went to storyboard-
ing, we already had the camera language for
the lm, and thats never happened before
in animated lms. We had a bible that de-
scribed every single frame in words like long
lens tracking medium shot of Despereaux as
he moves through the crowd, blurry backlit
characters buzzing in foreground extremely
out of focus moving through pockets of
overexposed cool light from above, other
parts of the frame fall o to black.
Every shot in the 120-page script had
notes on depth of eld, camera apertures,
and basic lighting elements. Gary [Ross]
thought in a lm sense, not an animated
sense, Blackbourn says. Ive been trying to
push 3D animated lms to embrace more of
the language of live-action lmmaking. Tis
was the process I had been aiming for.
Tat live-action sensibility changed the
look of the lm as well as the process by
which the crew made the lm. Ross surprised
the layout team with lens choices and com-
positions rarely used in animation. It was so
strange for me to see a character, right in the
middle of the frame, in animated lms, he
says. In live action, thats anathema.
Tus, in Despereaux, as in a live-action
lm, you will see a crowd thats not perfect-
ly in front of the camera, bits of characters
in the corner of a scene, camera shake, and
out-of-focus props in front of the camera.
Youll also see Dutch angles, which tilt
the horizon, and other techniques bor-
rowed from live-action cinematography.
To help accomplish this, Ross was a
stickler for precise depth of eld. I fought
early on for algorithms that accurately re-
ected depth of eld at specic f-stops and
lens lengths so the plane would fall o in
Flemish cities inuenced the Kingdom of Dors environments, while writer/producer Gary Ross
inuenced camera moves that mimicked live-action lms and incorporated precise depth of eld.
January 2009 21
CGI
nnnn
a natural curve, he explains. I wanted to
move through the 3D space as if it were a
painting brought to life.
When Ross moved the camera through
so-called mouse world, he wanted to do so
using a mouse-size camera. Gary asked
what the depth of eld would be in mouse
world if we were shooting with a mouse-
size camera focused two centimeters from
the lens, Blackbourn says. So, we built a
mouse-size crane and dolly rigs. Ten Gary
would say, Te shot is beautiful, but its
too perfect. I can feel the computer in this.
What can we do? So we put grains of sand
on the dolly track. Te sand was, in fact,
bumps on the camera curves.
Keys to the Kingdom
Mouse world was one of many environ-
ments in Despereauxs Kingdom of Dor de-
signed by Tomov, all within an aesthetic he
chose early in the lms development. Ev-
geni has been the consistent visual voice
throughout the whole process, says Fell.
For the environments, Tomov looked to
such 15
th
and 16
th
century Flemish towns
as Bruges, Belgium, and Amsterdam for in-
spiration. Te goal was to create a parallel,
imagined world through textures and light-
ing, he says. Tomov also referenced the
Flemish painter Vermeers use of light and
shadows to direct the viewers attention, es-
pecially in the human and mouse worlds.
Te human world is the least surprising,
Tomov says. We tried to make it stylish and
entertaining. Te most amazing room is the
huge kitchen in the castle where they cre-
ate the annual soup. For this, he designed
a huge and complex, Rube Goldberg-like,
wooden soup-making machine.
Tomov gave the mouse world, which
is close to the kitchen, a dierent mood.
Its a restrictive world where the mice are
taught to be afraid. Tey celebrate the rules
they live by: austerity, discipline, uniformi-
ty. So, its disturbingly neat and fastidious,
and the colors are muted.
A third world hides deep below the cas-
tles dungeon. Tis is the dark rat world,
where the mouse council throws disobedi-
ent little mice. For this more chaotic world,
Tomov moved from Vermeer to Hierony-
mus Bosch for inspiration. Tis world has
more extreme lighting, he describes. Its
more grotesque, a marriage aesthetically be-
tween the declining Roman empire and an
anarchistic medieval look. Te rat world
has an evil emperor and a coliseum with a
nasty house cat chained inside a cage.
To realize Tomovs designs, Framestore
put 25 primary characters, 12 secondary
characters, and crowds, most of them furry,
in 60 complex environments.
Te environments range from the
storerooms where the mice live, to a big
banquet hall with beautiful pillars and
staircases, to the rat world, to a farmyard,
says Ben Lambert, who supervised the 10
modelers at Framestore who created the
characters and an inventory of props that
mushroomed into the thousands. We
knew from the storyboard which models
would be close up, and they had extra de-
tails, but most of the props people see in the
movie are also subdivision surface models.
Te modelers relied on Autodesks Maya,
using Autodesks Mudbox and Pixologics
ZBrush for nal touches.
One of our main challenges was being
aware of the silhouette so that the props
didnt look overly CG, Lambert says.
Everything you see has some wear and
tear on it, like slightly cracked edges. What
really helped us was putting a slight curve
on the edge, a little bit of displacement
mapping to break up the straight lines you
get with CG models. We modeled as much
of that detail as we could.
Similarly, for the surfaces, Tomov asked
for painted rather than photographic tex-
tures. Evgeni didnt want the surfaces to be
photoreal, says Xavier Bernasconi, surfac-
ing and lighting supervisor at Framestore.
He wanted to use detailed textures, but he
wanted us to lose the edge of the detail. So
we had to brush away the high-frequency
At top: Riggers devised a system for generating hundreds of unique characters, including
these mouse children. At bottom: A mask created for a universal face in each species helped
the riggers quickly create animation tools for disparate characters.
January 2009 22
nnnn
CGI
details on the textures but retain the color
palette. For this, the artists used Adobes
Photoshop and Maxons BodyPaint 3D.
Once completed, the textures moved
through Maya into Pixars RenderMan.
Te challenge was in the quantity:
Te artists needed to surface 40,000-plus
unique assets, including props and charac-
ters. Everything in our sets was modeled,
Bernasconi says. All the objects in mouse
world are human objects. Houses are made
of drawers, with drapes on top. We had little
buttons and little ropes, and little needles
and carpets. And then we had rat world and
the human world. Honestly, when I saw rat
world, I thought, Ill never nish this.
Te artists couldnt paint each surface, of
course, so the team developed procedural
methods to drag and drop patterns that au-
tomatically surfaced marble, stone, wood,
rocks, and so forth with random variations.
Evgeni said, Yeah, thats nice, but its too
clean, Bernasconi recalls. So, Tom Maw-
bi wrote a shader that automatically created
dirt in the corners and streaks of water when
it rained. Te surfacing artists could place
the dirt and streaks, and control the shaders
attributes. To speed rendering, the team
baked the dirt map into point clouds.
Lighting the Way
Once surfacing was under control, Ber-
nasconi moved on to lighting. Light, in
fact, is a central theme of the lm: Light
going away, light coming back, and, of
course, a little ray of emotional light called
Despereaux. Tomovs designs based on the
Flemish painters called for rich lighting
with large areas falling into soft shadows.
Barry Armour, who had worked at In-
dustrial Light & Magic as a CG supervisor,
joined the Despereaux crew in that role and
later became visual eects supervisor for
the project. We used light very carefully,
he explains. We paid homage to the type
of lighting in Vermeer paintings. What we
did isnt groundbreaking technically, but it
was a real pain to achieve with furry char-
acters. Te approach I used was to light the
sets and pop in the characters using that
lighting setup, and add ll lights when
neededand, hopefully, the rst render
out of the box would be decent. Te char-
acters would move in and out of the light
and integrate into the scene.
It would have been impossible to calculate
ambient occlusion for every hair on a frame-
by-frame basis for the mice and rats, so Ar-
mour devised a clever method to achieve the
same result. Wed do an ambient occlusion
pass of the characters skin without the fur,
he says. And then, wed multiply in the fur
as a modulator of that overall ambient oc-
clusion. Tat gave us enough detail to retain
the soft look and still retain the shape.
To light the sets, the artists shot an envi-
ronment map from a central location and
used it as the ambient map. Tat pro-
duced some bounce light, Armour says.
We also could do indirect diuse lighting
using bright objects, like windows, to light
rooms and produce the soft light you see
in the lm. Te challenge was to keep the
light soft enough to look like a motivated
light source for these characters.
To avoid the cost of rendering physically
accurate volumetric light for the environ-
ments, Bernasconi wrote a shader that sim-
ulated volumetric light by spraying Render-
Man Ri curves. Te technique also gave
the artists specic control. Te Ri curve is
a normal curve that RenderMan can draw
as a ribbon, Bernasconi explains. So, I
wrote a simple shader that allowed artists to
change the shape to simulate a volumetric
light, tapering at the source and falling o
at the end. Artists could control the shape
using Maya curve ramps by, for example,
moving a point to change the fallo, and
then could see how the ray would behave
At top, Despereaux (left) lands in rat world, where his rescuer, Roscuro (right), shows him the one
shaft of light in the dark world that was designed to look more Hieronymus Bosch than Vermeer.
At bottom, lighting artists rst lit the sets, and then moved the characters in and out of the lights.
January 2009 23
CGI
nnnn
across its length. Compositors added shad-
ows for any objects crossing the rays.
We could draw as many rays as we
wanted, Bernasconi says. And we could
animate them to make them pulse. It was a
nice way to produce an artistic result.
Early in the lm, for example, theres a
scene with Pea and Despereaux. A bay win-
dow behind Pea lights the whole room. Its
a huge light for Despereaux, says Armour.
It helped us create Despereauxs scale. It
also creates a softer look, which pays hom-
age to Vermeer.
Populating the Kingdom
Each of the worlds in Despereaux had its
own set of charactersa community of
mice in mouse world, thousands of rats in
rat world, and the castle inhabitants and
villagers in the human world.
Gary [Ross] had this amazing scope in
the screenplay that we all loved, says Fell,
noting that because Ross didnt know what
was dicult in animation, he pushed the
crew to do things they might not ordinarily
try. For example, the sheer number of char-
acters is double what youd usually have in
an animated lm, Fell points out.
While Ross concentrated on camera and
lighting, Fell and Stevenhagen concentrat-
ed on character animation. As he had done
for Flushed Away, Fell created little biogra-
phies for each main character, with arc and
diagrams to show range of motion.
For the ve main characters, the mod-
eling team worked from maquettes and
drawings; for the rest, they used ortho-
graphic drawings. For the crowds, the rig-
ging team morphed standard male, female,
boy, and girl models into variations.
We devised a system to generate over
200 unique crowd characters in dierent
species and get diversity, particularly in
their faces, says rigging supervisor Nico
Scapel. We had as many people doing
faces as anything else. Framestore hadnt
done a feature before, so they didnt set
many boundaries. Tey might have lim-
ited the variety if they had had more expe-
rience. But we managed to do it, and they
got more than they expected. Te care that
went into each character, object, and envi-
ronment is amazing.
Te riggers started with a universal face
for each species, a mask they manipulated
to create the various crowd characters. Te
mask could travel from one extreme face
created by the modelers to anotheran
old face, perhaps, or a thin one. We could
then dial in regions of these dierent fac-
es, Scapel says. And, we had controls to
sculpt portions of the faces.
Te riggers generated so many faces that
they created catalogs of characters, and in-
vited the directors in for kill or keep ses-
sions. At the end, the directors settled on 90
humans, 60 mice, and 50 rats with unique
geometry that the texturing department
varied further. Surfacing artists could show
the directors in real time, using OpenGL
renders, those variationsa moustache, a
dierent colored shirt, short hair, and so
forth. It was like real-time costume de-
signing for the crowds, Bernasconi notes.
For facial animation, the riggers built
tools around shapes based on FACS ex-
pressions, with a combination of deform-
ers for sculpting controls. Mostly, we used
blendshapes that we applied to the univer-
sal mask so we could transfer them from
character to character, Scapel says.
For cloth and hair simulation, the rig-
gers devised a hybrid system that com-
bined rigging controls and simulation.
Wed take the result of a cloth sim and
use that as a corrective shape, Scapel says.
We basically baked it and put it back
into the rig.
Te result gave the characters a specic
look and saved calculation time. Similarly,
Princess Peas hair, which the stylists groomed
using guide hairs and tubes representing vol-
umes of hair, moved with a combination of
rigging and simulation. We ran a simula-
tion on the tubes, which represented clumps
of hair, but animators controlled the braids
in front with rigs, Scapel says.
For the princesss skirt and some crowd-
character costumes, the riggers devised
a simple system using dynamic curves.
Imagine a skirt becoming many vertical
strings with a surface that goes between
them, Scapel says. By manipulating the
strings, the artists could control the size of
the wrinkles or create a folded look.
The sophisticated character designs, subtle lighting, soft textures, and live-action pacing caused
directors Sam Fell and Robert Stevenhagens team of 70 animators to balance simple and real
for such delicate characters as Princess Pea, shown at left with her mother and father, and at
right, with Roscuro and Despereaux.
January 2009 24
nnnn
CGI
Delicate Performances
Seventy animators, divided into eight
teams, worked on the lm. Six teams
animated sequences, one handled crowd
animation using Massive and proprietary
software, and an eighth team worked on
nal xes. All the animators performed
characters in each of the worlds.
All the way through the lm, from the
character design to the pacing of the scenes,
we realized we were doing something dier-
ent, Fell says. Te characters have quite so-
phisticated designs, more sophisticated than
the stop-frame characters I had worked on
at Aardman, so we couldnt use a simplied
animation style. But, we couldnt copy real-
ity, either, because there would be no point.
It was a real learning curve for us and the an-
imation teamespecially when it came to
Princess Pea. Its quite easy to over-animate
a delicate character design like that. And, we
shot the lm like a movie, with long shots
where the characters really get to act.
Creating performances that t with the
subtle lighting, soft textures, and the live-ac-
tion pacing was a challenge for the anima-
tors. When a character has a big mouth
and eyes, is funny, rubbery, and moves a lot,
you can get away with murder, says anima-
tion supervisor Gabriele Zucchelli. But,
with a pretty character like the princess,
every gesture needs a delicate touch. Every
millimeter counts. Te mouth shapes have
to be right. Te shape of the eyes has to hint
at emotion. Its the dierence between using
a sword or a needle.
Te other hero characters, although
not as delicate, required the same light
and careful touch in animation. So did
the secondary characters. Te chef needed
an aristocratic attitude. Te jailor shows
a sweet side beneath his rough exterior.
Despereauxs parents worry. All the char-
acters, even though theyre stylized, needed
to move naturally, Zucchelli says, so we
needed rigs that behaved properly. We had
modules and mixtures of blendshapes,
sculpts, deformers, and lattices appropriate
to what we needed so we could tweak the
animation depending on camera angles.
To help with precision animation, the
animators could view the characters with
and without fur. Tey could also place a
fake iris and pupil in the rig to see how
rendered reections might aect eye lines.
Te modelers built the eyes like real eyes,
with a cornea that behaves like a cornea,
Zucchelli points out. And that slightly
skews the light. But to put reectivity into
the face rig would suck too much memory
to be interactive. So we had a little cheat
that gave animators the eect.
Although challenging, Zucchelli found
the work exciting. It was a refreshing expe-
rience, he says. Occasionally I felt, Oh my
God, I wish I could express the emotions
more openly. But, we found that the more
subtle the performance, the more the cam-
era loved it. We learned to communicate in
a naturalistic way, and it felt more real.
Looking back, there were many rsts
in this production. Tomovs rst 3D lm.
Te rst animated feature for Ross. Te
rst time Blackbourn had worked on cam-
era moves and composed shots before any-
one storyboarded the lm. Te rst time
some of the animators had created such
elegant, rened performances for characters
in a computer-generated lm. And, the rst
animated lm production for Framestore.
And yet, or perhaps because of all those
rsts, Despereaux works. Just as Ross had
woven the separate stories from DiCamil-
los book onto a lyrical screenplay, the de-
signs created by Tomov, the camera moves
led by Ross, soft lighting implemented by
Armour and his team, and subtle character
performances directed by Fell and Steven-
hagen blended into a cohesive lm unlike
any other animated feature.
Im proud of this movie, Ross says.
I think weve done something thats dif-
ferent and beautiful. I feel blessed that we
were able to do such a unique movieand
maybe a little spoiled. n
Barbara Robertson is an award-winning writer and a
contributing editor for Computer Graphics World. She
can be reached at BarbaraRR@comcast.net.
The lighting artists often used diffuse lighting from big, bright objects, like windows, to produce
soft light, especially for shots with tiny Despereaux, shown here reading a fairy tale rather than
eating the book, as he was supposed to do as a mouse.
40,089 individual assets
25 hero characters modeled,
surfaced, and textured
12 secondary characters
60 hero environments modeled,
surfaced, and textured
413,138 hairs on Despereauxs head
1726 nal shots delivered
1200 shots delivered during the last
month of production
278 people on the crew
126,248 frames altogether
CGW :1208_p 11/24/08 12:20 PM Page 1

Web/Virtual Reality
Virtual worlds are coming of age, and
theyre getting real...more real than ever.
During the past ve years, there has been
an explosion of sites that oer visitors to
persistent worlds a wide range of activities,
including playing games, having adven-
tures, and meeting people. In almost all
cases, the real attraction of 3D worlds is
communitythe chance to meet and in-
teract with people in a virtual space. And,
for many who visit 3D virtual communi-
ties such as Second Life, Tere.com, En-
tropia Universe, Active Worlds, and others,
playing in virtual worlds is creatively ful-
lling. People in virtual worlds can create
an identity, clothing, buildings, and land-
scapes (see Editors Note, pg. 2). As it turns
out, the drive to create is every bit as com-
pelling as the drive to blow things upthe
standard fare in most computer games.
Whats especially intriguing about this
new wave of interest in virtual worlds
is that 3D has the potential to become a
mainstream capabilitynally.
Obviously, 3D worlds are not new. Tey
arrived with browsers and the Internet, and
interest reached near-hysterical levels in the
early 1990s, during which time enthusiasts
asked, why have 2D browsers that mim-
icked printed pages when we could have 3D
browsers to wander around in? As so often
happens, the answer has been a long time in
coming, and its short and sweet: Why not?
LucasFilms Habitat is often consid-
ered the rst virtual world to use avatars,
though they were 2D. Te site gave people
an apartment and served primarily as an
animated chat. Habitat started in 1985 at
a time when most people didnt have com-
puters that could even run it, or Internet
connections fast enough to make the pro-
cess anything less than painful. Te spirit
of Habitat lives on in Habbo Hotel,
which also oers denizens their
own living space to use as a base.
One of the longest running
3D environments is Active Worlds,
started by Ron Britvich as WebWorld in
1994, which he ran on the Peregrine Sys-
tems servers after hours. Te project ew
under various company ags, and the site
was renamed AlphaWorld and, eventually,
Active Worlds, its current moniker. It was
kept alive by the stubborn persistence of de-
velopers who built and maintained a variety
of worlds. By the late-1990s, several other
companies emerged and got a running start
in this arena, including Tere, founded by
Will Harvey and Jerey Ventrella.
Te path to these online virtual commu-
nities was paved in large part by the eorts
of early 3D online content devel-
opers, such as Tony Parisi of Vi-
vacity. Parisi has been a ghter
on the front lines of the
Ghos ts
January 2009 26
January 2009 27
Virtual Worlds
nnnn
During the past
several years, virtual
community sites have become
increasingly popular and have
expanded to meet the growing
needs of their visitors. One of the
more popular sites is Second Life.
January 2009 28
nnnn
Virtual Worlds
VRML wars, the attempt to develop a stan-
dard format for 3D online, and his com-
pany is continuing work on the concept of
3D browsers. Agreeing that virtual worlds
have matured in the 10-plus years hes been
working in the eld, Parisi says, Tey are
becoming more diverse and cater to more
types of users and more mainstream us-
ers. As technologies emerge that allow
consumer-friendly virtual worlds running
inside a Web browser, the barriers to entry
are disappearing. Also, notes Parisi, the
barriers to content creation are eroding.
Today, 3D virtual communities are
home to millions of peoplea rough, seat-
of-the-pants estimate based on published
gures from the largest sites puts the num-
ber at approximately ve million. Second
Life mogul Philip Rosedale, chairman of
parent company Linden Labs, claims Sec-
ond Life is growing as much as 10 percent
every year. Other analysts put the overall
growth of virtual worlds, including gaming
sites, at 15 percent per year. Meanwhile,
market research company NPD puts the
total number of US subscriptions to online
worlds at $1 billion.
Every virtual world has a slightly dif-
ferent focus, and the worlds within these
worlds are often very dierent as well. Te
three worlds we examined closely for this
storySecond Life, Tere.com, and En-
tropia Universeall place a very strong
focus on creativity, all three have virtual
economies that the companies work hard
to protect and present as secure, and all
three are evolving rapidly as a result of resi-
dent participation.
Follow the Money
In the case of Second Life, by far the largest
and fastest growing of the sites, the company
has struggled with change on almost every
front, and not all of it is positive. Te rm
has had to ramp up its servers, it has to con-
stantly convince the outside world of its secu-
rity for young visitors (and, at the same time,
it aunts the raunchier aspects of the world
for adult visitors), and it has to contend with
a massive redistribution of wealth.
As a result of a federal investigation,
Second Life was forced to close down its
casinos. Moreover, the universe tightened
up its banking system, thus causing a con-
traction of Linden Dollars, which are po-
tentially vulnerable: Te Web is alive with
anecdotes about denizens being robbed of
their accounts. Last, Rosedale has retreated
from the CEO post to become chairman.
Te company isnt talking about why Rose-
dale is out as CEO, but ocials said they
were looking for a new CEO with more
operational and management experience,
though the money issue could not have
helped the situation. Recently, Mark King-
dom assumed that role.
Most chats with virtual world moguls
generally turn back to the money. Teyre
under pressure to demonstrate the success
of their brave, new businesses by making
lots of money, and theyre also constantly
challenged by residents and outsiders to
account for where the money goes. If the
money isnt controlled, theres the potential
for ination, and if residents are afraid their
money will disappear, they wont invest.
Entropias director of business develop-
ment, David Simmonds, is proud of the
banking system within the Entropia Uni-
verse, wherein users can cash out if they
want (however, there is no interest earned).
According to Simmonds, the Entropia Uni-
verse brought in $3.6 billion PED (Project
Entropia Dollars), or $360 million in US
dollars, in 2006. Sales of Linden Dollars in
Second Life, one form of revenue for the
company, is about $720,000 a month, or
about $8.6 million a year.
Marketing and sponsorships represent
another major source of money for the
virtual communities. Companies like Sun,
Intel, IBM, and Cisco use Second Life for
information hubs as well as meeting sites.
Te companies claim that their employees
get more out of meetings held in Second
Life compared to meetings on the phone
because they are more engaged, and no
doubt the novelty is an attraction.
However, its also worth noting that these
companies are also selling a bunch of servers
and networking equipment to rms build-
ing virtual worlds, so they have a vested
interest in making the business model for
virtual worlds pay o and grow. Tis is a
role these companies have long played in the
economy of virtual communities, and its a
valuable one because they help provide an
economic underpinning to support other
uses of virtual communities. Increasingly,
these technology companies are being joined
by retailers, universities, and media compa-
nies, as well as individual entrepreneurs.
Entropia Universe is ramping up to ac-
commodate specialized areas within the
Many of the virtual worlds focus on creativity, starting with the construction of a unique avatar.
January 2009 29
Virtual Worlds
nnnn
site. For instance, the Chinese government
has been developing its own planet in En-
tropia. Its own Cyber Recreation Develop-
ment Corp. has planned a planet containing
education, gaming, and entertainment areas
that, when fully realized, will support up to
a million users. Tis addition is expected to
add more than a million new residents to
Entropia. Considering that Entropia cur-
rently supports about 10,000 concurrent
users, this is a very ambitious project.
In addition, the Creative Kingdom, Inc.
(CKI), a thematic architectural rm, is
branching out and addressing the virtual
world. CKI is building a new studio com-
plex in Chiang Mai, Tailand, and the rst
project for the new studio is a new planet
for MindArk. Other projects from CKI
include the Palm Island, World Island, and
Madinat Jumeirah Resort in Dubai, UAE.
Tough the MindArk job is the rst virtual
project for the company, CKI and the gov-
ernment of Tailand are focusing on the
development of jobs in online creation for
the people in that country. MindArk hopes
the project will encourage other companies,
or countries, to form their own planet.
Meanwhile, MTV is building a world in
the online space Tere.com, where it can
appeal to its natural demographic: teens.
Additionally, the participation of uni-
versities is helping to grow the diversity in
virtual communities. Furthermore, uni-
versities are experimenting with holding
classes in 3D worlds. Te Los Angeles Art
Institute uses Second Life as a training tool
for students in Interactive Design as well as
a forum for online classes. Tis makes a lot
of sense in Second Life, where an aptitude
for graphics can generate revenue. Its also a
place for the university and its students to
show o their work. For example, scientists
and colleges are using 3D worlds to dem-
onstrate augmented reality techniques that
combine real-world objects with virtual ob-
jects. Yes, dear friends, the head-mounted
display is back. And Georgia Tech is only
one of several universities that have a site
on Second Lifes Augmented Reality Island
to show o its experiments.
Creating a New Reality
Arguably the real source of income in 3D
virtual worlds stems from peoples urge to
create. After all, standing around and chat-
ting tends to get old after a whileits kind
of like a long, awkward cocktail party un-
less you have something to do and to talk
about. You can even turn yourself into a
more interesting person. You can become a
real estate mogul who buys land, improves
it, builds on it, and sells it. If that doesnt
work, dance. In Second Life, dance is an
art form. Acquiring dance moves is an im-
portant pastime for some, and supposedly
there is work available as dancers in the var-
ious clubs to give the illusion that there re-
ally is a party going on. Its not the kind of
work your mother would approve ofbut
you are a dierent person in Second Life,
and so is she. For all you know, shes the
fairy dancing next to you.
Specic tools have been created for con-
tent creation in Second Life and Tere.
com, but users can also use third-party
tools. Active Worlds oers free access to
Caligari TrueSpace, an easy-to-use 3D
modeler. In contrast, Entropia Universe
oers its users in-house creation tools, and
that ts into the worlds philosophy, which
is based on its own conservation-of-energy
theory: Energy can neither be created nor
destroyedexcept by MindArk, Entropias
parent company.
MindArk has two goals in tightly control-
ling objects within the Entropia Universe.
First, the company wants to be sure that
creations complement the look of Entropia
Universe. However, it also wants to limit the
number of objects that are sold or bartered
in Entropia so that they maintain their val-
ue. In contrast, Second Life residents often
take pleasure in making gifts of clothing and
other objects rather than selling them. Its the
happy newbie who runs across these types.
With its philosophy of tightly control-
ling goods, Entropia Universe more closely
resembles the controlled worlds of online
games rather than 3D communities. Sim-
monds explains how he sees the mantra
that governs online economies: Tere are
people with lots of time and no money, and
people with lots of money and no time. In
Entropia, you cant make something from
nothing, he points out, so people can go
in and work with stu, very much like in
the real world. For instance, they can take
lowly jobs, like gathering dung for fertil-
izer, to earn money. People with property
can improve their land and allow hunting
or host other activities on it.
If youre a successful businessman, theres
Here, virtual visitors participate in a party, held within the Entropia Universe.
January 2009 30
nnnn
Virtual Worlds
a very good chance youll be a good busi-
nessman in Entropia, says Simmonds.
Most often, MindArk handles the work
of creating sites for advertisers in Entropia.
Te company uses Autodesks 3ds Max pri-
marily to build content for the companys
advertisers and clients. However, according
to Marco Behrman, Entropias CTO, com-
panies can build their own content and im-
port it into Entropia if it meets MindArks
specications.
Despite its strict rules governing content,
Entropia continues to evolve. In addition
to the new planets from China and Cre-
ative Kingdom, MindArk is incorporating
the CryEngine 2 from Crytek, which will
enable more advanced graphics (and raise
the level of computer needed to run Entro-
pia). Behrman insists that Entropia is in-
terested in creating areas where users might
have more freedom to create and share their
own content as long as the overall quality
and feel of Entropia is not diluted.
In comparison to Entropia, Second Life
is the Wild West. In general, if you can g-
ure out how to do it, you can do it in Sec-
ond Life, and that extends to skin texture,
dance moves, wardrobe, and sex (which in-
volves, presumably, plenty of imagination,
if not technical knowledge).
So, it should not be surprising, then,
that Second Life leads the pack in available
tools and tutorials. Taking advantage of
YouTube, Linden Labs oers tutorials by
the ever-enthusiastic character named Tur-
ley, who takes everyone from the hapless
Noob to the Adept through a series of easy-
to-follow steps to create custom avatars and
clothes, homes, and furniture. Users pay to
upload content into Second Life, but the
cost is minimal, and the average visitor can
get pretty far without spending a dime. In
Second Life, the economies of scale are at
work, and the number of visitors to Second
Life contributes to the availability of free
stu and things to do at the site.
Content can be created in Second Life
using basic tools that generate prims, or
primitives. Tis can be a fairly unreward-
ing process for those who have not spent
enough time with Turley and his tutorials.
Tere is also a free tool called Sculpty Paint,
which gives users the ability create 3D
models with textures and import them into
Second Life. Its tricky and not exactly in-
tuitiveusers save sculpture maps and tex-
ture maps, import them, and apply them to
prims in Second Life. Te approach can be
used with other 3D modeling tools as well,
but Sculpty was designed to make content
creation easy for Second Lifers.
Tere.com exists somewhere between
the tight control of Entropia and the more
freewheeling Second Life. Kids, it should
be noted, are often just as happy to be pro-
tected, and thats the philosophy of Tere.
com. Michael Wilson, president of Makena,
Tere.coms parent company, says there are
well over a million subscribers to the com-
munity, which oers a PG-13 environment
thats evident as soon as one sets a digital
foot into the realm. Avatars are not quite as
well endowed as those in Entropia and Sec-
ond Life. Tey wear slightly less revealing
clothing, andperhaps as a resultthe in-
teraction is quite open and friendly. An early
visit to Tere.com resulted in an interesting
chat with a young man (one presumes) who
mentioned that he was from Jordan.
In regard to content creation, Tere.com
requires people who want to build things
inside the world to sign up for its developer
program. Te site oers users an easy-to-
use tool called Style Maker for generating
content, though content created in other
3D modeling software also can be im-
ported. Independent of the DCC software
used, the models polygons and vertices are
limited because it will aect performance
within the world. Other DCC tools are
available as well, including a vehicle painter
that enables folks to paint objects from a
car to a surfboard.
However, anything created for Tere.
com must rst be approved by the com-
pany before it can be imported into the
Tere.com world. Content is monitored to
be sure its appropriate and for trademark
protection. Furthermore, Tere.com is add-
ing support for Collada, an open exchange
approach developed by the standards body
Khronos, which will allow easier integra-
tion of outside-created content.
Not surprisingly, notes Wilson, the largest
sub-economy of Tere.com is the market for
womens clothes: Tere are about 20,000
dierent womens clothing items, he says,
and they all t. Te next largest market is
for houses. Tere.com is also building up its
store of tutorials to help keep users engaged,
and the site fosters creativity in its users by
There.com is not as restrictive as Entropia nor as lenient as Second Life in terms of its virtual
activities, which include (from left to right) a fashion show, game show, and amusement park.
January 2009 31
Virtual Worlds
nnnn
sponsoring contests and showcases for art
and creations. It, too, has its share of ma-
chinima tutorials and authors.
The Why
All of this brings us to this question: Why?
Why spend good money on stu thats not
real? Why spend time searching out strang-
ers when it seems theres never enough time
to spend with RL (that would be real life
to you and me) friends and family?
Te answer: Because its fun. Playing in
virtual worlds oers people the same kind
of escape that others get from rst-person
shooters or from a romance novel. People
are more engaged, they meet people, and
they get to reinvent themselves. Studies
have found that some people self-actual-
ize through their avatars. People who cre-
ate thin avatars for themselves may start a
weight-loss program. Shy people can cre-
ate more outgoing alter egos in the virtual
world. Second Lifes Rosedale is convinced
that all online transactions will be carried
out through avatars in the future. Tere.
coms Wilson sees virtual worlds as an ad-
junct to other forms of communication,
but he, too, expects to see more and more
people communicating through avatars.
Tere is a virtuous circle forming around
content creation on the Web, particularly
when it comes to 3D. Tere has always
been a lot of interest around 3D modeling
and animation, but it is a hard craft to learn,
even with the arrival of low-cost tools, such
as TrueSpace, which has been acquired by
Microsoft, interestingly enough; SketchUp
from @Last Software, acquired by Google
(ditto intriguing); and the open-source tool
Blender from Ton Roosendahl. But what
good is all the eort to learn 3D if no one
ever gets to see and appreciate your work?
Online worlds change all that, and not only
do they provide a showcase, but they also
provide a marketplace. Te same is true of
video, machinima, and graphic images.
Among the fairly recent creations is a new
lm being presented through Cinemax.
Told in short installments, Molotov Alva
and His Search for the Creator: A Second Life
Odyssey tells the story of Alva, an accepting
soul who has abandoned real life to take up
permanent residence in Second Life. Alva
ponders the age-old question of existence,
with a Pirandellian twist: How do I dene
myself in this virtual world, and what about
the lmmaker, Douglas Gayeton?
What is the story of Molotov Alva? Is it
really a documentary? Not in the sense we
are used to, but lmmaker Gayeton opens
the question. If Gayeton tracks the process
and interactions of his ctional character,
why isnt it a documentary in the same way
that Borat is a documentary? Te digital
revolution means that all forms of media
are being fundamentally redened. Te
participants and creators in 3D virtual
communities are in the vanguard of those
who are doing the redening.
Te early pioneers who built Habitat, Ac-
tive Worlds and the
myriad universes out
there didnt really ex-
pect the evolution
of machinima. Tey
didnt anticipate TV
series combining real-
life instant messag-
ing, videos, and vir-
tual worlds, like the
hit Te Truth About
Marika, a self-re-
exive thriller/puzzle
presented by Swed-
ish television that
promised to get to
the bottom of myste-
rious disappearances
of young people in
Sweden.
Te show used re-
alistic newscasts, and
included a sub-plot
in which characters
accused Swedish TV
of being in on the
conspiracy. Clues were presented on the In-
ternet and in the TV show, and it included
characters living in Entropia. Te series was
addictive for many viewers who eagerly
worked through the clues.
Te early pioneers certainly did an-
ticipate visitors negotiating virtual worlds
with the help of 3D glasses and controls,
and theyre going to see more of that sort of
thing for industry as well as entertainment.
In essence, virtual online communities pro-
vide an architectural platform for scientists,
creative people, and hobbyists alike. Tere
really are worlds within worlds, and theyre
all evolving according to the whims and
desires of the people who live in them. n
Kathleen Maher is a contributing editor to CGW, a
senior analyst at Jon Peddie Research, a Tiburon, CA-
based consultancy specializing in graphics and multime-
dia, and editor in chief of JPRs TechWatch. She can be
reached at Kathleen@jonpeddie.com.
Create 3D Faces Today!
Take any photo and
turn it into a 3D head.
Use it as a morph with
any application that reads
OBJ, like Maya, 3DS Max,
Poser, Lightwave, DAZ
Studio, etc.
Available in WIN/MAC
$89.99
1. FaceShop
2. FaceShot 3D Camera system
3D Mesh on a budget.
FaceShot comes with 2
Canon EOS XT cameras,
mounts, hardware and
FaceShot 1.0 software.
Creates full 360 3D
heads with 30,000
polygons in minutes.
Starts at $1,799
(Compare to similar
systems costing 10x, 20x
as much)
www.abalonellc.com
CGW_1 FOURTH:CGW_1 FOURTH 10/23/08 5:18 PM Page 1
Massively multiplayer online role-playing games (MMORPGs) are just that, games. Yet, for
the millions of people who play them, they are a means of escaping from everyday life, oer-
ing the ability to become someoneor somethingelse.
MMORPG players invest a signicant amount of time living inside their virtual world
of choice, whether its World of Warcraft (WoW), Second Life, EverQuest, or some other al-
ternate-reality universe (see Editors Note, pg. 2). At times, players friendships inside these
games are just as strong, or even stronger, than their real-life bonds. Tis despite the fact that
they know almost nothing about their cyber buddies outside the boundaries of the virtual,
make-believe worlds in which they live and play.
So, who are some of the real people who inhabit these worlds? Are they anything like their alter
egos? And, what happens when reality starts intruding on fantasy, and vice versa? Tose are some
of the issues that are brought to light in the documentary Second Skin from Pure West Films. Te
movie, which introduces viewers to some real people who populate online virtual worlds, utilizes
in-game sequences to punctuate the information and contrast the real and alternate lives of
several gamers who are the subject of the live-action movie. Te documentary focuses on
how being an avid gamer has transformed and aected their lives over the course of a few
years, says Juan Carlos Pieiro Escoriaza, director and editor of Second Skin.
Second Skin follows the real and virtual lives of several
MMORPG players who log countless hours each week
playing their favorite online game. The documentary
incorporates the players avatars into the live action.
January 2009 32
nnnn
Film
Welcome to My Life
Te documentary is Escoriazas rst feature
lm. He, along with a small crew, began
working on the project in early 2006 and
completed it just in time for its debut at the
SXSW lm festival in the spring of 2008.
Since then, Second Skin has been shown at
other festivals and will hit select theaters
this spring, followed by a DVD release
shortly thereafter.
So what made Escoriaza choose such an
unusual subject matter? As the lmmaker
tells it, he was introduced to these alternate
worlds by two avid players: his brother,
Victor (one of the lms producers), and
a mutual friend. Just going into a virtual
setting for the rst time and seeing these
other people interacting with each other in
this alternate world was amazing, he says.
Our friend had all these responsibilities in
the game and was getting married in real
life. It took a strange turn because he was
so invested in the game that he had di-
cultly managing his time between the two
worldsthe real and the virtualbecause
both were important to him.
As it so happened, Escoriaza had started
a documentary company with Second Skins
other producer, Peter Schieelin Brauer, a
few years earlier, and he thought the yin
and yang of real and virtual lives would
make for good storytelling.
Te 94-minute documentary contains
three main story lines. Te rst focuses on
a mini guild from Fort Wayne, Indiana,
that includes a seasoned WoW player and
his wife, along with three neighbors who
each devote 20 to 50 hours a week playing
WoW. Tey are part of a larger group who
work together at a cell phone company, live
in close proximity to one another, and log
untold hours on WoW. Te second story
follows a couple who met virtually while
slaying dragons in EverQuest II, fell in love,
afterward traveled many miles to meet in
real life, and then moved in together. Te
third story introduces viewers to an MMO
addict who lost just about everything in
real life when he neglected reality for fan-
tasy and played MMOs 16 hours a day.
Te crew spent a good deal of time lm-
ing the subjects at their respective locales.
However, Escoriaza wanted to be sure that
the concept of a dual existencethe per-
son and the alter egowas apparent.
I wanted to mash the real and vir-
tual worlds in a way that you could co-
hesively understand them. I wanted to
show the players literally turning into
someone else when they were online, and
then turning back when they exited the
game, he explains.
Achieving that visual link required using
game footage, mixing the CGI with the
live action. For the actual game footage ac-
quisition, the group used Fraps, a free uni-
versal Windows application for real-time
video and screen capture. Te imagery was
then imported directly into Adobe After
Eects, where Escoriaza and Brauer began
adding the various controls, movements,
and so forth.
But oftentimes the real game footage
was, well, a little constrictive. Te move-
ment inside the game is not made for lm-
ing; its more about moving around inside
a 3D environment, Escoriaza says of the
reason behind the alterations.
As a result, the team had to go out of
the game to create so-called composites
from the game footage, using the WoW
Model Viewer to animate a loop of game
footage on a bluescreen. First, the editors
captured a still image inside the game,
and then stitched together the stills using
Adobes Photoshop. Next, they imported
that material into After Eects, where they
mapped the composite with a background
of another still image and elements of dif-
ferent avatars from the WoW Model View-
er. After we put a background behind [the
image], wed make camera movements,
like a pan, tilt, or dolly, to essentially ani-
mate these characters and creatively make
sequences that were more vibrant than we
would have had from just moving around
inside the game world, Escoriaza says.
According to Escoriaza, three quarters
of the game footage was taken from the
MMOs, while the remainder was created in
the WoW Model Viewer and in After Ef-
fects. Yet, even the in-game imagery was ma-
nipulated to some extent inside After Eects.
I would go in as an avatar cameraman in the
rst-person mode and capture a PvP (player
versus player) battle, for instance, or an in-
timate moment between the avatars. Ten,
later, I added eects to the images to make
them feel more real, explains Escoriaza.
However, for the EverQuest scenes, all
the footage was captured in-game (no blue-
screening was done). Players would enter
the game as their avatars, and Escoriaza
Second Skin follows the real and virtual exploits of Matt Elsworth (shown at right), a devoted
WoW player from Fort Wayne, Indiana. To the left is his avatar.
January 2009 33
Film
nnnn
January 2009 34
captured the footage on his computer using
Fraps. Later, the team integrated the in-game
imagery into the lm, which was edited in
Adobes Premiere Pro. My motion graph-
ics animator used After Eects, so he would
send me the source les, and we would swap
project les back and forth to do any quick
changes, says Escoriaza. Immediately the
new le would pop into my Premiere time-
line, and I could just press play to review it.
Real Challenges
According to Escoriaza, the biggest techni-
cal challenge the group faced was bring-
ing the avatars to life in a way that viewers
could relate to them. Te diculty was
capturing in-game shots the way we want-
ed them, he says. Its not easy to do cam-
era movement whereby you are completely
in sync with the other avatars. I fell o
clis quite a bit, and many times a monster
would kill me during lming. Sometimes
I couldnt reach the perfect spot for a shot
because the avatar just couldnt get there
its not always logistically possible.
Initially, the team tried to capture all
the in-game footage through Fraps. As the
work progressed, the lmmakers found
themselves using more and more of the
bluescreening technique, capturing still
images inside the virtual spaces.
To link the real and virtual scenes more
tightly, Escoriaza used a shaky-camera style
while lming in each. While After Eects
oers a number of presets to achieve this
look, the crew instead meshed together six
to 10 dierent eects, some of which were
generated from scratch.
Despite the dramatic increase in people
who play video games, whether single play-
er, multiplayer, or massively multiplayer,
there have been few attempts to combine
actual game footage with live action to any
extent. Yet, the action taken by Pure West
to do just that may become commonplace
in the near future. Tese avatars are be-
coming so personalized and are ingrained
deep into peoples psyche, and they are an
important part of themjust as important
as their real lives, says Escoriaza.
Who knows, shortly we may see a trend
toward home movies that not only include
real-life friends, relatives, and pets, but vir-
tual ones as well. n
Karen Moltenbrey is the chief editor for Computer
Graphics World.
Filming in the real world at times could be
nearly as challenging as doing so inside the
virtual worlds.
CGW_Half_HOR:CGW_Half_HOR 7/15/08 1:28 PM Page 1
nnnn
Film
Early Bird Ends Thursday, February 12, 2009
Register now and save up to 30% on conference passes
Visit GDConf.com for the latest sessions, keynotes, and news
Game Developers Conference


March 23-27, 2009 | Moscone Center, San Francisco
Attend the Visual Arts and Design Tracks at




































































CGW :109_p 1/5/09 3:45 PM Page 1
Autodesk
University (AU) has grown enormously in
the 16 years of its existence. It has grown
from a homey little conference with a few
tracks to a large conference and trade show.
And, over that same period, Autodesk has
grown and changed so that its a completely
dierent companyas it should be. CAD
is a dierent discipline, and no one has
stood still in the industry.
Te way Autodesk sees it, the point of
dierentiation is digital prototyping. Its
not a new idea, and Autodesk, for one, has
been preaching the doctrine for some time
now. For Autodesk, though, digital proto-
typing has become the club with which to
whack competitors in the project life-cycle
management (PLM) businessthe idea
being that better methods of building and
manufacturing are the true concerns of
CAD, as is project data managementand
not the squishy concept of PLM.
Autodesk has developed the idea of
digital prototyping broadly so that it en-
compasses every product the company
makes. Even its technology for entertain-
ment content creation is gradually getting
sucked into the digital prototyping lineup.
Autodesk sees digital prototyping as the
ability to plan, create, validate, and docu-
ment a design before building it and then
for maintaining it after production. So, it
includes conceptualization tools, includ-
ing a newly revived AutoSketch and Alias
Studio, as well as more exible sketching,
drafting, and modeling tools in products
like AutoCAD, Revit, and Inventor.
On the other side, digital prototyping
encompasses better visualization, and thus
includes Showcase, a version of 3ds Max
that is tuned for CAD visualization in 3ds
Max Design, and a program still in the Au-
todesk Labs skunk works called Newport,
which uses a game engine to enable fast vi-
sualization to try out ideas and collaborate.
And last but not least, because digital
prototyping means creating an absolutely
accurate and manufacturable digital ver-
sion, the data can be used to actually create
an object in the real world using CAM/
CAE, rapid prototyping machines, or
maybe hammers and saws. Te design in-
formation that went into building a prod-
uct can be used to create documentation,
to provide service documents, and it might
even be used when it comes time to recycle
or tear down what was designed and built
in the rst place. Tese are great ideas, but
so far were not living in a world where very
many companies really take advantage of
the cradle-to-grave approach.
The Big Push
At Autodesk University, Autodesk sought
to sell its ideas about workow as well as
to train users of its product in its way of
thinking. Te company is positioning itself
as the leader in CAD, and it can do that on
the basis of its user population. Autodesk
University had an attendance of more than
10,000most of whom are using Auto-
Autodesk and Stratasys
demonstrated a motor-
cycle built entirely from
parts generated by rapid
prototyping.
January 2009 36
nnnn
CAD
Autodesk University helps its CAD and manufacturing
users expand their knowledge base
CADand Autodesk is focused on mov-
ing them on to its advanced tool families
built around Revit and Inventor.
As part of the plan, Autodesk has been
building out its product line to incorporate
more analysis of every kind into its prod-
ucts. Te idea is that the more users who
can test the possibilities of a design, the
better the design. While analysis has tra-
ditionally been associated with mechanical
CAD and, specically, nite-element anal-
ysis, in reality, analysis is part of all aspects
of design. After all, the rst question of any
design is, will it work?
Autodesk has been steadily incorporat-
ing the traditional types of analysis into its
products with technology licensed from
Ansys. In addition, the company acquired
PlassoTech in the summer of 2007 to in-
crease its in-house analysis capabilities, and
in 2008, the company further increased its
portfolio with the acquisition of Moldow.
And, just a week or so after AU, Auto-
desk announced the acquisition of analysis
pioneer Algor for $34 million. With this,
Autodesk rounds out its analysis portfolio
and oers a suite a capabilities quite broad
in comparison to competitors PTC, Das-
sault/SolidWorks, and Siemens.
Like digital prototyping, Autodesk is ex-
panding the concept of analysis to include a
variety of approaches and disciplines, includ-
ing stress analysis, structural engineering, civ-
il engineering, site planning, environmental
evaluation, cost analysis, project planning,
and so forth. Te challenge of bringing the
cost eciencies exploited so well in the man-
ufacturing side of industry to architecture
means enabling contractors from dierent
companies and dierent disciplines to com-
municate, and Autodesk has been putting a
set of tools together to enable that.
Its All About Design
Designers are cool. Industrial designers
and architects are pretty much the rock
stars of the CAD world. Dont believe it?
Go to a Pecha Kucha, a hip party featuring
all kinds of designers who describe the pro-
cess to a project with 20 slides delivered in
20 seconds per slide. Autodesk rened the
idea with DesignSlamswhich combined
the Pecha Kucha concept with a poetry
slam. Designers had 20 minutes to deliver a
concept while rock music played and judges
watched the designers work. Tere were two
segments: one for architecture and one for
industrial design.
Design is cool, but designing with the
tools we currently have is hard.
Autodesk CEO Carl Bass, who some-
how manages to embody a stern practical-
ity with enthusiastic vision, prowled the
stage during his keynote, literally putting
the spotlight on cool, new developments
and products. He got a lot of help in the
oh-wow department from Autodesk CTO
Je Kowalski. Neither of them bothered
showing the audience the new stu in
AutoCAD.
However, Bass did quote Linus Pauling,
who said that the best way to have a good
idea is to have lots of ideas. Autodesk has
lots of ideas, and at AU, the company dem-
onstrated some of the work in progress. For
example, Autodesk showed a conceptualiza-
tion approach that enabled users to create
variations on a design using biomimicry
principles. Te idea is that once you have
a general shape, the computer can algo-
rithmically generate new variations on that
shape to help spark new ideas. Now, heres a
chicken-and-egg question: Are designs be-
coming more organic because the tools are
enabling surfaces and more uid shapes, or
are tools better able to handle and produce
organic shapes because designers are doing
more of this kind of work? Teres a cross-
pollination going on, and it will continue.
Its technology that has been around for a
while and has been demonstrated by Bent-
ley Systems as well as Autodeskand by in-
dependent third-party developers that have
created plug-ins for products such as 3ds
Max. Weve also seen similar approaches for
color, including Adobes Kuler plug-in and
AIR application. In a phone conversation
with designer Eric Adigard of MAD Studio
just after the keynote, he agreed that itera-
tive design is a powerful tool and says, Te
way for the designer to ght back against
the computer is to demand that it come up
with alternatives.
Autodesk also brought its ideas about
digital prototyping out to the stagein a
spectacular demonstration of the capabili-
ties of rapid prototyping and new materi-
als, a motorcycle built entirely out of parts
generated by a Stratasys rapid-prototyping
machine was lowered to the stage and
started up. Its lights came on, and an en-
gine noise emanated from the cycle. But, as
a matter of due diligence, its worthwhile to
add that Stratasys competitors Objet and
NextGen both expressed doubts that the
Douglas Look of Auto-
desk Labs demonstrates
how far more intuitive
working in Inventor
might be with a multi-
touch computer. In this
case, Autodesk worked
with Perceptive Pixel
to create a multi-touch
interface.
January 2009 37
CAD
nnnn
January 2009 38
motorcycle was actually started, and this
reporter does have her gullible moments.
Whats New
Progress doesnt stop, it speeds up.
Kowalski expanded on the idea of put-
ting the computer to work. Each year, he
said, more computer power is produced
than the sum of all that has come before.
Kowalski told the audience that the future
is coming more quickly than we ever real-
ized. Building on the iterative computing
idea, he asked, Why does the computer
make us tell it what to do, as if its resources
were valuable and ours are not?
Kowalski built on the idea of iterative
design and provided a sneak peek of what
Inventor might be able to do, and maybe
even pretty soon at that. He demonstrated
direct modeling being used on an Inven-
tor model. To digress a minute, the next
big revolution in mechanical design is an
old revolutiondirect modeling, which is
a friendlier approach, letting designers try
out dierent ideas interactively. Its opposite
is constraint-based modeling, which ensures
that a manufacturable design is produced
but is restrictive when one just wants to try
out some ideas. Autodesk demonstrated the
ways in which both approaches could be
used together on the Autodesk University
stage, and its an idea that is going to keep
Inventor current with its competitors.
Te idea of using the computer to gen-
erate ideas and alternatives is another as-
pect of the broadening concept of analysis,
as Autodesk is putting it to work. On one
hand, theres traditional analysis that takes
into consideration variables such as stresses,
materials, heat resistance, and aerodynamics,
and demonstrates areas that need attention,
giving the designer an idea of where to try
out alternatives. Te point to iterative design
is that the designer quickly generate alterna-
tives. Te concepts can be used together as
well as separately to nd the best design or,
better yet, nd a brand-new idea that hasnt
been tried before.
Kowalski showed o other ideas that are
being tried out in the labs, including com-
puting in the cloud with Project Showroom,
which Autodesk has just made available on
its Autodesk Labs Web site. Its a tool an in-
terior designer might use to show a kitchen
and change material nishes interactively
so the customer can see what a wood oor
might look like versus tile, or white cabinets
versus black. Tere are processors out in the
cloud beavering away at processing while
the desktop client updates relatively quickly.
Autodesk describes it as borrowing.
Autodesk suggests that users could rent the
use of additional processors as they need it.
Both Kowalski and Bass in a later interview
made it clear that this isnt a new business for
Autodesk; rather, Auto desk would team up
with a company that is oering serverfarm
access. Amazon, for instance, is developing
The DesignSlam gave designers 20 minutes to produce a concept while rock music blares, the
audience shouts, and judges mill around. Here, the challenge was to rethink the watch using
AutoSketch and/or Studio. Brothers Rich and Glenn Walters of Brooks Stevens Design seemed to
totally enjoy the experience.

CAD
c H1 R f OR bl Gl 1AL
l MAGl HG AR1 5
A1 O51OH UHl VR5l 1 Y
ROGRAM5
3b AHl MA1l OH
l H1RAc1l V
Mbl A
GAM AR1
cHARAc1R
AHl MA1l OH
www. cbl AU. cOM
wAL1HAM, MA wA5HlHG1OH, bc

u
a
g

u
a
r

u
u
c
a

r
u
r
e
CGW_THIRD_vert:CGW_THIRD_vert 10/24/08
January 2009 39
cloud computing as a lucrative hobby for
all its server processors. Google, too, has a
bunch of servers that conceivably have some
spare time. Te idea is that companies in
the information business are likely to have
to build out huge server capacity for peak
times, but that capacity might not be used
to the fullest. Tere are server cores available
for other work.
If you think about it, human proces-
sorsour little or big brains, as the case
may beare subject to the same challenge
as semiconductors, both CPUs and GPUs.
Tere reaches a point at which one just
cant do it all by thinking harder or work-
ing faster. Semiconductors hit the wall, and
we get smoke coming out our ears. So the
answer in the case of semiconductors is to
gang up processors to do a task, and its the
same for uswell gang up semiconduc-
tors to try out more ideas for us, or we can
gang up brains for better collaboration.
The Next Generation
Tis is a trend happening throughout the
digital content creation world as the industry
tries to make the shift from tools that force us
to adapt to them, to tools that adapt to our
tasks and actually expand our capabilities. In
a way, the rst step was really just getting the
process of design and documentation into
the computer. Now, were ready for more.
Bass said two things during Autodesk
University that pointed to the ways in which
the company would like to change the pro-
cess of design. He said, I dont understand
why a $50 game should look better than a
design product costing thousands of dol-
lars. And, talking about interfaces, he said
the napkin just accepts the pencil. Much
of the work going on in the labs is to try and
make the software attractive and enticing to
play with, and the process as simple as trying
out ideas with a pencil and paper.
Along those same lines, Autodesk has ac-
quired the aptly named Mudbox, an interac-
tive modeling tool that lets users push, pull,
carve, and cut a model into shape, just like a
sculptor might do. Well, Autodesk, being the
company it has become, isnt likely to reserve
that technology for entertainment content
creation when it could be so useful for 3D
design and surface design. Demonstrations
of Mudbox at the Auto desk Labs booth on
the exhibit oor showed the software paired
with an HP multi-touch display so that the
user could easily pull and push a model into
place using both hands. In addition, Au-
todesk prompts the following: Why couldnt
multi-touch be used to make navigating in
3D with tools, like Inventor, much easier.
And all this discussion seems to leave Auto-
CAD, Autodesks stalwart CAD tool that
is the most widely used CAD tool in the
world, lying neglected in the dust. Its not,
so were told. Autodesk wants to push Auto-
CAD forward into the new millennium, to
use all the ideas being kicked around, such
as freeform design, dynamic interaction, and
iterative design, to make it a more useful tool
for the rst step in getting into a buildable,
or manufacturable, form. In a casual conver-
sation, Luc Robert from the RealViz team,
said, Our job now is to make AutoCAD
cool again. Now that might be a big job, but
it is kind of job one for Autodesk.
As Autodesk University so clearly dem-
onstrated, the companys real power base is
founded on those 10,000 or so people ex-
citedly chattering in the halls at AU. Tey
lled the classrooms where new techniques
were taught and new products were dem-
onstrated. Teyre hungry for tools that can
help them do more. Tis year, Autodesk
seemed to be saying that its not a matter of
moving people o AutoCAD and onto new
platforms, but rather to move AutoCAD
forward so its a vehicle allowing its users
to do more and, along the way, incorporate
more tools into their tool set.
Kathleen Maher is a contributing editor to CGW, a
senior analyst at Jon Peddie Research, a Tiburon, CA-
based consultancy specializing in graphics and multime-
dia, and editor in chief of JPRs TechWatch. She can be
reached at Kathleen@jonpeddie.com.
w
w
w
.
f
o
c
a
l
p
r
e
s
s
.
c
o
m
focal press
b o o k s
F O R C R E A T I V E S,
B Y C R E A T I V E S
G
a
m
e

A
r
t

C
o
m
p
l
e
t
e
E
d
i
t
o
r
:

A
n
d
r
e
w

G
a
h
a
n
I
S
B
N
:

9
7
8
0
2
4
0
8
1
1
4
7
5
$
6
9
.
9
5
T
h
e
C
o
m
p
le
t
e
G
u
id
e
t
o
G
a
m
e
A
u
d
io
, 2
e
B
y
A
a
r
o
n
M
a
r
k
s
I
S
B
N
:
9
7
8
0
2
4
0
8
1
0
7
4
4
$
3
9
.9
5
D
i
g
i
t
a
l

C
h
a
r
a
c
t
e
r

D
e
v
e
l
o
p
m
e
n
t
B
y

R
o
b

O

N
e
i
l
l
I
S
B
N
:

9
7
8
0
1
2
3
7
2
5
6
1
5
$
6
9
.
9
5
D
i
r
e
c
t
i
n
g

t
h
e

S
t
o
r
y
B
y

F
r
a
n
c
i
s

G
l
e
b
a
s
I
S
B
N
:

9
7
8
0
2
4
0
8
1
0
7
6
8
$
3
9
.
9
5
Visit www.focalpress.com!
View video tutorials, chapter samples,
podcasts and more. Sign up for
specialized e-newsletters with
tips and specials throughout the year.
| e a r r m a s t e r c r e a t e
Focal Press helps
youmaintain your
competitive edge.
Buy online
or in your favorite
bookstore today!










Y I YY T A E R C Y
R
B
RR I T A E R C R
o
OOO F
oo k o oooo bbbb
c r p c p l ca c o f




EEE S E VVV
, S E SS EE
ss
V
s
I
k
sss ss eee r


































t t u o gh u o pecials thr d s s an tip
s w er tt wsle e ed e-n z peciali s
p gn u . Si e r o d m ts an s podca
er s t p a h , c ials r o t eo tu w vid ie V
m .co s s e r ocalp .f www isit V





m
o
c
.
s
s
e
r
. e year h
with
r o p f
, s le p am s
! m





e r o tto ks boo
ur f o r in y yo o
n uy o B
v titi pe mmp co
a tta ain um o yyo
s ee l Pr re Foca
e r c r t s a r m e a r |





y!
p
l
a
c
o
f
.
w
w
w
aay od da e tto
e itte rri o av ffa
e nlin
. ee. gge e ed dg vve
ur o ain y yo
s s help
e t e a r
CGW_THIRD_vert:CGW_THIRD_vert 12/23/08
40 January 2009
Most people would not classify the 3D imagery found in a
video game or a lm as art. Perhaps its because they are
looking at moving sequences rather than still frames. Yet,
if folks had the chance to see the beautiful sketches that
form the basis for the animated imagery, they would indeed
agree that the term art is tting.
A few months ago, The Art Institute of CaliforniaSan
Diego, in collaboration with High Moon Studios, presented
a gallery exhibition of artwork from Robert Ludlums The
Bourne Conspiracy video game. The third-person action
title features a blend of hunter-prey activity with dramatic
escapes, all designed around the Jason Bourne characters
signature combat style, popularized in the trio of feature
lms and described in the best-selling novels.
The Art Institute had a special interest in the game:
Several of its alumni assumed a role in the development
of the game.
Given the success of the Robert Ludlum trilogy and
the popularity of the Game Art & Design and Media Arts
& Animation program, The Art Institute of CaliforniaSan
Diego was thrilled to showcase the talent of its alumni
and faculty member who took part in the games evolu-
tion, says Jody Auslund, public relations/communication
coordinator at the school. The focus of the exhibit was
to showcase the talent of our alumni and the caliber of
student coming out of our programs.
This is the rst time the school has hosted such an exhibit,
though its graduates have worked on numerous popular
games in the past, including Guitar Hero World Tour for the
Wii, Midnight Club Los Angeles, and EverQuest II.
The eight students who worked on the Bourne game
include: Charles Bradbury, environment artist; Tyler
Wanlass, associate prop artist; Don Ott, associate prop
artist; Jason Copeland, prop artist; Carlos Dominguez,
technical artist; Jess Morris, cinematic animator; Chad
Campbell, associate character setup artist intern; and
Michael Vincent Castro, associate game designer. Alan
OBrian, a former Game Art & Design instructor at the
school, assumed the role of animator.
All the images showcased at the exhibition were crafted
in Adobes Photoshop CS2, and each image was created to
provide a specic mood.
I get to play for a living. As someone who transitioned
out of the engineering industry into games, its a dream
come true, says Dominguez. Karen Moltenbrey
41 January 2009
The Art of
the Game
These two pages contain concept art from the video game
Robert Ludlums The Bourne Conspiracy, created by the
artists at High Moon Studios. Several alumni and a former
instructor from The Art Institute of CaliforniaSan Diego
held roles in the games production.
This is the type of title that just about every Game Art &
Design student strives to be a part of in their professional
career; its great to see recent grads, some less than
a year out of school, participating in a project of such
high caliber. We are extremely proud of our graduates,
and its amazing to see their hard work come to life in a
game such as this, says Jean Branan, director of career
services.
42 January 2009
I
ts been often stated in the pages of this magazine and elsewhere
that the VFX, animation, and post industries are a people busi-
ness. Thats why recruitment is so importantits the recruit-
ers job to secure the human talent that is so key to a companys
ongoing success.
So what happens when your hard-fought-for talent ups and
leaves after making magic on a major motion picture? How do you
fill that empty seat with equal or greater talent? How do you keep
tabs on someone youd like to bring back when more work is avail-
able? And, how do you deal with an economic environment that
portends less production, at lower budgets and tighter deadlines?
You ask the recruitment experts. And we did just that. Four re-
cruitment executives from top companies in the US, Canada, and
the UK offer their take on the strengths, weaknesses, opportunities,
and threats they perceive that may help or hinder their efforts as
they proceed into the new year.
Dominic Davenport
CEO/Founder
Escape Studios
London/New York
www.escapestudios.com
Escape Studios has multiple missions. The company, a reseller of post
and visualization tools, also provides the industry with animation and
VFX services. But, as founder Dominic Davenport says, Our mainstay
is to train and educate, and that includes job placement.
Strengths: One strength of recruitment in this industry is the
diversity of talent. The majority of our business is repeat business
based on clients finding our people incredibly useful and rebook-
ing or asking for more experienced or specific roles to complete the
task. Visualization customers want to adopt a more post or filmic
pipeline, and theyre using more high-end tools and delivering the
type of images we see in movies and high-end animated features
today, to sell a new city in the Middle East or in Singapore.
Weaknesses: For the recruitment industry, the main weak-
nesses are going to be finding people, especially for postproduction,
who have the specialist skills necessary to deliver high-end content
at a time when budgets are being cut. I think theres going to be a
crisis in salaries, at least at the high end. As the level of complexity
expected by the public continues to rise, their eye becomes more re-
fined, and youre just going to have to throw more people at projects
to deliver the same level of quality in shorter amounts of time. With
less money around, either a production moves to a locality where
a workforce is cheaper or rates have to be cut. In the post industry
over here [in the UK], Ive heard rumblings of certain companies
cutting rates across the board.
Opportunities: There is opportunity for talent to change fo-
cuslike getting involved in projects using their skills for Web activ-
People Persons
Finding the best employees for the job
BY KEN MCGORRY
Escape Studios Dominic Davenport: Talent may also look beyond the
entertainment industry to CG-oriented visualization.
Dont Break the Bank!
Game Developer Tools and Technology
Workstations / Monitors / Input devices
Studio Storage
Outsourcing Studio Services
MARCH:
NOVEMBER:
AUGUST:
MAY:
Sponsored white papers, case studies, and technical articles that
will help you finish your projects on time and under budget!
Stay informed with upcoming CGWSpecial Reports
packed with recession-fighting tips
from top industry suppliers.
CGW Middleware:1208_p 12/30/08 3:51 PM Page 1
44 January 2009
nnnn
Knowledge & Career
ity. With the boom in the
Internet as a marketing ve-
hicle, along with the move
away from the traditional
30-second commercials,
there is a hybrid activity
that will evolve which uses
all of the high-end skills
for much lowlier output.
For the [boutiques], there
will be a great deal of work
flying their way. And with
the global economy in the
state its in, theres oppor-
tunity for people, having
acquired certain skills, to
be useful not just to postproduction, or the games industry, or the
visualization industry, but across them all.
Threats: Its hard to imagine, but I see a potential slowdown
in the media industry, with less content being produced in general.
Theres always a sorting of the wheat from the chaff in terms of the
level of talent thats available. The entertainment industry has done
well in times of economic strife, but I do think that less stuff is go-
ing to be produced.
Outlook for 2009: Its going to be a challenging year, but then
again, when isnt it? Weve all got to keep the levels of quality and
creativity high and try and enjoy what were doing. Otherwise,
whats the point?
Dennis Hoffman
Senior vice president and general manager
CIS Vancouver
Vancouver, Canada
www.cis-vancouver.com
Last January, Deluxe acquired Rainmakers visual effects and postpro-
duction facilities in London and Vancouver. Renamed CIS, the shops
joined Hollywoods CIS studio. Hoffman oversees recruiting for the
Vancouver location.
Strengths: Finding strong leadership and talented people in
all aspects of visual effects is a requirement for us to grow ourselves,
our tool sets, and our stature within the VFX community. We look
in the local community school systems, which are very good, for
growing artists. We look more to the US and London for seasoned
artists, to help grow the whole facilityand theres New Zealand
and Australia, as well. Higher-end artists learn of CIS Vancouver
via word of mouth, advertising, and our knowing where people are
and when their projects are coming to an end, and then actively
nurturing those relationships.
Weaknesses: One challenge is the artists interest in moving
to a new location. As VFX has matured and grown more global,
the artist base is much more open to going to different places. One
trap when you are looking at somebodys reel is ascertaining what
he or she really was responsible for on the reel. Also, as schedules
get tighter and tighter, you have to make faster and faster decisions.
You cant go through a long process. If we want to meet face to face,
with London and Los Angeles, one of the weaknesses is not being
able to sit across the table from somebody.
Opportunities: With VFX, the education process never
stopsmore mature artists help teach the young artists. The
work can go globalyou can be considered for stuff you wouldnt
have been [considered for] in the past. Its an opportunity for Van-
couver. And were in the same time zone as Hollywoodstudio
people like the idea of not having to coordinate a 3 am phone
call. Moreover, we have relationships with other facilities in other
countries for simpler things, like roto and matchmoving, and this
exposes us to other people who we may then be able to bring over
to our facility.
Threats: The opportunity is that were a global industry, and
the threat is that were global. You either ignore itin which case
it truly becomes a threator you figure out how to make it part
of how you run your business. We work in a creative environment,
so the recruiting of people is the key factor in the longevity and
strength that you build within your facility. The success of that goes
back to how effective you are in recruiting the right people.
Outlook for 2009: The year looks to be filled with lots of op-
portunities. I anticipate the first half to be slower than normal due
to the delay at the end of this summer because of the anticipated
SAG strike. However, we are currently seeing an impressive increase
in bidding for features and television, and we project that VFX work
will be robust from April through the end of the year.
CIS Vancouvers Dennis Hoffman: As schedules get tighter and tighter, you have to make faster and faster decisions.
January 2009 45

Knowledge & Career


nnnn
Jon Lewis
Co-founder
ICA Creative and WhosCreative
Fairfield, CT
www.icacreative.net, www.whoscreative.com
ICA Creative specializes in locating, evaluating, and placing creative
talent in the post and broadcast industries. The Web site WhosCreative
provides job seekers with online tools for marketing their talent on the
Internet.
Strengths: Studios should use several methods to find
talent. These days, everyone has a Web site. However, finding an
individual site can be nearly impossible. Companies like Whos-
Creative.com showcase VFX artists work. Using these sites, you
can pinpoint exactly the type of artist youre looking for and then
quickly view their demo reel or storyboards. However, the most
effective method for finding good artists is building relationships
with talent you already have. They can refer like-minded, talented
artists you desperately need.
Weaknesses: I see a weakness in attracting new talent using
free jobs boards. Lets say you are in need of a 3D Maya generalist
for a freelance in-house job starting next week. By placing an ad on-
line, you are subject to hundreds of responses, all with varying levels
of skill. Your work is disrupted because you have to wade through
the hundreds of responses to find the right person, confirm they are
local to your area and have the right skills, check their reel (make
sure the work on their reel is actually theirs), and then match per-
sonality. Working with a recruiter, such as ICA Creative, is more
cost- and time-effectiveits their job to qualify, evaluate, manage,
and place excellent artists.
Opportunities: When youre searching for new talent, you cre-
ate a wide web of opportunity. While constantly opening doors,
you may run across someone who isnt right for the project today,
but has potential for other jobs tomorrow. By ignoring that person,
you will lose out on an up-and-comer who could be perfect a year
down the road. Managing green talent is just as important as find-
ing new talent.
Threats: Sometimes during a down economy like this, a studio
will have to let someone go. Or perhaps an internal conflict results
in a valuable artist leaving the company. What a studio needs to
keep in mind is that during that artists tenure, he or she built rela-
tionships with clients. Those relationships have an immediate im-
pact on the companys current and future business. It is important
to nurture existing talent and to build a loyal relationship so that
work doesnt go elsewhere.
Outlook for 2009: Opportunity will always exist in a down
market, and this will be proven true in 2009. Budgets will con-
CGW_Half_HOR:CGW_Half_HOR 9/15/08 10:17 AM Page 1
46 January 2009
nnnn
Knowledge & Career
Digital Domains Lala Gavgavian: The directors
we work with and the shows themselves are a
big draw.
tinue to get bruised and battered, more so
in 2009. There will be much more com-
petition to get the gig (less work out there,
way more people looking), and there will be
much more pressure for artists to come off
their rate.
Lala Gavgavian
Director of recruiting
Digital Domain
Venice, CA
www.digitaldomain.com
Digital Domain, founded in 1993, is known
for its Oscar-winning VFX for films as well
technical and artistic innovations in commer-
cial and video game production.
Strengths: A fairly consistent strength
we enjoy is that Digital Domains projects
typically include groundbreaking visual ef-
fects, and the challenge of creating them
attracts great talent. The shows themselves
and the directors we work with are another
big draw. We also benefit from the high-end
tools and technology that we use, as they at-
tract artists looking to further their skills in
particular areas.
Weaknesses: Because our core offer-
ing is a service business, we usually arent
able to forecast our needs out much more
than six months at a time. This means we
arent always able to commit to great tal-
ent when its available. Also, it is becoming
common for studios to farm out a single
project to multiple companies, so when
shows ramp up, many studios are looking
for the same type of talent at the same time.
This cycle can create challenges in securing
particular disciplines that are in demand.
Opportunities: Our recruitment ef-
fort absolutely leads to discovering new
stars of VFX. Internships in particular
are a great opportunity for discovering and
fostering talent. Its fantastic to see interns
turn into superstars as they gain more pro-
duction experience.
Threats: Digital Domain has been for-
tunate in that weve had a full slate in 2008
and have been able to keep folks busy. How-
ever, the ebbs and flows of the VFX busi-
ness sometimes dictate a ramp down, usu-
ally through the summer months, which
does create a situation where we may have
to let some talented people go. For all of us
in the recruiting side of visual effects, the
risk is that we wont be able to rehire talent
when we need them, as they may commit to
another project elsewhere. That is a reality
that just comes with the territory.
Outlook for 2009: I think the demand
were starting to see now for hybrid artists
who can cross visual effects, games, and ani-
mated features will only grow in the coming
year as studios look to cross-utilize talent
more and more. The good news for artists is
that they have more freedom and opportu-
nities to work on a wider range of projects.
The difficulty for recruiters in visual effects
is that we now face even broader competition
for talent, not only from other VFX studios,
but from the games and animation indus-
tries as well. n
Ken McGorry is a consulting editor for Post maga-
zine, CGWs sister publication. He can be reached at
mcgorry@optonline.net.
Get your FREE 1-year subscription to Computer Graphics World.
What are you waiting for?
www.cgw.com November 2008
$6.00 USA $8.25 Canada
D
o
g
g
e
d
P
u
r
s
u
it
Disneys R&D develops new
tools and techniques for the
loose brush style in Bolt
www.cgw.com October 2008
$6.00 USA $8.25 Canada
The artists and
the Madagascar
zoosters encounter
new challenges on
the savannah
In Plain
Sight
www.cgw.com September 2008
T
h
e

A
p
p
r
e
n
t
ic
e
LucasArts unleashes
a new force in
computer gaming
$6.00 USA $8.25 Canada
Subscribe online by logging on to www.cgw.com, select Contact Us from
the navigation bar, then Subscription Customer Services.
Subscribe today!
Available in print & digital formats.

CGW:1108_p 11/24/08 3:50 PM Page 1
VIDEO/VFX
Enhanced Effects
Autodesk announced extensions for the
Autodesk Smoke 2009 nishing effects
system and the 2009 versions of Autodesk
Inferno, Autodesk Flame, and Autodesk
Flint visual effects systems. Now avail-
able to customers with a subscription, the
new software extensions provide many
new features, such as an enhanced oat-
ing point tool set, support for additional
compressed formats, improved interoper-
ability with the Autodesk Lustre, Smoke
editorial workflow enhancements, real-
time deliverables for mastering to multiple
formats without rendering, and more than
80 new particle system presets for Inferno,
Flame, and Flint.
The company also introduced the
Autodesk Lustre 2009 digital color-grad-
ing system and Autodesk Incinerator
2009, a real-time system accelerator
for Lustre. Lustre 2009 is designed to
improve the color-grading workow with
its stereoscopic grading, project manage-
ment, multilayer timeline shot prioritization,
customizable sensitivity settings for the
Autodesk Control Surface, and improved
interoperability with the Flame, Inferno,
Flint, and Smoke visual effects and nish-
ing systems. Incinerator 2009 harnesses
parallel processing, server, and networking
technologies to provide real-time color
grading for demanding digital-intermedi-
ate, feature-lm projects. Autodesk Lustre
2009 is now available for Red Hat Enter-
prise Linux and Microsoft Windows XP
operating systems, whereas Incinerator
2009 is shipping for Linux systems.
Autodesk; www.autodesk.com
TEXTURING
Luxology has upgraded its ImageSynth
image processing tool to Version 2. Image-
Synth 2, infused with up to 10 times
faster performance, enables artists to
process images into seamless textures,
synthetic images, and backgrounds, as
well as generate high-resolution images
from low-resolution sources. Available in
a stand-alone version and as a plug-in
for Adobe Creative Suite 4, ImageSynth
is especially useful when working with
noisy images, such as those depicting
crowds or beach pebbles, as examples.
The tool creates royalty-free images that
can be tiled together seamlessly to create
backgrounds, brushes, or texture maps.
Version 2 boasts a new algorithm threaded
for use on dual-core systems, improved
memory management to handle larger
scenes, and new control over chunk size
and orientation. ImageSynth 2 is now avail-
able for $99 to new users and as a $49
upgrade to existing ImageSynth users.
ImageSynth works with Adobe Photoshop
CS3 or higher.
Luxology; www.luxology.com
DATA TRANSLATION
Okino Computer Graphics, a provider of
3D data translation solutions, is shipping its
third-generation, solids-based Catia V4+V5
R19 and ACIS SAT CAD importers. Based
on components from Dassault Systmes
and Spatial, the importers provide accurate,
high-quality import conversions because
they share the same runtime code used
by Catia V5 modeling software and the
Spatial ACIS modeling engine. Okinos
CAD data-optimization routines enable
complex assemblies to be imported, opti-
mized, reduced, composed, and loaded
into 3D packages, such as Autodesks
3ds Max and Maya, NewTeks LightWave,
Softimage XSI (now owned by Autodesk),
Maxons Cinema 4D, Adobes Director,
Googles SketchUp, and others. The Catia
and ACIS CAD importers provide various
data-import options, extensive tessellation
controls, model healing and erroneous
surface repairs, and selective layer import
for Catia V4 and V5 les.
The company also announced that a
secondary data-conversion pipeline in
its PolyTrans CAD enables STK users to
evaluate the new Collada model import
capability of AGIs STK software. Okinos
PolyTrans CAD converts native CAD le
formats, such as ACIS SAT, Alias .wire,
Autodesk Inventor, AutoCAD DXF/DWG,
Catia V4+5, DWF, IGES, JT Open, Para-
solid, Pro/Engineer, Rhino-3D, Solid Edge,
SolidWorks, STEP, STL, XGL, U3D, VDA-
FS, XGL, and X3D. Users gain control over
nal scene complexity, as well as tools for
small parts removal, robust hierarchy/scene
optimization, and polygon reduction.
Okino Computer Graphics; www.okino.com
For additional product news
and information, visit CGW.com
48 January 2009
WIN MAC
SOFTWARE
MAC
January 2009, Volume 32, Number 1: COMPUTER GRAPHICS WORLD (USPS 665-250) (ISSN-0271-4159) is published monthly (12 issues) by COP Communica-
tions, Inc. Corporate ofces: 620 West Elk Avenue, Glendale, CA 91204, Tel: 818-291-1100; FAX: 818-291-1190; Web Address: info@copprints.com. Periodicals
postage paid at Glendale, CA, 91205 & additional mailing ofces. COMPUTER GRAPHICS WORLD is distributed worldwide. Annual subscription prices are $72,
USA; $98, Canada & Mexico; $150 International airfreight. To order subscriptions, call 847-559-7310.
2009 CGW by COP Communications, Inc. All rights reserved. No material may be reprinted without permission. Authorization to photocopy items for internal
or personal use, or the internal or personal use of specic clients, is granted by Computer Graphics World, ISSN-0271-4159, provided that the appropriate fee
is paid directly to Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923 USA 508-750-8400. Prior to photocopying items for educational
classroom use, please contact Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923 USA 508-750-8400. For further information check
Copyright Clearance Center Inc. online at: www.copyright.com. The COMPUTER GRAPHICS WORLD fee code for users of the Transactional Reporting Services is
0271-4159/96 $1.00 + .35.
POSTMASTER: Send change of address form to Computer Graphics World, P.O. Box 3296, Northbrook, IL 60065-3296.
M
o
d
e
l

b
y

D
e
a
n

A
m
i
r

D
e
p
a
y
,

I
D
E
,

I
n
c
.
WIN LINUX
WIN MAC
WIN
2009 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, ATI, the ATI logo, FirePro and combinations thereof,
are trademarks of Advanced Micro Devices, Inc. All other brand names, product names, or trademarks belong to their respective holders.
Images courtesy of Will Burdett at the University of Hertfordshire, PTC, Youngwoong Jang, and Todd Daniele
Convergence of pure performance and your vision
You have a vision. ATI FirePro

workstation graphic accelerators give you


the power to help realize it. The robust portfolio of graphics accelerators
delivers leading performance and reliability across the broadest spectrum
of engineering, scientifc and DCC applications and operating systems. ATI
FirePros display engine produces over a billion colors on 10 bit displays
and delivers the power needed to manipulate huge data sets and support
large displays.
Go to ati.amd.com/FirePro to learn more about ATI FirePro

family of
products.
Dominate your design.













Do


omina


ate you


ur desi


gn.



























s displ o r eP ir F
in ineer of eng
ers leadi deliv
o er t w the po
e a vi v ou ha Y
Convergen


er v es o oduc ine pr y eng la
C app tifc and DC , scien ng
eli r e and manc or f ing per
obus he r T . e it ealiz help r
kst or w

o r eP ir TI F AAT sion.
nce of pure performa


olors on 10 bit r a billion c
ting s a tions and oper plica
oade oss the br acr y iabilit
aphics a r olio of g tf st por
o t a eler c aphic ac r tion g ta
ance and your vision


s y displa
TI AAT . ems st sy
trum est spec
ors t a eler c c
ou e y iv ors g
n




an dv 2009 A


. ed v eser ts r igh ll r . A nc , I es evic o D icr ed M nc
. ts oduc pr
ti.amd a o o t G
s y ge displa lar
ers t and deliv
s displ o r eP ir F


TI log TI, the A , A wlogo o r r MD A , the A MD . A
n m o lear t o r eP ir om/F .c d
. s
o man er needed t w the po
er v es o oduc ine pr y eng la


, eof f, tions ther ombina o and c r eP ir , F go
fa

o r eP ir TI F AAT e about or
ta sets a e huge da t nipula
olors on 10 bit r a billion c

amily of
t nd suppor
s y displa


t our mages c I
k ademar e tr ar
an dv 2009 A


y of He ersit t the Univ dett a ill Bur W esy of
ll other . A nc , I es evic o D icr ed M anc dv ks of A
. ed v eser ts r igh ll r . A nc , I es evic o D icr ed M nc


, and oong Jang oungw Y C, , PT e dshir or tf er
adema , or tr t names oduc , pr and names r br
TI log TI, the A , A w logo o r r MD A , the A MD . A


odd Daniele T
. e holders tiv espec o their r ks belong t r
, eof f, tions ther ombina o and c r eP ir , F go




CGW :109_p 12/23/08 10:13 AM Page 1
BES T VI S UAL EFFECTS
2008 Paramount Pictures. All Rights Reserved.
PARAMOUNTGUILDS.COM
His name was
Captain Mike Clark
Hed been on a tugboat
since he was seven.
CGW :109_p 12/23/08 10:12 AM Page 1

You might also like