Make your design unforgettable with

ATI FirePro 3D graphic accelerators.

3d Studio Max 9 OGL SPECapc SM Comparison

Give yourself the power and speed to take projects to the next level. To bring your imagination to life. To finalize jobs before
their drop-dead dates. To make the competition green with envy. After all, ATI FirePro 3D graphic accelerators detect,
optimize, and tune applications for superior performance. Plus, they give you rock-solid reliability for extreme productivity.
What does that mean for your final digital content? Images that make people stop, stare, and stare again. And for you?
A digital design force to be reckoned with.
The time has come. Put ATI FirePro 3D graphic accelerators to work for you.
ATI FirePro

Nvidia Quadro

V7750 vs FX3800

V5700 vs FX 1800

V3700 vs FX 380

Learn
more at www.amd.com/atifirepro/dcc
Image courtesy of Anton Bugaev.

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. Other names are for informational purposes only and may be trademarks of their respective owners. 47056A

47056A_Firepro_Ad_CGW_June09_HHOG_FP.indd 1

6/3/2009 4:34:18 PM

June 2009 Volume 32 Number 6

Innovations in visual computing for the global DCC community

Features
COVER STORY

12

The Shape of Animation

shapes reflect the artistic direction and story line for Disney/
Pixars Up, the first stereoscopic 3D film from Pixar.
12Geometric

By Barbara Robertson

Living Art

& Hues creates some lively effects for Night at the Museum: Battle
the Smithsonian, bringing art and artifacts to animated life.
22ofRhythm

26

By Barbara Robertson

Its a Mad, Mad, Mad, MadWorld

22

32

Departments
Editors Note

Courting Controversy

artistic freedom at risk in the gaming industry, particularly with recent

2Isgovernment
and special-interest group involvement?

Spotlight

technology. The Foundrys Nuke 5.2. Eyeons SWAT

4andcertifiSLIcationMulti-OS
program. Blackmagics routers, converters, and other offerings.
Products Nvidias Quadro FX 5800, 4800, and 3800, and NVS 295,

Havoks Havok AI. Caustic Graphics CausticRT. Autodesks Flare and other
offerings. Allegorithmics Substance Air. News Outsourcing in the
gaming industry is on the rise.

Viewpoint

10Gimbal lock in the CG industry.

Careers

45How animators can grow their skills.

group of distinguished game developers form a new company and create

26AMadWorld
, a title with an unusual black-and-white, graphic-novel style.

By Martin McEachern

A Bit of Difference

When it comes to imaging, is 10-bit that much better than 8-bit? A growing
of users think so, and are willing to pay the price to have it.
32number

By Alex Herrera

Elemental Effects

create some devilishly difficult scenes, including soaring

sets and large crowds, for the live-action film Angels & Demons.
38VFXvirtualstudios

By Barbara Robertson

Adding POW! to Prime Time

on a tight television production schedule, Mechnology gives a

new team of superheroes their powers in the now-defunct TV series
42Working
The Middleman.

By Debra Kaufman

SEE IT IN

What makes the Hannah Montana

show work?
Red shoots on the rise.
Trends in broadcast design.
J.J. Abrams on Star Trek.

Back Products

47Recent software and hardware releases.

Whats on your mind?
Have a question about a product? Curious about how an effect was done? Want to
get peer opinion about something you are working on? Or, do you just want to voice
your opinion about something, anything, pertaining to the digital content creation
industry? If so, start a string in the new Forum section on cgw.com.

ON THE COVER

Pixar set the bar high for CG animated feature films when
it released Toy Story in 1995. Fourteen years and nine films
later, the studio, in conjunction with Disney, once again has
carried computer animation to new heights with the spectacular film Up, this time releasing it in stereo 3D, pg. 12.
June 2009

GuestEditorsNote

Courting Controversy

mmediately upon its release, the graphic-novel styled video game MadWorld (see Its a
Mad, Mad, Mad, MadWorld, pg. 26) became a lightning rod for the outrage of media
watchdog groups the world over. John Beyer, director of the UK conservative specialinterest pressure group mediawatch-uk, was quick to vocalize his disgust over the games
content, even urging the BBFC (British Board of Film Classification) to deny the game a
rating, which would have effectively banned it from being sold. On March 10 of this year,
the day the game was released, the National Institute on Media and the Family issued a
press release lashing out at Nintendo for tarnishing the familyoriented image of the Wii by opening its doors to the violent
video game genre. (The game is played on the Wii platform.)
The games publisher, Sega, already buckling under pressure, announced late last summer that it would not release the game in
Germany, fearing a backlash from local media watchdog groups
and an ill-informed public there.
The game is, without question, graphically violent. Literally.
All the action centers on killing and dismembering, though it
is done in a black-and-white comic-book style, with red accent:
blood. The moral outrage, however, almost seems laughable con- Contributing editor
sidering the violence plays out with all the realism of a Wile E. Martin McEachern
Coyote cartoon. Nonetheless, these media watchdog groups, hyper-sensitized to game violence, are a threat to artists, their livelihood, and the artistic maturity of the medium, especially if developers and publishers concede to their wishes in fear
of stoking their ire. They have the power to foment public opinion, influence mainstream
media, and pressure ratings boards, such as the ESRB and the BBFC, to strip games of
their classifications, essentially crippling their marketability and distribution.
Even worse, with the economy fighting its way out of the recession, Bloomberg.com is
reporting that Activision Blizzard, the worlds most powerful publisher reportedly with
$3 billion in cash and no debt after the merger with Vivendi, is now looking to gobble up
smaller developers as cheaply as possible. This follows in the footsteps of another gaming
giant, EA, which has in recent years acquired Mythic Entertainment, Phenomic Game
Development, Hands-On Mobile, Digital Illusions CE, and Headgate Studios, while still
holding a 15 percent controlling interest in Ubisoft. With the industrys wealth and power
consolidating in the hands of a few gaming conglomerates, the opportunities for creativity
and innovation could only dwindle under these pressures. What could result is a gaming
industry that looks much like the film industry, where media conglomerates crank out
safe, generic, homogenized art, carefully assembled in a boardroom, where every creative
decision is driven by market demands and the need for maximum profitability. Needless to
say, anything that would threaten a market-friendly classification would be avoided.
We understand there are differences among the ethical sensibilities of each country,
says Atsushi Inaba, one of the four famed developers who founded Platinum Games, creator of MadWorld. However, we believe that adults are capable of distinguishing fantasy
from reality, and understanding that a game is a completely virtual experience. And only in
this virtual world are we given opportunities to do things that are not acceptable in the real
world. At the same time, adults have to communicate to their children that a game world is
totally different from the real world. The media and [these watchdog groups] have to think
about this. Yes, MadWorld is a violent action game, but the violence is not gratuitous; its
not like you can do anything you want to do. In fact, if you take the time to play to the end
of the game, the story line delivers a very strong anti-violence message, so we just want the
media to understand and report the contents of the game accurately.
Inaba remains vehement in justifying the games violence and protesting the media castigations of his art. Whether MadWorlds story delivers an anti-violence message or not is
continued on page 48
ultimately irrelevant to the majority of game artists. For centuries,
2

June
2009
August
2008

The Magazine for Digital Content Professionals

E D I TO R I A L

KAren moltenbrey
Chief editor

[email protected] (603) 432-7568

36 east nashua road
Windham, nH 03087

Contributing Editors

Courtney Howard, Jenny Donelan,

Audrey Doyle, George maestri,
Kathleen maher, martin mceachern,
barbara robertson

WIllIAm r. rIttWAGe

Publisher, President and Ceo,

CoP Communications

SA L E S
lIsA blACK

national sales manager

Classifieds education recruitment
[email protected] (903) 295-3699
fax: (214) 260-1127

Kelly ryAn

Classifieds and reprints

[email protected]
(818) 291-1155

editorial office / lA sales office:

620 West elk Avenue, Glendale, CA 91204

(800) 280-6446

PRODucTIOn
KeItH KnoPF

Production Director
Knopf bay Productions
[email protected] (818) 291-1158

mICHAel VIGGIAno
Art Director

[email protected]

CHrIs sAlCIDo

Account representative

[email protected] (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 [email protected]. 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

LightWave 3D

New Quadro FX Offerings

Nvidia recently introduced its most advanced top-to-bottom
line of Quadro professional GPU solutions in the companys
history, including the flagship Quadro FX 5800.
Spanning from sub-$100 entry-level to stand-alone visual
computing systems, this series includes: the 4GB, ultra highend Quadro FX 5800 ($3499) and FX 4800 ($1999); the
FX 3800 ($1199) at the high end; the mid-range FX 1800
($699); the entry-level FX 580 ($199) and FX 380 ($149),
and the NVS 295 ($159) business solution. The Quadro
cards are available through system manufacturers such as
Dell, HP, and Lenovo, and Nvidia channel partners.
Separately, Nvidia introduced its SLI Multi-OS, which
allows users and visualization applications, for the first time,
to take full advantage of multiple Nvidia Quadro GPUs from
a single graphics workstation in a virtualized environment.
Built into the new Quadro FX 3800, 4800, and 5800,
SLI Multi-OS allows users to tap into the advanced visualization and compute capabilities of the Quadro GPUs
for full graphics performance within a virtualized system.
SLI Multi-OS works with Parallels Workstation Extremes
virtualization software and Intels VT-d technology, assigning both the host and the guest virtual machine their own
dedicated GPU.
Available in the new HP Z800 workstation, these technologies deliver application performance nearly identical
to systems configured with a dedicated OS and GPU. The
SLI Multi-OS supports selected combinations of Windows
XP, Windows Vista, and Linux operating systems.

Eyeon Forms SWAT Teams

The Foundry announced an upgrade version of its Nuke compositing software. Nuke 5.2 features new pre-comp tools that facilitate collaborative workflow, Python UI improvements, metadata
support, and the ability to register multiple Viewer Process
Gizmos for user-defined viewer LUT processing, including new
support for 3D LUTs and OpenGL GLSL shaders.
Additionally, Nuke 5.2 enables video monitor output through
Blackmagic and AJA Kona and Xena devices, and introduces
a RED R3D Redcode format reader that brings the full range
of picture information into a full 32-bit float-processing environment. Nuke 5.2 is priced at $3500.

Eyeon Software, maker of the Fusion compositing application, introduced its new SWAT certification program to
ensure that new and established VFX artists receive the
training and support they need to realize their full creative
potential.
With the new SoftWare Artist Training (SWAT) umbrella program, the company has expanded its international
network of trainers, colleges, online resources, and in-house
product support personnel to ensure that every Eyeon artist
is up-to-speed on the latest versions of Fusion, Generation,
Rotation, and Vision.
Eyeon will deploy trainers to assist new customers onsite to ensure that their artists are able to transition seamlessly to the new software. In addition, the SWAT umbrella
consolidates all the available training options for Eyeon
applications through Eyeons VFXPedia.com resource Web
site. Training partners on track to become SWAT-certified
include Class-on-Demand, cmiVFX, Digital Tutors, NADS,
NYU, Seneca College, Sheridan College, and others.
Innovation is no longer just about the technology; its
about knowledge of the tools, says Joanne Dicaire, director of business development and marketing at Eyeon.
Seasoned veterans dont want to feel like beginners when
they move to a new application. Shake artists, for example,
have done some of the best VFX work in the world, and
when they move to Fusion, we want them to produce work
that reflects their true abilities right away.
The SWAT program goes beyond the compositors work.
With Eyeons new Generation products, facilities now have
a means to visualize and manage their whole pipeline with a
degree of ease not previously available.

PRODUCT: COMPOSITING

PRODUCT: TRAINING

PRODUCT: GRAPHICS CARDS

The Foundry Unveils Nuke 5.2

During the past several weeks, a number of companies

introduced major new hardware and software products
at the Game Developers Conference (GDC) and at the
National Association of Broadcasters (NAB) show. We
have highlighted some of those here in the Spotlight
and Products (pg. 47) sections. Additional product and
company news from these conferences can be found
online at www.cgw.com.

June 2009

New Multibridge Pro has SDI, HDMI and Analog

editing with multi channel audio for only $1,595
Multibridge Pro is the most sophisticated
editing solution available. With a huge range
of video and audio connections and the
worlds first 3 Gb/s SDI. Advanced editing
systems for Microsoft Windows and Apple
Mac OS X are now affordable.

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 file formats.

Connect to any Deck, Camera or Monitor

Microsoft Windows or Apple Mac OS X

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.

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 film resolution 2K, 1080HD, 720HD, NTSC and PAL
for worldwide compatibility.

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 film using
real time 2048 x 1556 2K resolution capture and playback.
The Drawn Together images are courtesy of Comedy Partners.

Multibridge Pro

$1,595

Learn more today at www.blackmagic-design.com

Blackmagic Releases
Host of Offerings

Havok Reveals New

Middleware SDK

Blackmagic Design recently unveiled several new products,

including two new Videohub routers: a compact 16x32
model for under $5000, and the Enterprise Videohub, a
144x288 model for the largest broadcast and postproduction facilities that sells for under $30,000. Both sport 3Gb/
sec SDI, built-in deck control routing, redundant power and
switchable SD, HD, and 3Gb/sec SDI, and re-clocking. In
addition, Blackmagic updated the Videohub router software
to 4.0, which includes a new design with touch screens. The
update is free to all Videohub customers.
The company also rolled out OpenGear Converters with
SD/HD auto switching, DashBoard network control, and
3Gb/sec SDI, which are selling for $495. It also unveiled
DeckLink Optical Fiber, the first 10-bit SD/HD broadcast
capture card with both optical-fiber SDI and regular SDI,
for $495; HDLink Optical Fiber, a 3Gb/sec optical-fiber SDI
and conventional SDI monitoring solution for SDI, HD, and
2K monitors for $795; and new, bi-directional Mini Converter Optical Fiber for $495. Aside from a number of other
releases (which can be found on cgw.com), the firm also
introduced Ultrascope, a combination PCI Express card
and software waveform monitor for $695.

Havok, known for its physics software, rolled out a new

offering, Havok AI, which provides developers with solutions
for pathfinding and advanced character interaction in highly
dynamic game environments.
Havok AI meets the challenge of pathfinding with its new
platform-optimized SDK that features a robust automatic
nav mesh generator that creates the nav meshes in seconds
from complex game levels of hundreds of thousands of triangles, allowing for the rapid iteration of level content. With
Havok AI, the pathfinding is fully integrated and dynamic, not
static, with additional layered dynamic avoidance technology to handle thousands of moving obstacles in real time
with high fidelity. Fully extensible and customizable, the solution includes a hierarchical pathfinder that is multithreaded
and works on all the major gaming platforms. Furthermore,
with a predictive local steering module, characters predict
the movement of obstacles and adapt accordingly, moving
plausibly through the complex and often congested situations that arise when environments become dynamic.
Havok AI features out-of-the-box integration with Havok Physics, Destruction, Animation, and Behavior. Havok AI can be
purchased as a stand-alone product or with Havok Physics.

PRODUCT: VIDEO CARDS

PRODUCT: GAMING AI

Outsourcing within the Gaming Industry Is on the Rise

Think Services Game Groups Game Developer Research
has revealed the results of its 2009 Game Development
Outsourcing Report, which finds that the proportion of
respondents whose studios use outsourced game development rose 10 percent between 2007 and 2008, from 76
percent to 86 percent.
Almost 200 professional game developers were polled
anonymously to construct a comprehensive look at the
segment. The developers were asked questions about their
studios use of outsourcing, their outsourcing budgets and
plans, regional factors, and more.
The report finds that game development outsourcing is on
track to grow to be an even larger practice than it is now. Half
the participating developers who do not yet use outsourcing say they plan to begin doing so, and 95 percent of those
who already use outsourcing expect to continue. Other find-

NEWS: OUTSOURCING
6

June 2009

ings include the fact that a major contributor to the growth of

outsourcing is the increasing cost and bandwidth required to
create a high volume of assets for modern console systems.
Respondents reported that Microsoft Xbox 360 and Sony PlayStation 3 are the two platforms for which elements of game
creation are most often outsourced.
The report also finds that outsourcing is on the rise in terms of
overall budget allocation. The high end of outsourcing budgets
increased nearly twofold, with the proportion of companies
planning to spend $2 million or more on outsourcing rising to
almost 20 percent in 2008.
Additionally, the report discusses overall budgets, reasons
for outsourcing, the selection process for choosing firms
to outsource to, and the regions of the world with which the
respondents worked. The report can be purchased at www.
gamedevresearch.com.

ICE!
R
P
W
NE

Intensity Pro introduces professional HDMI

and analog editing in HD and SD for $199
Intensity Pro is the only capture and playback
card for Windows and Mac OS X with HDMI
and analog connections. Intensity Pro allows
you to upgrade to Hollywood production quality
with uncompressed or compressed video capture
and playback using large screen HDTVs.

Connect to Anything!
Intensity Pro includes HDMI and component
analog, NTSC/PAL and S-video connections
in a low cost plug-in card. Capture from HDMI
cameras, VHS and Video8 decks, gaming consoles, set-top boxes
and more. Playback to large screen televisions and video projectors.

Beyond the Limits of HDV

Microsoft Windows or Apple Mac OS X

HDVs heavy compression and limited 1440 x 1080 resolution can

cause problems with quality and editing. Intensity Pro eliminates these
problems and lets you choose from uncompressed video, Online
JPEG and Apple ProRes 422 for full 1920 x 1080 HDTV resolution.
Now you can capture in 1080i HD, 720p HD or NTSC/PAL video.

Intensity Pro is fully compatible with both Adobe Premiere Pro

on Windows and Apple Final Cut Pro on Mac OS X, as well
as Motion, Color, DVD Studio Pro, After Effects, Photoshop,
Encore DVD, Combustion, Fusion and many more.

Playback to your Big Screen HDTV

Use Intensity Pros HDMI or analog output for
incredible big screen video monitoring. Unlike
FireWire based solutions, Intensity uses an
uncompressed video connection direct to Final
Cut Pros real time effects renderer. No FireWire compression means
all CPU processing is dedicated to more effects and video layers!

Intensity Pro

$199

Learn more today at www.blackmagic-design.com

Caustic Graphics Readies New Raytracing System

Newcomer Caustic Graphics introduced CausticRT, a massively accelerated raytracing system for achieving breakthrough
levels of quality in interactive, cinema-quality 3D computer
graphics. The Caustic development platform, which includes a
raytracing accelerator card and SDK, is now available to qualified developers.
Raytracing duplicates the natural physics of light, creating
stunning images by meticulously tracing the path of light to
and throughout any given scene. Light rays naturally scatter
in many different directions (incoherent rays), and tracing and
shading them require memory access to many disparate parts
of the scene, a previously impossible caching task. CausticRT,
based on a new algorithm that addresses this issue, organizes
incoherent rays into a data flow that takes advantage of the full
computational power of CPUs and GPUs. CausticRT does not
displace a GPU or CPU in a graphics system; rather, it acts as
a co-processor that traces rays and schedules the results in a

manner that allows GPUs or CPUs to shade them as efficiently

as they do with rasterization.
Caustic Graphics is working with an emerging developer
community, such as Splutterfish, to create new or to port existing
renderers and applications to Caustic so artists and designers
can take advantage of the photorealism and visual effects that
make raytracing more compelling than rasterization.
CausticRT includes the CausticOne raytracing accelerator card
and the CausticGL programming API thats based on OpenGL.
CausticOne, an optional co-processor that works with CausticGL, unlocks the ability of the GPU/CPU to shade efficiently and
allows it to render 3D imagery up to 20 times faster than it currently can. A supporting SDK includes documentation, access to the
support portal, and a one-year subscription to hardware and software updates, as well as technical support. The CausticRT platform is priced at $4000 and includes CausticGL, a CausticOne
card, and one year of firmware and software updates.

PRODUCT: RAYTRACING

Autodesk Launches Flare

In addition to announcing the 2010 releases of its Flame and
Inferno visual effects systems, Autodesk presented Flare, the
new creative software companion to those hardware offerings.
The company also released a new version of Smoke, and 2009
Extension 1 releases of Lustre and Incinerator.
Flare 2010s aim is to bring the productivity of the Flame artist
to a new level without sacrificing any of the creative tools to
which users are accustomed. Flare bridges the gap between

2D and 3D, enabling artists to handle some of the Inferno and

Flame work via software while those systems are occupiedat a
fraction of the price of the hardware. Featuring the core creative
tool set of Flame and Inferno, the software is designed to
boost creativity, expand capacity, and develop talent for Flame
and Inferno customers. The product is intended for advanced
creative tasks, such as compositing and interactive design, as
well as support tasks, such as rotoscoping and keying.

PRODUCT: COMPOSITING

Allegorithmic Releases Substance Air

Allegorithmic recently rolled out Substance Air, a new texturing
middleware for the development and distribution of online and
downloadable games (MMOs, Free2Play, and XBLA/PSN).
The offering enables game developers to dramatically reduce
the size of downloadable applications, while providing a solution
for advanced user-generated content.
According to the company, Substance Air will improve the
online game development and delivery process by creating new
opportunities for artists in regard to user-generated content and
in-game asset customization.
Substance Air is part of Allegorithmics new generation of

PRODUCT: TEXTURING
8

June 2009

middleware for authoring and generating textures that feature

visual quality and graphical detail. It enables users to create
detailed graphics that are only a few kilobytes in size, thus
increasing the amount of high-quality content without overweighting the games client size. Substance Air also features
a new run-time engine optimized for generating textures at high
speed, even on low-spec devices, and a flexible authoring tool
to improve ease of use, with unique features designed to help
artists optimize their production pipeline.
Substance Air pricing models vary based on the type of game
and gaming platform.

By GerGely Vass

CG

Avoiding Gimbal Lock

o place an object in 3D space, its transformation properties need to be specified: position, scale, and orientation. The
first two attributes are easily definable by three numbers for
each. The meaning of the x, y, and z positions and scale parameters
are easy to understand, visualize, manipulate, and animate for artists. However, that is not the case for orientation.
Using an xyz triplet (three angular values) to manipulate an
objects orientation may become impossiblefor instance, during
some configurations of the three angles, such as gimbal lockand
lead to major problems when animating these values. Gimbal lock
is a phenomenon known for a long time, and it has caused severe problems long before computer graphics emerged. According
to NASA documents on the Apollo space program, pilots had to
keep a close eye on the Gimbal Lock warning light while maneuvering the spacecraft in order to avoid unwanted and dangerous
malfunctioning in the guidance and control systems.
The orientation, or angular position, of a rigid object has three
degrees of freedom. By holding a camera in our hands, for example, we can pan left and right, tilt it up and down, or roll it
without changing the point of interest. The most common and intuitive way to define these attributes is the use of Euler angles: The
orientation is represented by three consecutive rotations around
the main axes of a reference frame. However, the order of the rotational axes is something the industry has never agreed on, so
it is essential to supply this information if we transfer animation
data using Euler angles. Each major 3D application has a way to
change the order of rotations.
Using three gimbals, it is possible to construct a physical device,
a gimbal system, based on the principle of Euler angles. A gimbal
A former Maya TD and instructor, Gergely Vass
eventually moved to the Image Science Team of
Autodesk Media and Entertainment. Currently
he is developing advanced postproduction tools
for Colorfront in Hungary, one of Europes leading DI and post facilities. Vass can be reached
at [email protected].

10

June 2009

is a pivoted device, most often a ring, which rotates around a single

axis. By mounting a gimbal inside another one, the inner ring rotates around an additional axis, increasing the degrees of freedom
by one. Defining the orientation of an object with Euler angles is
like putting it inside a virtual three-gimbal device, and then rotating each ring by the corresponding angle. (Once again, the order
of rotations and the coordinate frame axes must be agreed upon.)

Astronauts, like animators, try to avoid gimbal lock. Shown here is the
Apollo 15 control panel with the eight ball indicating the gimbal-lock
danger zone with red.

The outer ring can represent the tilt, the middle ring the pan, and
the innermost ring the roll. However, most of the publications on
Euler angles refer to the three attributes as yaw, pitch, and roll, as
used in aerospace applications.
Having full control over the three degrees of freedom, one could
conclude that Euler angles (or gimbal systems, in general) are the
perfect way to describe orientation. Unfortunately, there are configurations wherein we lose one degree of freedom: the gimbal
lock. In this state, one of the gimbal rings is rotated such that it
aligns perfectly with another. In this situation, the entire range of
rotations is unreachable, and we may need to first re-orient the
locked gimbal in order to rotate the ring arbitrarily. If the angles
are near the gimbal-lock state, the gimbal system becomes unstable, as even small rotations (or round-off errors of the numerical
representation) may yield unexpected results.

Viewpoint

n n n n

e
m
s
r
)

d
n
s

d
e
l
t
f
e
s
l

It is fairly easy to get close to gimbal lock:

We simply need to rotate the middle angle,
or gimbal, by 90 degrees. Because two of the
rotational axes have become aligned, changing the first or the last angle in the chain results in a rotation around the same axis (see
images above). The only way to leave this
locked state is to adjust the middle rotation
again and move it out of the danger zone.

Gimbal Lock in CG
Computer animators controlling characters body parts or any kind of object using only three angular values should try to
avoid gimbal lock. Investigate your setup
before you start to animate, and change the
order of the axes so the middle axis is the
least one used. If all three degrees of freedom are needed, hierarchical transformations (parenting an additional node) may
help. If readability and human interaction are not important, such as under the
hood of game engines, then quaternions
and rotational matrices are used instead of
Euler angles. Quaternions are four-dimensional representations, ideal for computing
smooth transitions between poses. While
gimbal lock never happens when using
quaternions, they are hard to read and conceptualize. In spite of all the issues of lock-

A three-gimbal system is analogous to the Euler angles. In the above left image, all the axes are at
their default positions. If the middle (red) gimbal is rotated 90 degrees, as seen in the above right
image, the blue and the green axes become aligned, losing one degree of freedom.

ing gimbals, Euler angles are the industry

standard for describing rotations.
Real-world gimbal systems have been
known and used for a very long time now.
The gyroscopeinvented nearly 200 years
agois a fine example: At the heart of this
device is a platform, with one or more spinning wheels or disks mounted in gimbals.
The large angular momentum of the spinning elements forces the central platform
to remain almost perfectly fixed, regardless
of the motion of the devices shell, as the
gimbals let it rotate freely. By installing sensors to detect the rotation of each gimbal,
we can measure the relative orientation
(or attitude) of any object as it relates to
a fixed reference system. That is how the
Inertial Measurement Unit (IMU) of aircraft, spacecraft, and watercraft, including
guided missiles, works. The three gimbals
hold the central stable platformwith the
spinning elementswithout introducing
any notable external torque, which would
force the platform to lose its original orientation. That is where gimbal lock becomes
a real problem: By losing one degree of
freedom near this state, the fixed platform

cannot rotate freely, thereby risking the loss

of reference.
This may be a good time to finish reading the magazine and begin watching the
1995 movie Apollo 13 featuring Tom Hanks
(and some great visual effects). You will
be surprised how much they talk, or yell,
about avoiding gimbal lock and losing the
reference of the Inertial Measurement Unit.
The IMU of the Apollo space program
was a three-gimbal system, and pilots were
required to navigate the spacecraft so they
did not approach the 10-degree danger zone
around gimbal lock. The current attitude of
the vehicle was displayed on the Flight Director Attitude Indicator, or the eight ball,
as pilots referred to it. If the indicator entered
the red danger zone centered at yaw 0 and
180-degree poles, and the stable member
lost its attitude reference, the gimbals had to
be re-aligned in-flight against star references.
So, the next time you are fighting gimbal
lock in a 3D animation package or game
engine, dont panic. Just think of the Apollo
13 crew members, who were losing oxygen
and electrical power in a crippled spacecraft
at the same time. n
June 2009

11

Character ModelingAnimation

Pixar Animation Studios once again

redefines CG animated features with
its 10th film, Disney/Pixars Up
By Barbara Robertson

12

June 2009

Character ModelingAnimation

square. A circle. Two fundamental shapes in geometry, two

iconic shapes in computer graphics. And, the simple basis for
the visual stylethe shape languageof Disney/Pixars 10th
animated feature, Up, one of the most sophisticated ever in
terms of story, design, and technology. Directed by Pete
Docter (Monsters, Inc.), co-directed by Bob Peterson, and
written by Docter and Peterson, Up is an action-adventure, a comedy, a most
unusual buddy film, a circle-of-life story, and Pixars first film in stereo 3D.
Carl Fredricksen, 78 years old and voiced by Ed Asner, is the square and the
films central character. His default expression is a scowl. But, thats now. When
the story begins, hes eight years old and enchanted by the exploits of the pilot
Charles Muntz, who he sees in a black-and-white newsreel claiming to have found
a prehistoric bird, a missing link, in Paradise Falls, South America. On his way
home from the movies, little Carl meets Ellie, a bouncy tomboy, who is also a fan
of Muntz and his motto, Adventure is out there. Ellie is Carls first circle. They
marry, and we watch their lives spool forward. Carl becomes a balloon salesman.
They save money for an adventure out there, but something always stops them
a flat tire, a house repair, medical bills. Even so, theyre happy as they grow old; life
is colorful. And then, Ellie dies. The color fades from Carls life. He stays
in his house and talks to Ellies picture. He becomes rigid.
But, two events happen. First, Russell, an eight-year-old Wilderness Explorer, shows up on Carls doorstep, much to the grouchy
old mans annoyance. Russell is a round little boy, full of joyful enthusiasm. He needs to help an elderly person to earn his last badge.
Second, Carl finds himself sentenced to life in a retirement home. That spurs him to fulfill his promise to
Ellie. As the retirement home orderlies wait at the
curb for Carl to emerge from his house,
10,000 balloons rise up from behind the house and lift it from
its foundations. Carl is on
his way. He sits back,

The personalities
of Dug, Russell, and Carl
show on their animated faces
as they hang onto Carls balloonpowered house, while Kevin the
bird catches an unruffled ride
on the roof.

Images 2009 Disney/Pixar.

June 2009

13

Character ModelingAnimation

content, in his easy chair as his house floats

past skyscrapers. And then, Russell knocks
on the door.
When Pete and Bob pitched the story,
with Bob reading, it brought John to tears,
says Jonas Rivera, producer, referring to the
directors Docter and Peterson, and to John
Lasseter, Pixars chief creative officer.
The idea originated with Docter and
Peterson noodling around with an escape
fantasy, of just floating away. They wanted
to find a unique character, one they hadnt
seen before in an animated feature. Docter
drew a grouchy old man holding a bunch of
brightly colored balloons. They laughed, the
brainstorming began, and soon the balloons
were floating the old man in his house.
All they needed to do then was discover
why the old man was in a floating house,
where he would go, and what would happen when he got there. It took three years
or soPeterson, Docter, and story artist
Ronnie del Carmen

Russells multiple layers of clothing and Carls thick jacket and boxy trousers created unique
problems for the cloth-simulation team to solve. At first, character developers tried making Carls
hair thick, but it was too distracting.

started in 2004. When the Ratatouille production pulled Peterson off the Up project
for a while, Docter brought in Thomas
McCarthy, writer and director for The Station Agent, who took notes for six to eight
months, as Rivera puts it.

ers are big

uare. His trous
sq
is
ad
he
is
dnt start
re. H
s tall and squa
uare. But, he di
ad
sq
e
he
ar
e
s
re
nd
th
s
ha
e
H
d a little
on his
Ellie, his face ha
the age spots
n
et
m
ve
t
E
.
rs
fi
xy
d
bo
an
d
and
an
ght
e rigid than Ellie
n he was only ei
or
he
m
y
W
.
sl
ay
ou
w
vi
at
ob
th
life
l are in
ily are
pictures of Car
he and his fam
e,
l,
us
til
S
ho
r
s.
ei
es
th
dn
In
roun
edding.
e see at their w
ther are
her family, as w
of the two toge
es
ur
ct
pi
le
hi
w
fter Ellie dies
square frames,
an oval matte. A
ith
w
es
m
fra
re
e circles are
in squa
g balloons, all th
llin
se
s
op
st
l
ar
comes totaland C
d his design be
an
e,
lif
s
hi
m
es boxed in,
gone fro
aws from life. H
dr
ith
w
e
H
.
re
ne threatly squa
at is, until someo
Th
s.
ay
w
s
hi
memostuck in
his homespun
in
e
lif
e
iv
us
cl
ens his re
lp of 10,000
then, with the he
rial to Ellie; and
venture, Pixar
ars. For his ad
balloons, he so
ic circles to
ily of enthusiast
provided a fam
ges.
chy old guys ed
soften the grou
l were in
allenges for Car
The technical ch
s to show
facial expression
s
be
cu
a
ng
gi
rig
lhouettes
creating the si
in
d
an
n,
io
ot
em
ersized
despite his ov
d
te
an
w
s
or
at
anim
s a result,
ide pant legs. A
w
,
xy
bo
d
an
et
jack
character it
most complex
e
th
m
hi
lls
ca
Pixar
has created.

14

June 2009

We spent a lot of time on the story,

Docter says. The actual production was
only during the last year and a half. Its our
10th film now, and we have really good
people who have honed their craft. We
could push the production schedule.

Simplexity
As Docter, Peterson, and del Carmen
worked to refine the story, production design began developing the visual language.
The story is about a house pulled by balloons floating in the sky, Docter says. We
needed to create a world where that was
possible. The story pushed us to a level of
stylization wed never done before.
Ricky Nierva, the production designer,
named the resulting look simplexity.
This is a movie about age, about the authenticity of life, Nierva says. Our big
challenge was not to make it too photoreal. Otherwise, why not make it live
action? On the other hand, if you
pull too much detail away, it looks
like cheap CG. So we looked for
the sweet spots for characters, environments, and details.
For design inspiration, the
character and production designers
chose the work of Mary Blair, a Disney artist who developed the color
and style for Cinderella, Alice in
Wonderland, and Peter Pan,

Character ModelingAnimation

and who was art supervisor for such films

as The Three Caballeros and Saludos Amigos,
set in South America. They also looked at
George Booth cartoons, and Hank Ketchums Dennis the Menace.
To create storybook backgrounds, the
artists used simple shapes for bushes in the
jungle, eliminating any leaf that didnt face

the camera, and stylized the shading and

modeling for Carls house. We wanted
Carls house to feel small, like a dollhouse,
notes Steve May, supervising technical director, as if big hands made small pieces
and placed them by hand.
The trim around the fireplace doesnt
quite line up. The fireplace has soot, but if

Carl and Russell travel in Carls balloon-powered house through skies filled with
clouds in approximately 150 shotsfluffy, white clouds and dark, stormy clouds.
The clouds are volumetric, constructed from spheres using a system developed
by TD Alexis Angelidis. In the entire film, 400,000 different spheres create the
clouds; the storm alone uses 85,000.
Alexis can place a sphere, maybe even as large as a football field, explains
Gary Bruins, effects supervisor, and then run a script that breaks it into a fractal
distribution of spheres with different radiuses. The scripts helped with general
modeling, but Alexis built the clouds by mostly adding sphere onto sphere.
The various-sized large spheres described the overall cloud shape; the smaller
spheres added detail.
Each cloud model had two shader versions, one for the surface and another
for the interior. The interior allowed object mattingthe house, for instancebut
no camera motion blur. The surface supported camera motion blur but didnt
allow embedded objects. We had a switch that we could flip on a per-shot
basis, Bruins says.
To scatter light within the clouds, TD John Pottebaum created a 3D density
map. In addition to lighting artists using their tools and Johns 3D density map,
Alexis replicated, in the shader, the lighting response of water droplets in the
cloud, Bruins explains. And, his volumetric caching system was a huge help in
optimizing rendering. Barbara Robertson

n n n n

you look closely, you see that the soot is actually brush strokes made with a dry brush,
as are the bark on the trees and the dirt on
Russells face. Youll see this style everywhere
in the movie, May says. We used a variety
of procedural and paint components in the
shaders, but generally we had a painted brush
stroke in a paint pass with every shader.
In addition to Carl, Ellie, Russell, and
Charles Muntz, the main characters include
a 13-foot-tall bird that emerges from the
jungle after the balloon-driven house lands
on a tabletop mountain in Venezuela. Russell names the bird Kevin, but the dog that
also pops out from the brilliant jungle
names himself: Dug wears a collar that
translates his thoughts into words. Suddenly, Carl now has two family members
circling him. Kevin is colorful and silly,
with a round body and long neck, and
Dug is an enthusiastic, overweight golden
retriever/lab mix that loves everyone. Carls
a tough old cookie. It takes more than one
buddy to bring him back to life.
For the characters, modelers working in
Autodesks Maya adopted a less is more
philosophy. Carl is three heads tall with
stocky arms and legs. Russell is a little egg.
Dug has a big, round nose. We removed
anything extra, May says. The characters
dont have ear holes or nostrils, for example, not even Dug.
It was character supervisor Thomas Jordans team of approximately 30 people who
had first turned the concept art, maquettes,
and expression sculpts for characters into
3D models that could move and that met
the design goals. Each of the main characters had its own unique problems to solve.
It was more of a challenge than usual to
finalize these characters, Jordan says. No
one can tell you what simplexity means. We
had to find it. The sketches and maquettes
didnt work in 3D once we started animating them.
Animators gave the riggers drawings of
expressions they wanted to hit, and the
riggers, working in Pixars proprietary softJune 2009

15

n n n n

Character ModelingAnimation

Beautiful Balloons
The effects departments role was to create the believable elements that connect
the audience to the stylized world. By animating the balloons in a realistic way,
the audience more easily believes the house could float, maintains Steve May,
supervising TD. When they buy that concept, they believe Carl is going someplace. And, when Russell shows up on his front porch, they believe he really is
in peril. Our primary goal is to help tell the story. Our secondary goal is to make
something beautiful.
At first, knowing it would be difficult to run a rigid-body simulation on 10,000
balloons, Pixar tried to float the balloons procedurally within a modeled canopy.
They had no notion of each other, though, says Gary Bruins, effects supervisor.
And, they intersected all over the place. It didnt look
good. TD John Reisch did tests using rigid-body
dynamics with Open Dynamics Engine (ODE)
through Autodesks Maya that produced the
believable motion they wanted, but the system was able to simulate only 500 balloons.
When Eric Froemling split ODE away from
Maya, though, and created an improved,
stand-alone version that ran in Pixars pipeline,
they upped the number. Considerably.
With ODE installed in Pixars pipeline, Reisch
created a system in Maya that generated the initial
state of the balloon simulation, then moved that state to
ODE as a stand-alone without using data communication.
Releasing it from communicating with Maya released the bottleneck, says Bruins. We could do 50,000 balloons, but because
they werent intersecting, we needed only 10,000 to fill the canopy.
The canopy provided artistic control for the overall shape of the
grouping. The simulation can push the balloons out of that shape,
but when the rigid-body sim was in its rest position, it had the same
topology as the canopy, Bruins explains. Thus, to maintain the art-directed shape,
the team decided to move the strings that tied each balloon to Carls chimney in
a separate simulation.
The strings are more complicated than the balloons because getting them
to bend and conform around the balloons requires more control points, Bruins
says. If all the strings had to avoid the balloons along their paths, they would
become tangled and twisted.
So, in the first pass, the crew simulated the balloons with strings that didnt
know about other balloons. Then, they brought that simulation in as baked data
and simulated only the strings. We thought about reducing the number of
strings, but as it turned out, we didnt have to, Bruins says. The strings collide and interact with surrounding balloons, but they dont influence the behavior
of the balloons. For close-up shots, though, they would simulate balloons and
strings together, and hand dress the hero balloons. Barbara Robertson
16

June 2009

ware, shaped the mesh topology created

by the modelers to match. We set up the
mesh so the rigging controls created the
shapes the animators wanted, Jordan says.
But in the case of Carl, we spent quite a bit
of time and a lot of trial and error to give a
square the ability to have human emotions
that are not symbolically square.
One problem was Carls mouth, which
stretches in a hard line across his huge bottom jaw. We had to make some compromises, Jordan says. We softened the corners, but we tried to do it in a subtle way so
you would feel it without seeing it.

Cranky Old Carl

Animators working with Carl discovered
that economy in motion matched his simple style. Wed put his face in a pose and
let the pose speak for the character, says
Scott Clark, animation supervisor. Carl
feels more alive if you dont animate him
than if you do. Sometimes, wed just put a
blink on him and not much else. Hes old.
He doesnt bend or twist too much.
Animating Carls clothes, however, became a technical problem. To give viewers the feeling that he lives in a miniature
world, the designers increased the size of the
weave on Carls herringbone wool jacket to
look as if someone had scaled up a dolls
jacket to fit someone five feet tall. And,
it isnt just the textures, says Jordan. Its
the way the fabric moves, as if someone cut
a dolls jacket from a life-sized jacket. But
nothing in real life looks like a scaled-up
dolls jacket. So we had to guess. Then, we
worked with animation to do walk cycles,
have him sit and stand and interact with
Russell to see if he was believable and fit
the design of the film. We didnt want him
to feel like a doll.
In the end, Carl looks like a little old
man in an oversized suit, a square, boxy
shape with wide pant legs. And that caused
problems with cloth simulation. We
wanted the miniature look with the oversized clothing, and to see the lines of ac-

Character ModelingAnimation

tionthe bends of knee, the lines of the

elbow, the position of the shoulders relative
to his head, May says. But, when Carl
bent his little legs inside his big trousers
as he walked, you didnt see the knee bend
with normal cloth simulation; it looked
like he was walking with straight legs.
Our animators got really upset, May
relays. Theyre used to doing subtle and
important differences in poses to sell the
acting of the scene. To fix that problem
and to give Carls clothes a thick look with
a minimum number of wrinkles that the
artists wanted, the cloth-simulation crew
used targeting with modeled surfaces that
the simulation would approximate, and
shapes based on baked simulation that they
isolated to certain areas of the body.

Holding onto Russell

The problem with Russell was his eggshaped face. Hes basically a balloon, Jordan says, and, my gosh, that was challenging. The smallest facial expression would
change him and make him look too old, or
too young, or not appealing. Giving him

Special shaders accented the ability of lighters to change host Charles Muntzs expression from
kindly to sinister.

a chin to separate his head from his neck

helped, but often the riggers would make
their best guess and then adapt the rig based
on notes from the animators and directors.
Russell is very caricatured, Clark says.
And, hes a kid, so he has to jump around
and be active. But, if we moved his neck
too much, hed feel gummy, so we built
limitations into the model.
When Russell jumps around, he moves
several layers of clothinga shirt, a sash on

with a
s a tiny body
ha
he
S
.
le
rc
d bare feet.
llie, is a ci
utiful smile, an
r
ea
Carls wife, E
b
a
,
se
no
n
s dirt under he
, a little butto
e
clothes and ha
us
n
w
ho
o
r
d
ei
eballoon head
th
-m
in
rs hand
photos
ea
er
w
H
he
e.
S
at
.
m
cy
soul
Shes boun
e meet
laymate and
r is loving. W
hes Carls p
he
S
.
ut
o
ils
ab
na
er
ng
ng
fi
back
erythi
during Carls
ames, and ev
,
fr
lm
d
fi
e
un
th
ro
f
o
in
are
eginning
g girl in the b
her as a youn
with Carl in
lif
r
llow he e
fo
e
w
d
an
,
story
sequences
st emotional
o
m
e
th
f
o
one
feature.
r an animated
fo
d
te
ea
cr
after Ellie
ever
ure begins
Carls advent
motivation. To
e underlying
th
is
e, prohe
S
s:
ie
d
ugh the movi
ro
th
t
iri
sp
s
gave her
carry Ellie
icky Nierva
R
r
ne
ig
es
d
at shows
duction
magenta, th
r,
lo
co
lic
o
b
remind us of
a sym
and skies, to
s
er
w
o
hino crefl
in
up
ichael Giacc
M
r
se
o
p
m
at plays
her, and co
llies theme th end,
E
as
tz
al
w
the
ated a
d turns until at
with twists an
actionan
omes
when it bec
eme.
adventure th

top, a neckerchief on top of that, his backpack straps, and the backpack itself with
between 20 and 30 Wilderness Explorer
gadgets hanging from it, all animating independently. For cloth, we have a two-stage
process, May says. Animators perform
the character without clothing, and then
simulation does the clothing after theyre
done. But the backpack is a rigid-body
simulation that runs nearly in real time.
As the animators put Russell through his
poses, they could see the backpack and all
its parts moving, and because the software
converts the dynamic movement into keyframe data, they could change the simulated movement.
This is the most complex clothing
weve ever done, says May. And then,
we attach these two complicated
characters, Carl and Russell,
by a hose and a rope to each
other and to a house thats
attached to 10,000 balloons.
The animation rig is amazingly complex. Depending
on how hard Carl pulls on the
hose, it might move the house,
the wind might push the balloons,
and Russell is pulling, too.
This is the situation: Carl has wound the
hose attached to the house, floating above,
around his shoulders. Hes tied a rope from
Russells backpack to a knot on the hose.
June 2009

17

The tethers form an upside down y with

the house on top. Because animators wanted control, they used a sophisticated rig
to move the tethers, rather than rely on a
simulation to create the motion. The house
moves independently, sometimes reacting
to balloon simulations, sometimes moving
with keyframe animation with the balloons
reacting to that.
In most shots, we didnt see

Carl tied a rope from Russell to a hose hes using to pull the floating house. Creating the final image
at top required water simulation (above, far left), character animation with rigid-body simulation
for Russells pack and a sophisticated rig for the tether (above, second from left), cloth simulation
(above, second from right), and shading and lighting (above, far right.)

all these things at once, but when we saw

all three, generally animators performed
the characters sans clothes and with the
tether in a rough position for the house,
May says. Simulation ran dynamics for
the clothes and, if

necessary, improved the animation of the

tether. And, effects ran the simulation
for the balloons and sometimes for the
house. In other words, animation first,
then cloth simulation and effects working
in parallel.

Iridescent Kevin
Kevin, the bird, needed to be a character
Eight-year-old Russell, a Wilderness Explorer who inadnever seen before and dramatic enough to
vertently stows away on Carls adventure, takes Ellies place as
give Muntz, the pilot who was Carl and
a circle in Carls life. Russell rolls joyfully through life, excited, constantly
Ellies hero from the newsreel footage,
circling around. He looks like a little egg with stubby arms and legs and no
a reason to spend 50 years looking
neck, and he wears a backpack loaded with Wilderness
for her. She was the hardest charExplorer camping gear, a neckerchief, and a sash
acter because we walked the fine
covered with Wilderness Explorer buttons.
line between too unreal and too
The technical challenges for Russell were in
cartoony, Jordan says. Pete
creating facial features and expressions that
said that when the audience
looked appropriate on his smooth, young,
first sees her, he wanted her
oval face, and in simulating all his layers
to look like shes made of gold
of clothing and gadgets.
when the sun hits her, but that

she really could exist in nature.

The art department found images of a Monal pheasant from the
Himalayas that has iridescent feathers
in metallic green, purple, red, and blue
colors, a copper-colored tail, turquoise
blue skin, and a blue crest like that of a
18

June 2009

Character ModelingAnimation

Kevin is a rare, 13-foot-tall flightless bird, colorful and

silly. Carl and Russell happen across her after they fly
inside Carls house to South America. A supposed missing link, she is the object of (Carls hero) Charles Muntzs 50-year-long quest. In Kevins
shape and coloration, she, too, reminds us of Ellie: Kevin has a
round body, a long neck, and brilliant iridescent feathers.
The technical challenges for Kevin were those brilliant iridescent feathers.

peacock. The crew was able to see a pair in

a nearby animal sanctuary, and those birds
served as reference material for Kevins colors. Kevin became a 13-foot flightless bird
with an orange beak, a long, iridescent blue
neck, and purple feathers. John Lasseter
always says that before you can caricature
reality, you have to understand reality, Jordan explains. To be believable, it has to be
inspired by something that could exist.

For animation inspiration, the studio brought an ostrich to the campus, and the animators took
pictures as it walked around outside, but
they werent bound by the real birds physical limitations. Kevin is a made-up creature, and probably the character we had the
most liberty with, says Clark. Russell uses
her like a pogo stick. Pete always said that if
you could hear the sound her brain made, it
would be like a dial tone.
For Kevins feathers, the technical direc-

Net Knots
In one sequence during the film, a character gets caught in
a bolo net. To create that net, rather than using a sheet of
cloth with a texture or geometry attached, TD Eric Froemling drew on the same rigid-body simulation system used to
float the 10,000 balloons carrying Carls house.
I had first played with this idea on a promo spot for Cars,
says Gary Bruins, effects supervisor. Theres a shot where
Mater runs into a velvet rope, and I tried connecting spheres
to approximate the rope. I talked to Eric about the idea, and
he gave it a shot.
Heres how it worked. Eric strung together a series of
spheresrigid bodieslike a string of beads hanging in a
doorway, Bruins says. With enough spheres, we could approximate the behavior of a rope. He made a net by knotting

tors used one Ri curve for the quills, with

hundreds more coming off the sides to
produce each barb. For Finding Nemo,
we modeled feathers using wide hair or
a single piece of geometry with a clever
shader, May points out. For Kevin, we
wanted more geometric complexity. And,
because we used individual pieces of geometry, we could model a range of feather
types from flat feathers that we might have
done with the shader before, and also fluffy
under-feathers.
To color Kevins feathers, the artists created a shader that gave the directors the
ability to choose which colors turned on
and off, and at which angles, as well as texture maps painted with a ramp of all the
colors the feathers could exhibit, to control
the iridescence. The order in which colors
appear in the map is the order they shift as
the angle changes between the camera and
the light source, Jordan says. Its a fairly
big cheat, but Pete wanted the ability to
turn the iridescence off.

Barking Is So 20th Century

For the dogs, the character supervisors and
animators spent time in a doggie day-care

the ropes into a 2D grid. Any time a character interacted

with the net, we got a more believable response from the
rigid bodies than we would have with cloth. Because it had
open squares, if a limb went through an opening, there was
no motion on the net.
To render the net, TDs ran Ri curves through the centers
of the spheres and shaded the curves with rope textures.
Similarly, Froemling stitched together a series of triangles
and made a quilt of rigid bodies to create a tarp that dropped
onto the balloons. As a result, the TDs could more easily
make the tarp interact with the rigid-body simulation for the
balloons than if they had used cloth simulation for the tarp.
Bruins intends to keep exploring this use of rigid bodies
to do soft-body simulations. The nice thing is that you can
easily detach the strings of spheres and get ripping and rupturing, so it has a lot of potential for large-scale cloth and
special cases. Barbara Robertson

June 2009

19

Character ModelingAnimation

center, consulted a behaviorist nearby, and,

of course, studied their own dogs. If you
look at their designs, Clark says, Dugs
nose is really big, almost a Snoopy nose.
But we do things with the behavior of the
rig to flavor the animation more toward realism. We have controls to give them that
little flavor, but we consciously chose to do
dog behavior.
In addition to the happy-go-lucky
Dug, the film features Muntzs pack of
hunting dogs, which he trained to serve
himeven to the point of serving dinner,
albeit somewhat sloppily. Where Dug is
round and happy, the pack dogs are an-

The bad master in Up is Charles

Muntz, voiced by Christopher
Like all the dogs
in Up, Dug
Plummer, whose design is as
wears a high-tech
co
lla
r
that translates
angular as that of his hunthis thoughts into
spoken words, bu
t Dug is
different from the
ing pack. Petes challenge
rest of the pack.
H
e
s round;
th
eyre angular. He
to us was he wanted the
s the happy-go-lu
cky
nerd of the hunt
audience to feel like
ing pack, sent ou
t
by the others on
Muntz was a warm,
a hopeless missio
n
to get him out of
the way. And that
loving grandfather,
s
when he adopts
Carl and Russel
who could become
l.
For the animator
s, the goal was
sinister at the drop of
to give all the do
gs realistic
a hat, Jordan says. The
behavior, yet st
ill have
them fit within
crew accomplished that,
the
caricatured style
in particular, by rigging
of the film.
his eyes and eyebrows to be
appealing or scary, and
through shading.
Athena Xenakis
was able to find the
right shading details on his face, the circle of life really is, and always was.
such as crows feet, that could be And, Pixar reminds us, for the 10th time,
enhanced or hidden, and to make that the medium is deep enough to hold
his eyes look sinister when com- any story, whether funny, adventurous,
bined with certain lighting, Jor- scary, emotional, heartwarming, heartdan points out. When hes hold- breaking, heartpounding, or, sometimes,
ing a lantern and talking about all of that rolled into one. That is, one
Russell, Dug, and Kevin helped soften Carls edges, but
meeting his childhood hero Charles Muntz gave Carls
his obsession with Kevin, we meticulously designed and accomplished
adventure its most exciting turn.
goosed the lighting to bring out film named Up.
gular and serious. Alpha, a Doberman, the creepy details.
MuntzCarl, and Ellies childhood Barbara Robertson is an award-winning writer and a
leads the pack with Beta, a Rottweiler,
and Gamma, a bulldog, right behind heroforces the climax of the film. Carl contributing editor for Computer Graphics World. She
him. All the dogs talk via a high-tech col- learns what a well-rounded adventure in can be reached at [email protected].
lar that, among other things, translates
their thoughts into spoken words. Peterson voices Dug and, with some technical
tricks, Alpha.
For Pixars first adventure in Disney Digital 3D-land, the studio focused inward,
Pete and I are lifelong dog lovers, Peterturning stereo into a window, rather than lobbing gags out from the screen and
son says. So we looked for a unique way to
into the audience. We used [stereo] 3D to help tell the story, says director
have Carl and Dug talk. The collar allowed
Pete Docter. When Carl is stuck in his house, we kept it flat, but when hes
us to get the behavior of a real dog and hear
standing on a cliff, we wanted to feel the wind on his face.
its thoughts; Dug could be scratching his
Stereoscopic supervisor Bob Whitehill devised a depth script based on the
ear while hes talking.
color script. When Carl and Ellie are young and life is colorful, the images have
The real behavior of Alpha, Beta, and
depth. When Ellie dies and the color fades from Carls life, the images are flat.
Gammas pack, on the other hand, is often
Then, Russell shows up. The scenes get deeper and deeper. Notes supervisfrightening, especially when they turn their
ing TD Steve May, Everything [at Pixar] has always been 3D, so, technically, it
hunting prowess toward Carl, Russell, and
was just matter of adding a camera and rendering. Barbara Robertson
Kevin. But as we learn and Peterson affirms,
There are no bad dogs, only bad masters.

Deep Emotions

20

June 2009

CGW :808_p

7/16/08

11:45 AM

Page 1

CGLive Action

Finally, someone shot a feature film inside

the Smithsonian Museums. And, such
a film. Directed by Shawn Levy, the 20th
Century Fox feature magically brings to
life works of art, historical figures large and
small, creatures, statues, and Einstein bobbleheads in a sequel aptly named Night at
the Museum: Battle of the Smithsonian.
The battle features, as did the first Night
at the Museum, Larry Daley (Ben Stiller),
the museum guard from the earlier film,
now an entrepreneur and inventor; Jedediah (Owen Wilson) as a miniature cowboy; Octavius (Steve Coogan), another
miniature; and Teddy Roosevelt (Robin

R&H animators could cut loose with the

Einstein bobbleheads and have fun creating
facial expressions for the cartoony characters.

Williams). Introduced in the sequel are

Kahmunrah (Hank Azaria) and Amelia
Earhart (Amy Adams). But, enough about
the actors. The fun in this film centers on
watching familiar inanimate objects and
iconic artifacts take on a life of their own.
Rhythm & Hues (R&H) was the primary visual effects house for the film, and
22

June 2009

a crew of approximately 300 in its Los Angeles and Mumbai and Hyderabad, India,
studios created an animated octopus, the
Einstein bobbleheads, various sculptures,
airplanes, falcon heads for the Horus, a
squirrel, and digital doubles. CafeFX concentrated on bringing paintings to life.
Co-visual effects supervisor Raymond
Chen, along with Dan Deleeuw, led
R&Hs work on the 535 shots. The fun of
this project was that we had different kinds
of work. It wasnt like having an orange
cat for 500 shots, says Chen. We had a
furry creature but also hard surfaces, metal,
marble, and octopus flesha lot of different problems to solve.
The octopus was the most difficult character, the bobbleheads the most fun. At
Rhythm & Hues, artists model with Autodesks Maya, and for displacement, they
use Pixologics ZBrush and Autodesks Mudbox. For most other tasks along the pipeline,
the crew uses the studios proprietary tools.
To handle the eight octopus tentacles,
which were like limbs without bones, character riggers created a system that included
procedural animation. The director wanted
something anthropomorphic and not too
cartoony, Chen says. The rig had splines
controlled with lots of knobs, automated volume preservation so that if you extended a
tentacle, it would get thinner as it stretched,
and methods that helped the tentacles
and suckers slide over each other.

To prevent interpenetration, geometry

on the ground or another tentacle could displace a tentacle overtop. When animators
dragged one tentacle over the top of another,
they didnt have to worry about the contact,
Chen says. The rig would cause the bottom
tentacle to push the top one up.
After the animators completed their work,
the effects department slimed the octopus
based on proximity to other surfaces, using geometry rendered with transparency.
When the tentacles wrap around Ben Stiller, we tracked in geometry for him, Chen
explains, and put slime in the areas closest
to the tentacle. As a tentacle got farther away,
the slime would stretch and then snap.
The marble statue of Abraham Lincoln,
by contrast, had an opposite challenge: It
couldnt look stretchy or rubbery.
The
animation

A
r
s

At left, actor Ben Stiller tosses water from a Turner painting animated by CafeFX onto an octopus
created and animated by Rhythm & Hues, while Amy Adams looks on. At right, a special rig helped
the crew at Rhythm & Hues lock CG falcon heads onto stunt actors shoulders to create the Horus.

style was more constrained, but even in

the rig, we limited the area of influence for
movements in the mouth, says Chen.
Similarly, to stiffen the cloth, the
simulation artists hid the movement inside the folds.
In one scene, Lincoln smashes
through a window and attacks the
Egyptian god Horus, a humanoid
with the head of a falcon. Horus
emerges from an effects-heavy
netherworld filled with particle
smoke and mist. Transforming
flocks of these bird-like creatures from
the mist and chaos of the underworld
was a healthy amount of work, even
though its a handful of shots, Chen says.
We did some things with procedures to
warp and transform the geometry.
The creatures bodies were stunt actors filmed on greenscreen, onto
which the artists fit

A constrained animation style and limitations in the

rig helped animators create a performance for the
statue of Lincoln without losing its marble essence.
Images 2009 Twentieth Century Fox Film Corporation.

the bird-like heads using a rig that helped

lock down the feathers in the transition
zone. To create the feathersa number of
which are metallic and inlaid with jade, several are dreadlocks with metal claws at the
end, and many others are like more typical hair and furthe crew used multiple
techniques. Some were cards of geometry
with opacity maps, Chen says. Some were
hair or fur groomed in layers to look like
feathers. The dreadlocks and metallic feathers used simulationscloth sims to swing
the dreadlocks and rigid-body dynamics to
swing the metallic feathers.
The challenges for the squirrel and the
bobbleheads centered more on animation
artistry than on technology, for the most
part. The squirrel is a scale gag, Chen
says. We see him most of the time from
Octaviuss point of view, and hes one of
the miniature diorama guys. So, part of the
joke is that the squirrel rears up and appears to be roaring, but hes still a small,
cute squirrel. The studio used its welldeveloped fur tools to keep the critter looking fluffy, even when it interacts with the
White House lawn, which the crew
created with CG hair.
The animators had the most freedom in animating the Einstein
bobbleheads. Modelers worked
from scanned practical models to create the caricatured
versions of the renowned
scientist. The bobbleJune 2009

23

n n n n

CGLive Action

heads were great to work with because of

their proportions, Chen says. Einstein has
a huge head thats constantly bobbling, and
short arms and legs. The animators didnt
have to worry about keeping him stiff like
the sculptures. He had a lot of dialog, so the
animators could be expressive. It was about
getting the best performance.
In addition to these characters and creatures, R&H created digital doubles that the
crew used in whole and in part for Stiller
(Larry), Adams (Amelia), and several other
actorsespecially for Coogan (Octavius),
who rides the squirrel. And, the studio animated such famous sculptures as Auguste
Rodins The Thinker, Edgar Degas La
Petite Danseuse de Quatore Ans dancer,
Jeff Koons Balloon Dog, and an abstract
sculpture by Isamu Noguchi.

Leaping Off the Wall

The artistry at CafeFX, on the other hand,
centered on paintings in the National Gallery. When Ben Stiller gets hit with a
snowball thrown by a kid in Agnes Taits
painting, its the clue that everything comes
alive in the museum, says Scott Gordon,
visual effects supervisor at CafeFX, referring
to Taits famous Skating in Central Park
painting of winter revelers from 1934.
For background paintings, the animation is subtleyou might see a bit of a wall
breaking, or a curtain moving slightly. But
some paintings played a more active role. For
example, Stiller grabs a pitchfork from Grant
Woods American Gothic. And, in another
shot, a JMW Turner seascape saves the life of
Rhythm & Hues octopus. The gag is that
the octopus is having trouble breathing,
Gordon says. So Ben Stiller grabs the painting and heaves water onto the octopus.
Gordon, who had been a CG supervisor
at Mass Illusions for the painted-world sequence in What Dreams May Come, which
won a visual effects Oscar, relied on tool
kits from RE:Vision Effects, whose founders won a technical achievement award for
those tools. Pete Litwinowicz and Pierre
24

June 2009

CafeFX brought several paintings in the National Gallery to life, including this seascape by JMW Turner,
using tools from RE:Vision Effects enhanced with additional features and techniques whose origins
trace back to the optical-flow particle-manipulation technology used for What Dreams May Come.

Jasmin had created products based on the

ideas in the optical-flow particle-manipulation software we used for What Dreams
May Come, Gordon says. In its simplest
form, you can track the motion in a filmed
scene and attach paint strokes to it. And,
you can do a lot of things with the paint
strokes and the way they appear. For example, you can base the paint strokes on
sprites created in Adobes Photoshop from
scanned brush strokes, or use algorithms
to generate the strokes, and rendering to
make the strokes look like paint.
We contracted with RE:Vision to add
some new features to their products, Gordon says. For example, they have a feature
that builds random variation into the brush
strokes, and we wanted to break down the
variation to control how much was huebased, saturation-based, and luminancebased. Thats just one example of many.
A number of the paintings started as liveaction footage that CafeFX alteredmanipulating the photography, as in American Gothic, to stretch and squash the
filmed actors into the shape of the farmers
in the Grant Wood painting. Some paintings were entirely CG, as was the Turner
painting, except for foreground characters
in the last scene. For modeling, CafeFX
used Maya, and for the water, Maya particles with Side Effects Houdini procedures.
And, of course, the RE:Vision tools.
We tried to be faithful to each painting
in every way we could, Gordon says. For
Jackson Pollocks swirling Convergence,
they pushed Maya particles along splines
using a system developed by CG lead Scott
Palleiko. And, CG supervisor Will Nicholson caused Roy Lichtensteins Crying
Girl to wipe her real tears. That was a

combination of CG animation, various

contour renderings, and some compositing tricks supervised by Theresa Rygiel,
Gordon says. It was quite complicated to
make the face look exactly like what we expect Lichtensteins idealized girl to look like
when she turns around.
For most paintings, the artists tried to
reproduce the painters brush strokes in oil,
charcoal, and other mediums, and then
scanned those into Photoshop. Applying
such brush strokes to live-action footage
made it possible to create the illusion that an
actor filmed smashing a bottle is behind the
counter in Edward Hoppers Nighthawks.
When the artists took liberties, it was
in adding motionpouring water from
a seascape and causing clouds in the sky
to move in Turners dramatic London
from Greenwich Park, for example, and
in adding touches like god rays to Albert
Bierstadts Among the Sierra Nevada
Mountains. Bierstadts painting was the
one most like What Dreams May Come in
terms of its setting, Gordon says. Its only
a three-second shot. We put in enough
movement to see it, but not so much it took
away from the foreground. It wasnt always
about what looked right in the painting,
but what looks right in the shot.
With effects ranging from setting into
motion a paintings captured moment in
time to causing Abraham Lincoln to rise
from his marble chair, these studios have
shown, once again, that, in the hands of
artists, computer graphics tools can make
the impossible look real. n
Barbara Robertson is an award-winning writer and a
contributing editor for Computer Graphics World. She
can be reached at [email protected].

26

Gaming

June 2009

Gaming

banned in Germany, ignited a

global firestorm of controversy
upon its release, and was hailed
as one of the most graphically innovative games of recent
years (see Guest Editors Note, pg. 2). It comes from four
of the brightest minds in game development today. Their
combined credits include some of the most successful and
groundbreaking games of our time: Resident Evil, Devil May
Cry, Okami, and Viewtiful Joe. So when legendary developers Atsushi Inaba, Shigenori Nishikawa, Tatsuya Minami,
and Hideki Kamiya joined forces under the banner of the
newly formed Platinum Games, expectations ran high for
the first effort from the four creative powerhouses.
What they unleashed upon the gaming world this spring
left the markets collective jaw on the floor: a visceral, violent,
visually stunning action game for the Nintendo Wii that unfolds like a Frank Miller graphic novel in blood-splattered
black and white. Released to widespread critical acclaim and
controversy surrounding its over-the-top comic-book sadism, MadWorld puts players in the boots of Jack Caymen.
A mechanic and former marine with a retractable chain saw
affixed to his arm, Jack fights his way through a hyper-stylized city where citizens are pitted against one another in a
blood-sport game show called Death Watch, a la Stephen
Kings The Running Man.
The third-person action game is the first of its genre for
the Wii platform, following Jack via a traditional, behind-

the-back point of view. What isnt traditional is the gameplay. Inaba and his team tailored the action, the story, and
Jacks fighting repertoire to remain true to
the physicality of the Wii platform, focusing the gameplay around close-combat, meleetype weapons, and fighting styles, rather than the
more common aiming and firing.
We didnt use the infrared pointer. We
didnt want a game where the player had to carefully point and click, says Inaba. To that end, players
control Jack with the Wii Remote and Nunchuk, using his
hands or whatever object is at hand, including street signs,
garbage cans, and telephone polesto impale, dismember, and so forthfoes in ways befitting the most extreme
Itchy & Scratchy cartoon.

Aesthetic Violence
The game, however, is not just
about killing. Beneath the absurd violence lurks a story
line that is deceptively labyrinthine and full of surprises.
Nishikawa says MadWorlds
development was guided by
two creative objectives. The
first was to create a stark, illustrative, black-and-white graphicJune 2009

27

n n n n

Gaming

MadWorld uses a black-and-white aesthetic in which the games stylized violence plays out. While
visually different, the look challenged the artists in maintaining visual clarity, particularly with the
sweeping camera moves amid detailed imagery.

novel look as a backdrop to emphasize the

violence, represented by red blood. The second key objective was to make the violence
more comical than perversean outlet for
the games beautiful and boldly different
universe, rather than an end in itself.
While Nishikawa admits that the team
drew inspiration from Frank Millers Sin
City, Inaba says they had a broader agenda
of emulating the American comic-book
style, though borrowing from both Japanese and international comic-book artists
to create a visual identity unique in the history of gaming.
The project came from a desire to not
only make something fun and attractive
for the Wii, but to introduce a level of
violence that would make it unique among
titles for the platform, where violence has
been taboo, says Inaba. From the outset,
we were not aiming explicitly for a graphic-novel aesthetic, but for a simpler, more
elegant art style. To accomplish this, we
immediately chose a black-and-white art
style. We were also intent on heavily stylizing the violence, making it humorous with
cartoonish characters, rather than nauseating. That was important to us.
When asked if the hardware limitations
of the Wii led in any way to the stripped28

June 2009

down simplicity of the graphics, Inaba

is quick to protest. Absolutely not. We
wanted to create a powerful visual impact,
a brand-new and wholly original art style,
he says. That was our mandate.

Death Watch
As outlandish as the violence is, it is motivated by an intriguing story line that corkscrews though a roller-coaster ride of twist
and turns. In the games back story, terrorists, called The Organizers, have besieged
the fictional Varrigan City, cutting off the
island transportation and communications
links to the rest of world. After releasing a
virus upon the population, The Organizers announce that any person who kills
another will be inoculated with a vaccine,
thus turning the city into the stage for the
twisted game show Death Watch, complete
with color commentators Howard Buckshot Holmes and Kreese Kreeley calling
all the action.
As soon as Jack enters the game with
the help of a sponsor named Agent XIII,
The Organizersled by the mysterious
Noarealize that Jacks mission is not
merely to win, but to expose the secrets of
Death Watch and bring down the nefarious cabal of high-powered politicians and

pharmaceutical CEOs behind it. First,

he tries to rescue the mayors daughter
trapped within the city, but when he finds
her, she refuses to leave with him. Next,
he saves a doctor named Leo, who was
also trapped within the city but mysteriously managed to obtain the vaccine. The
two become allies, and as Jack comes closer to unravelling the truth and becomes
the darling of both viewers and advertisers
alike, the organizers conclude that they
must eliminate him from the game.
In the final battle with reigning champ
The Black Baron, Agent XIII reveals himself as Lord Gesser, a powerful politician.
He explains that Death Watch was created
to quench mankinds thirst for blood and
was orchestrated by a pharmaceutical company named Springvale. The company lost
billions of dollars in the last presidential
election and saw the virus as a way to recoup its losses. As it turns out, Leos father
was the head of Springvale Pharmaceuticals. And so the plot thickens.
For Inaba, the plot was crucial. We
didnt want the violence to come easily to
the player; we wanted it to be motivated
by something deeper than mindless blood
lust, he says. In the end, we also wanted to
give a message that says no to the violence.
Thats what we tasked our writer with.
Inaba concedes that the plot is reminiscent of The Running Man. What was
more important, from our very first script
meeting, was to show the contrast between
the people who have to fight and survive in
this closed-off space, and the mad people
who exist outside the violence, cheering
it on, enjoying it as a TV program, he
adds. Also, with a game thats so relentlessly exciting and freneticboth visually
and physicallywe were afraid the player
would become exhausted quickly, grow
accustom or even bored with the pace
of the action, and lose the motivation to
play. To prevent this from happening, we
needed a good story to keep their eyes on
the game.

Gaming

Visual Restraint
As awe-inspiring as the graphics may be,
the visual presentation increased the challenge of maintaining visual clarity. In early
stages of the game, disorientation would
set in quickly, especially with the sweeping camera movements, which often led
players grasping at thin air rather than the
golf club or signpost in their midst. In
these early builds, says Inaba, the blackand-white graphics led to considerable
eyestrain. The key to solving the problem
was to control the on-screen movement
and fine-tune the texture maps into subtle
gradations of black and white. This eventually alleviated the eyestrain. Thankfully, we
never had to alter the gameplay or shorten
the time limit on the levels to combat the
problem.
Despite the monochromatic color
scheme, which could potentially flatten out
the images, the industrial and urban environments are sculpted three-dimensionally with complex lighting effects. From
the building-lined streets to the industrial

interiors paneled with computers and walls

of spikes, the environments are bathed in
real-time lighting. Because of the hardware
limitations of the Wii, however, it was too
expensive to employ light maps or extremely complex layered texture schemes for the
environments. Instead, the team used normal maps as often as possible to paint details into the monochromatic sets, because,
according to lead character designer Masaki
Yamanaka, if the geometry gets extremely
complex, with too many vertices, it can take
forever to load in the Wii.
Using normal maps instead of complex
geometry also gave the artists much finer

control over the visibility of the environment and the balance of black and white.
They also tried very hard to implement
real-time reflection mapping, but in the
end, couldnt find any way to stylize it to fit
with their comic-book sensibility. So, we
ditched it, says Yamanaka.
When Jack walks into a well-lit area, his
face blanches with sudden illumination;
entering the darkness, his face blackens as
stark-white highlighting lines accentuate
his chiselled features. To apply the blackand-white art style to Jack Caymen, we created two types of black-and-white textures:
one for well-lit areas, and another for dark
ones, says Yamanaka. Both were created in
Autodesks Softimage.
In fact, the entire project, from modeling to effects animation, was helmed in
Softimage, and driven by a custom game
engine rather than the popular Unreal Engine 3 from Epic. All the environments and
characters are rendered with simple blackand-white shaders. However, one of the
biggest challenges involved maintaining the

n n n n

All about Character

The games relentless carnage was an everpresent challenge to the character modelers. According to Yamanaka, the violence
was the foundation of the games macabre
humor, so the modelers and programmers
worked tirelessly to make the flesh as destructible as possible. Programmers handled the severing of limbs, while modelers
created the remaining stumps and guts in
Softimage.
Creating bold, striking, unique designs
for each of the main characters was crucial
for differentiating them after theyd be rendered monochromatically. We were afraid
they would blend into the background,
says Yamanaka, so we really had to focus on
creating clear designs, especially for the enemies. Using the texture maps to individualize the main charactersJack, Lord Gesser,
Noa, and Leowas a constant struggle that
resulted in almost obsessive-compulsive
finessing right up to the release date.
We would twist the black-and-white
balance, shadows, and shadow textures

Even though the game has a comic-book style, the imagery is three-dimensional and contains
complex lighting effects. Normal maps gave the artists more control over the visibility of the backgrounds and helped them balance the use of black and white.

graphic-novel look in the effects animation,

especially in the smoke and explosions,
since particle animation tools are typically
tailored for lifelike realism.
We studied the look of explosions and
smoke in graphic novels intensely, says
Yamanaka. To prevent the smoke from
looking like a group of particles, we used
a post filter in Softimage and did extensive edge extraction to each smoke-effect
group, eventually creating the comic-book
style smoke shader.

endlessly until we ran out of time, Yamanaka adds. If we had any more time,
Id still be adjusting them!
Using Softimage, modelers created the
main characters and the end-of-level bosses
using approximately 2000 to 5000 polygons, limiting characters of lesser importance
to around 1500. For much of the writhing,
wriggling, and detaching of body parts, the
team used Havok IK, which was also employed for the hard-body destruction. To
handle more extreme effects, such as bodies
June 2009

29

n n n n

Gaming

All the environments and characters are rendered with simple black-and-white textures, while the
effects, such as smoke and explosions, required a post filter in Softimage.

being crushed in the compactor of a garbage

truck, the team wrote original programs.
Making Jacks movements mimic those
of the player and the Wii Nunchuk demanded an extreme degree of flexibility, rotational freedom, and realism in his
animations. With a swiping motion of the
controller, the player can hack through an
opponent with the chainsaw. By twirling
it, the player can swing a foe over Jacks
head before hurling him into an object.
Its a constant challenge, says Yamanaka,
to link the characters movements to the
players so that he responds perfectly.
To meet that challenge, the team created
nearly 2000 animation cycles for all the bipedal main characters, including Jack, in
Softimage. About 60 percent of those animations were motion captured using systems
from Vicon and OptiTrack. To blend those
cycles smoothly, Yamanaka says riggers used
Hermite interpolation, a method of interpolating values between key points to form
smooth curves, in the blending system.
Aside from linking Jack and the gamers
movements, another constant challenge was
maintaining movement in the nonplayer
characters. When the movement stops, the
character dies, as the old animation adage
goes. We had to keep the enemies doing
something all the time, says Yamanaka, be
it making them punch each other, act frightened of something, or just laughing. There
isnt a single moment anywhere in the game
where the enemies are standing totally still.

Effects in Red
Polo with chain saws is how Jack describes the game he finds himself part of.
30

June 2009

From golf clubs to baseball bats, spiked

clubs to lampposts and knives, Inaba and
his team developed Jacks arsenal around
melee weaponry designed for close-quarter
combat. During the conceptual stage, Inaba analogized Death Watch to a baseball
game, wanting the players actions to feel
like those of a batter, which would, in turn,
be organic to the Wii.
To accomplish this, we avoided guns
and the infrared pointer. We wanted the
gameplay to be more intuitive and refreshingly different than other titles for
the Wii, Inaba says. Gun violence is too
common in video games, and worse, its
visually boring. Over-the-top violence was
our mandate, and the only way to achieve
it was through an increasingly creative and
sometimes bizarre set of weapons.
Such weapons, however, are bound to
make things messy. Blood is cheap in Varrigan City; its everywhere, not just as a consequence of violence, but as the linchpin
of the art design and the central metaphor
for the games theme of bloodshed for entertainment. The group created the blood
splatter in Softimage using a combination
of meshes and particle animation. However, because there were so many variations
in the kill moves, the team had an especially difficult time making the trajectory
and movement of the blood flow unique
to each move.
Imbuing the blood with a sense of realistic movement was extremely challenging,
because in a world that is only black, white,
and red, the only tool you have to create
movement is movement, says effects designer Yoshikazu Hiraki. If you want

something to stand out in a monochromatic world, you have to make it move. To

that end, we often had to increase the volume of the particles and the rate at which
theyre duplicated.
For the splashes of blood that stain Jack
and his enemies, artists modeled several
meshes in Softimage, which are then laid
over the character geometry during gameplay. For rain, smoke, dust, fog, wind, and
fire, the artists again used the same system
of particles and polygonal meshes. To create the smoke, fog, and dust, they duplicated the particles, created meshes from
which to emit them, and then added wind
effects to make the particles drift along the
vectors. For the huge plumes of billowing black smoke, they added a silhouette
to each particle aggregate to separate them
from the rest of the smoke cloud.
Programming the blood spray from one
body to another or from one surface to
another was crucial to the games success.
That challenge fell to programmer Masumi Tarukado. If the spray reaches a wall or
trickles to the ground, Tarukados program
uses four collision detections to determine
if the liquid hits the receiving surface. If
all four are affirmative, then the surface is
stained blood red with a texture map. If
blood splashes against another characters
body or if Jack is struck by a backlash of
blood, Tarukados program searches the
characters mesh for the polygons that
would be stained, and then creates UV and
vertex indexes for them, which are then
used to place the blood meshes onto the
character.
With the Wii, it is possible to also have
unique indexes for each vertex or set of UV
coordinates. Therefore, the engine can create the blood mesh using vertex information derived right from the character mesh,
or it can use UV information, which it then
uses to draw the blood mesh using newly
created UV coordinates. Using UV information, however, can cause Z-fighting, in
which the blood mesh can penetrate the

Gaming

characters mesh. By using the vertex information of the characters mesh, however,
the vertices of the blood mesh will be the
same as the characters, which means the
Z-scores are exactly the same, eliminating
any intersection between the two surfaces,
explains Tarukado.
To ensure that the lighting remained consistent with the visual style, Tarukado and
his fellow programmers additionally created
a system by which the artists could quickly
adjust the grayscale gradation on the particles, meshes, and other lighting effects. They
also added blooming effects to enhance the
fireworks and other electrical effects.

Maintaining Control
The games control scheme is a combination of button mashing and wild flourishes
of the Nunchuck that often can send the
camera reeling around the on-screen action. To combat the unruly and disorient-

ing camera, the team included a reset camera feature so players could reset the view
behind Jack. According to Tarukado, the
developer had to keep the players camera
controls simple for two reasons. The first
reason was to help the player maintain the
breakneck pace of the games playability.
The second was the limited numbers of
buttons on the Wii compared to other
platforms. We choose to give the player
control over camera reset functions only after theyve performed an attack maneuver,
rather than completely devolve control to
the player throughout the game, he adds.
The broad, exaggerated, almost Looney Tunes-esque style of animation nearly
precluded the use of ragdoll physics for
the NPCs. Because Jack has so many attack variations and can be almost infinitely
creative in his kills, we had to be equally
creative in crafting the enemys reactions
to the trauma. We spent a lot of time, in

n n n n

the early stages of development, trying to

perfect those reactions using Havok physics, but could never quite attain the right
exaggeration in the movement, so we created most of them manually, says Tarukado. However, you can still see ragdoll
animation in some placeslike when the
enemies are hit by a bat and sent hurling
into a rose bush (wall of spikes).
Whether or not the violence in the game
will negatively impact sales will become apparent in the months to come. Or, perhaps,
the unique graphic-novel look of the title
will attract players who are growing tired of
the same types of first-person shooters that
are currently dominating the market. Either
way, the designers deserve kudos for thinking outside the same-old game box. n
Martin McEachern is an award-winning writer and
contributing editor for Computer Graphics World. He can
be reached at [email protected].

June 2009

31

Trends & Technologies

inally, 10-bit imaging has come of

age. A long time in the making, the
technologys newfound maturity is
sparking interest that extends beyond its past
niches and across a range of professional applications. To what can we attribute 10-bit
technologys rise in popularity? A confluence
of recent developments in both the graphics
and display industries has created a tipping
point, one that is seeing an improved generation of 10-bit technology not only entrench
itself deeper into existing markets, but spill
over into new spaces, as well.
Offer customers a choice between 10 and
8 bits of precision, and with all else equal,
they, of course, will select the higher option.
And that explains the biggest stumbling
block to 10-bit acceptance: The two have
not historically been anywhere near equal,
not when it comes to price or availability.

Medical applications demand an increasing mix

of 2D and 3D images, in both 10-bit grayscale
and 30-bit color. Above, such imagery is being
viewed on a Barco Coronis Fusion 6MP DL
diagnostic display.

The CAD industry can benefit from 10-bit

imaging, giving extreme detail to this car
prototype created by Daniel Simon of CosmicMotors Galaxion, using Autodesks Alias 2010.

32

June 2009

Trends & Technologies

Todays de-facto 8-bit standardyielding 256 shades of grayscale, or 16.7 million

shades of color (256 shades of red, green,
and blue)has proven more than adequate to address the needs of mainstream
applications. The computer industry has
served that majority well, building a broad
set of commodity productsdisplays and
graphics, hardware and softwareall in
support of 8-bit systems.
But not so for 10-bit. Professionals with
the need to push beyond 8 bits have faced
a host of shortcomings in the marketplace,
with options few and prices high. As a result, only a few of the most demanding,
absolutely-must-have users could justify the
deployment of 10-bit grayscale or 30-bit
(3x10-bit) color. The medical imaging community is among this group. Delivering four
times the number of displayable shades as
8-bit data, 10-bit precision improves radiologists speed and, more importantly, the
accuracy in reading high-fidelity scan source
from a range of technologies, including
X-ray, magnetic resonance imaging (MRI),
and computed tomography (CT).
Additional discrete shades can be allocated to fit a narrower luminance range,
dedicating more displayable intensities to
precisely that range where subtle details
are concentrated, often varying by image
and scan technology. The image reader sees
more detail and spends less time twiddling
with windowing, leveling, brightness, and
contrast. The end benefit to the physician,
hospital, and patient: more precise reads
delivered in less time.
While the medical communitys been
the most committed and consistent consumer of 10-bit display technology, its
done so in spite of what the market has
offered, not because of it. Medical installations adopted 10-bit technology because
of the overwhelming importance of getting
an accurate diagnosis quickly, and in pursuit of that goal, it has had to accept more
than a few drawbacks.
Niche markets typically attract niche

solutionsoften just a handful

and usually proprietary. This had
been the case for 10-bit imaging,
where proprietary solutions led to
high prices, few standards, and
little in the way of interoperability. Avoiding a single-source
supply often was simply not
possible; to piece together a
working end-to-end solution
was taxing enough, let alone
to attempt something heterogeneous. Being locked into
a proprietary solution translated into long
product cycles, yet it came with no guarantees of long-term support. And should
the desire ever arise to take advantage of
attractive technology advancements developing beyond the bounds of the current
implementation (think of the explosion in
capabilities and performance of 3D graphics hardware), well, they were simply out of
reach for the 10-bit user.
Fortunately, the medical imaging community no longer has to go it alone in supporting 10-bit technology, as help is coming from other professional imaging arenas,
including one thats both well funded and
highly motivated. Hollywoods quest for
ever-superior image quality is leading to
the adoption of 30-bit color precision, raising awareness of the technology and stimulating the development of more capable,
interoperable, and cost-effective products
and technologies.

Hollywood Goes 30-bit

The film industry has surely and steadily
been making the move to 100 percent digital workflowsend-to-end and at every
step along the way. A linchpin in realizing
that vision, 30-bit color not only delivers
superior fidelity, but also has been instrumental in achieving that elusive goal of uniform color, from one artists desk to another,
throughout the production pipeline.
Nevertheless, the term film is becoming a misnomer, as that end-to-end digi-

n n n n

Digital mammography has been a driving

force in the push to 10-bit imaging.

tal process now extends all the way to the

theater, with digital movies already being
shown on a new generation of digital projectors replacing stacks of film cans. Now
in use at thousands of theaters worldwide,
digital cinema has arrived and should gradually replace traditional film projection.
To meet or exceed the visual experience
of 35mm or 75mm film, digital cinema
proponents and adopters set a quality target of 4k (4096 pixel)-wide spatial resolution and 10-bit precision per channel. That
target has been met with a range of 30-bit
capable, 4k projectors now on the market,
including Sonys SXRD product line.
To a lesser degree, 30-bit color is making a mark in broadcast markets, as well.
While the transmission of higher-precision
video streams to consumers is still a long
way away (if ever, for a variety of economic
issues rather than technical limitations),
30-bit has found its way into the content
creation side of the industry. Keeping the
content at the highest quality for as long
as possible in the production process
through blending, mixing, and overlays
is always a good thing.

Open Standards
Medical imaging may have been the
cornerstone for 10-bit display precision,
June 2009

33

n n n n

Trends & Technologies

but it is the film industrys adoption that

has spurred development and expansion
of a more cost-effective and interoperable
30-bit color ecosystem. Championed by
high-profile studios, with support from
the computer graphics and display industries, momentum for 30-bit color has been
building along three fronts: the establishment of standards to improve interoperability, the dramatic improvement in the
availability and pricing of 30-bit displays,
and the emergence of comprehensive highprecision support in graphics hardware.
Emerging as broad-based unifying standards for the storage and transmission of
10-bit image data, OpenEXR and DisplayPort, among others, are providing the glue
that allows heterogeneous, multi-vendor
10-bit solutions to take hold. In the absence of a formal 10-bit image format from
the computer industrys de-facto standard,
Windows, Industrial Light & Magics
(ILMs) OpenEXR standard volunteered
to fill the gap. Adopted from the half
data type of Nvidias Cg shading language,
OpenEXRs 16-bit floating-point format
provides high dynamic range in a compact
format, with a 10-bit mantissa to sustain
10-bit fidelity throughout the rendering
and postprocessing workflow.
With 24-bit color as the de-facto standard for most display markets, video in-

terfaces havent formally supported 30-bit

pixel formats. Yes, vendors can and do
build 30-bit display systems around DVI,
yet DVI has never officially specified a 10bit-per-channel format, forcing providers
to specify their own formats.
The emergence of Video Electronics
Standards Associations (VESAs) DisplayPort should improve the status quo.
Showing promise as a unifying video interface not only to replace both VGA and
DVI, but also to provide a cleaner path to
HDMI, DisplayPort formally specifies a 10bit standard format (as well as 12- and 16-bit
formats shared with HDMI 1.3). DVI, on
the other hand, did not. With a breadth of
support across markets, DisplayPort should
provide a lower-cost 10-bit alternative to
SDI (and HD-SDI), an existing standard
(ITU and SMPTE) confined today to
high-end professional video applications.

LCDs, Projectors, and GPUs

Until recently, CRTs ruled the 10-bit display world. Early LCDs couldnt match the
CRTs affordability, nor did they measure up
on critical metrics, such as luminance and
contrast. Yet, just as the LCD has supplanted the CRT in virtually every other arena,
it has done so, too, in the most demanding
professional spaces and for both 10-bit grayscale and 30-bit color applications.
LCD technology, display quality,
and volume manufacturing have
improved to the point where CRTs
are typically no longer considered
for medical imaging applications.
The LCD has even dethroned the
CRT in its role as the de-facto
standard for broadcast-reference
monitors. Barco, a longtime leader
in professional display products,
recently introduced the first LCD
monitor to achieve Grade 1 (EBU),
A 24-inch, color-space calibrated, 30-bit color
LCD monitor from HP, created in partnership with
DreamWorks, makes color differences apparent.

34

June 2009

a set of demanding specifications for brightness, contrast, and blackness.

Similar to its rise in the consumer space,
the LCDs momentum in professional markets stems from its superior form factor and
the dramatic reduction in price over time.
Witness Hewlett-Packards DreamColor
LP2480zx display, a 24-inch, color-critical LCD monitor designed in partnership with DreamWorks Animation. The
LP2480zx not only displays 30-bit color
images (10-bit per channel), but it can do
so with custom presets calibrating the display to the gamuts of seven-color spaces,
including sRGB, ITU Rec. 709, and Adobe RGB.
With that image fidelity now available
at approximately $2000, an LCD monitor
like HPs will not only allow studios to realize that color-consistent, end-to-end digital
vision, but it will open a lot of eyes in spaces that havent historically employed 30-bit
color technology. And its not just the advent of high-quality, 30-bit capable LCD
monitors on the desktop that promises to
change the playing field for higher-precision imaging. Rather, it isperhaps even
more sothe emergence of the high-resolution, 30-bit LCD projector, engineered
to deliver better-than-film quality to the
movie house.
Of course, the display is only half the
equation; 10-bit capabilities also must be
up to snuff in graphics hardware. Fortunately, advancements in GPU technology
in recent years have delivered just that,
starting with the adoption of top-to-bottom 32-bit floating-point pipelines. Working off the 32-bit architectural foundation,
Nvidia and AMD have both extended precision10-, 12-, and 16-bit integer, as well
as floating-point formats like OpenEXRs
through the GPUs back-end paths and out
to the display.
Early support focused on true 3x10-bit
DACs and dual dual-link DVI ports, but
todays GPUs are not only capable of driving 30-bit digital display streams across

Trends & Technologies

Collaborative viewing is a natural fit for high-resolution, 30-bit displays.

DVI to support todays platforms, but

they now comply with the first truly interoperable 10-bit standard, DisplayPort.
And were not talking a few pricey, niche
models, either. Support for 30-bit color
over DVI and DisplayPort is a feature now
spread across the breadth of professional
graphics product lines.

practices, where the light box remains a

common fixture.
Another opportunity opened up by recent improvements in 10-bit display technology is the exploitation of the synergy
between 10-bit precision and cutting-edge
GPU technology. Today, after all, considering an imaging solution with 10-bit display
as the one and only goal is shortsighted.

10-Bit Appeal
All the appeal of recent improvements in
10-bit quality, cost, and interoperability
wont be limited to new markets, as existing ones will reap the same benefits. Consider the potential impact of DisplayPort in
studios, for example. Offered a less-costly,
broad-based alternative to 10-bit SDI, more
customers should be able to justify further
deployment of 10-bit capable systems.
And with the scope of medical imaging
continuing to expand through the spread
of picture archiving and communication
systems (PACS), IT buyers will certainly
be pleased to find their choices wider and
their prices lower, opening opportunities to improve the existing infrastructure.
Hospitals wont necessarily equip twice as
many reading rooms if solution costs are
cut in half, but any dollars saved would
go to good use: to upgrade existing reading stations with multiple high-resolution
displays, thereby allowing more effective
side-by-side analysis, or to extend PACS
and 10-bit display into other corners of the
hospital, including operating rooms.
Beyond hospital borders, a wider range
of economical, 10-bit digital imaging solutions will mean that more will find their
way into a physicians individual or group

Real-Time Rendering
Keep in mind that 10-bit precision (and
full 30-bit color) is just one of many features afforded by todays professional
GPUs. Select Nvidias Quadro FX or
AMDs ATI FirePro, for example, and you
not only get the full 30-bit color precision
(with 10-bit grayscale, to boot), but a highperformance, full-featured, programmable
3D GPUand everything that goes along
with it. Such solutions offer capabilities
that extend not only beyond 10-bit precision, but even beyond conventional, polygonal 3D rendering.
Consider simple 2D pan and zoom,
seemingly trivial graphics operations in this
age of elaborate 3D special effects and immersive gameplay. With scan resolutions today reaching 2k x 2k (and beyond), reading
medical images without fast, hardware-accelerated pan and zoom can waste a lot of valuable time in the reading room. Yet, many of
the proprietary solutions historically offered
to medical applications have lacked even this
most basic of hardware features.
With todays professional GPUs, highspeed 2D zoom and pan come free and
fast, trivial cases of the more complex nonaffine perspective transformations com-

n n n n

mon in 3D rendering. And their onboard,

programmable scaling filters process more
samples with more precise filter responses,
yielding high-quality, high-resolution images in real time.
While display quality is the top priority
for many applications in medical imaging, a
visualization solution that can handle more
than the simple 2D display of 10-bit grayscale has become a must-have. IT buyers
demand maximum return for dollars spent
and are anxious to build a more broad-based,
flexible digital imaging solution, thereby
unifying support for the display and rendering of imaging data generated by a range of
scan technologies in both 2D and 3D.
For example, companies serving medical imaging marketssuch as visualization
and analysis software provider Vital Images
and Mercury Computer Systemshave
exploited the Nvidia Quadro FX cards
used primarily for fast 2D imaging, having
them also serve double duty as real-time
3D volume visualizers.
Similarly, Barco Medical Imaging is utilizing the AMD ATI FirePro cards for fast and
accurate representation of raw CT data and
more. With todays advanced CT scanners
producing scans of 1mm per slice, volumes
can run into the thousands of high-resolution slicesfar too many for a physician
to navigate effectively in 2D. Add into the
mix emerging scan technologies like positron emission tomography (PET) and the
increasing use of 4D visualization (3D volumes varying across time), and high-performance volume visualization has moved
from nicety to that of baseline requirement
for radiology departments.
PACS, 10-bit, 3D rendering, multimodal volume visualization, blending
high-quality grayscale and color: The
technologies go hand in hand with todays
increasingly integrated medical imaging
solutions. By combining best-in-class 3D
GPUs with 10-bit display precision and
innovative software, professional graphics
products seemingly can handle it all.
June 2009

35

Trends & Technologies

lars, any premium demanded by 30-bit color technology is trivial by comparison.

In the CAD realm, design
staff shouldnt expect to see
an across-the-board transition to 30-bit color. With
24-bit color more than adequate for the purposes of
Effective seismic analysis, using data such as this from Halengineering, 30-bit wont
liburton Landmark Software and Services, must detect subtle
have a broad impact, at least
gradient changes and reiterate analysis in high precision.
not anytime soon.
New Spaces
But for CAD stylers, those shaping and
Medical imaging provided the foothold, selling a products look and feel, the story is
and through the promotion of standards different. Theres no compromising on reand wider market opportunities, film pro- alism and fidelity, making 30-bit upgrades
duction has helped push technology and ac- for stylers desks more common, while
cessibility while driving prices down. With presentation rooms and auditoriums are a
the advent of a wider, more mature, cost- natural fit for the new generation of 4k, 30effective base of products and technology, bit projectors.
more application spaces are primed to take
And demo rooms are just one example,
the plunge into the 10-bit precision waters. as digital cinema projectors are finding
Which ones? Well, with fewer limita- appeal well beyond the theater. Consider
tions to accept, and fewer dollars required, wall and mega-display installations, such
the technology has truly become open game as reality centers, simulators, and CAVE
for any application that simply cannot com- environments, in applications like aeropromise on display precision. Yet, a few can- space, geoscience, and academia. For such
didates particularly stand out: oil and gas immersive display applications, one 4k
exploration, CAD, collaborative viewing, projector can replace several lower-resoludigital content creators (small to midsize), tion projectors, eliminating the hassle and
artifacts of edge blending and refresh synand surveillance/computer vision.
Sharing the medical communitys need chronization. Complementing the 10-bit
for fast and accurate image analysis is the capable display with 10-bit graphics hardoil and gas industry. Collecting 16 bits of ware then becomes a very appealingand
data per seismic sample is now the norm, affordablenext step.
Big-budget Hollywood media producproviding the critical precision to allow detection and refinement of the subtle gradi- tion got the ball rolling. But 30-bit colors
ent changes that might signal a rich find appeal in DCC has not been limited by
and to more effectively filter out geologic the size of the studio, but by the size of
anomalies leading to dead ends. With the the budget. Image quality and consistency
interpreters eye being a critical link in the are paramount. Offer smaller and midsize
iterative process of isolating areas of inter- DCC businesses a financially feasible 30est, refining geometry, tweaking filters, and bit option, and theyll make the move, as
re-visualizing the display of 16-bit volume well. Graphic artists, ad agencies, videogdata with 10-bit precision represents a logi- raphers, small businesses, or sole proprical and valuable progression. In an indus- etorsall are potential 30-bit users, and
try where misinterpretation of image data that prospect hasnt been lost on vendors
could mean tens of millions in wasted dol- like HP. Looking beyond Hollywood bor-

n n n n

ders, HP, for one, is seeing the aggressive

pricing of its new 30-bit LCD monitor as a
game-changer: a courier killer, eliminating the costly practice of producing and
shuttling high-quality prints to clients, or
even as a new tool for the weekend wedding photographer.
Additionally, 10-bit precision has been
incorporated into the baseline feature set for
most image sensors, A/D converters, and
DSPs targeting surveillance, security, and
other complex image analysis applications.
And as in oil and gas exploration, complex
image analysis with 10-bit precision will
mean quicker positive identifications and
fewer false ones.

Next Step
While the technology is not for everyone,
it is a natural step for demanding professionals. Yet, 10-bit display precision isnt in
the foreseeable future for most mainstream
applications. After all, 8-bit-per-channel
color is pervasive, cheap, and entrenched,
and the eyes ability to discern shades isnt
getting better. So, theres little doubt that
moving from 8-bit to 10-bit in fact offers
diminishing returns.
But for applications that cant compromise on image quality, any return is valuable, diminishing or not. With historical
stumbling blocks withering away, 10-bit
technology has moved beyond the radiologists digital light box and into higherprofile spaces, like digital content creation
and visualization.
Costs are down, and choices and compatibility are up. Presented with improved
technologies, engineered into a wider range
of affordable, interoperable hardware, more
professionals in more spaces will soon be
sampling the benefits of 10-bit precision. n
Alex Herrera is a senior analyst with Jon Peddie Research
and author of the JPR Workstation Report series reference guide for navigating the markets and technologies
for todays workstations and professional graphics solutions. He can be reached at [email protected].
June 2009

37

Digital SetsVFX

Visual effects studios create CG explosions and virtual

environments for Angels & Demons

he clock is ticking on an unstoppable Illuminati time

bomb hidden somewhere in Rome. Only one man, Harvard religious expert Robert Langdon, can distinguish angels from demons and find the bomb in time to prevent an
attack on the Catholic Church. To do so, Langdon must get
inside the Vaticans archives to discover the path to illumination. But, the Church does not hold the questioning symbologist dear to its heart.

38

June 2009

The Catholic Church didnt hold dear the filmmakers who

had also made the movie The Da Vinci Code, either. Based on
the controversial novel by Dan Brown, the 2006 film directed
by Ron Howard had sparked harsh criticism from the Roman
Catholic Church before its release. Thus, well before postproduction had wrapped on that film, Angus Bickerton, visual effects supervisor, began preparing for Columbia Pictures
Angels & Demons, Browns prequel turned into a sequel.

Images 2009 Columbia Pictures Industries, Inc.

By Barbara Robertson

Images 2009 Columbia Pictures Industries, Inc.

Digital SetsVFX

Bickerton says, As we wrapped up, Todd

Hallowell [executive producer, second unit
director] said there was every chance wed
be doing a sequel, so because we could
have problems, maybe we should photograph the key locations in Rome whilst we
were all still together.
For The Da Vinci Code, the visual effects
studio Rainmaker (now CIS-Vancouver)
had circumvented the Churchs restrictions
and created the interior of Saint-Sulpice by
projecting photographs of the interior onto
low-res geometry, and then placing actors
filmed on a partial set into the virtual environment (see On Holy Ground, July
2006). Todd liked that methodology,
Bickerton says, explaining that the Church
had not given permission to film inside
Saint-Sulpice. Using photographs of the
actual location worked really well. We had
a partial set, but 85 percent of the church
was the virtual environment.
With this in mind, Hallowell and a photographer traveled to Rome, where most of
the sequel takes place, and did the Angels
& Demons tour, as Bickerton puts it. Although none of the post houses used those
photos for textures or photogrammetry,
the images served as reference material.
Ultimately, each studio working on the
film essentially approached the project in
the same way, Bickerton says. They rephotographed the environments and derived geometry from online sources and
from the photographs themselves. To do
so, teams from the visual effects studios
joined the throngs of tourists and shot still
images with digital cameras. Although the
Church wasnt as hostile to the sequel as
the crew had feared, the filmmakers still
couldnt shoot actors inside the churches,
nor could the VFX houses take laser scans
of most locations.
We didnt pretend to be tourists, we
had tripods, but it certainly wasnt a proper
setup for a shoot, says Graham Jack, CG
supervisor at Double Negative (dNeg).
Bickerton selected four visual effects

n n n n

houses to work on the film, the same stu- like it, Bickerton notes. They like it low
dios he had worked with on The Da Vinci res and rough for cutting because they have
Code: Double Negative, CIS-Vancouver, somewhere to go, and because they can say
the Moving Picture Company (MPC), and its a work in progress when they show it
the Senate. After seeing Cloverfield partic- to the studio. But what these temps do is
ularly, we decided dNeg was the best op- give them a picture of what Tom Hanks
tion for our big explosion sequence, Bick- is standing in front of, which affects the
erton says. Once we had chosen them as dynamic of the shot and lets them cut for
our primary company, we based ourselves picture content.
in London.
Double Negative
also created the exteriors of St. Peters Basilica and St. Peters
Square, the crowds
that fill the square,
a CG helicopter,
daytime establishing
shots, and exteriors
for the funeral shots.
CIS-Vancouver built
CIS-Vancouver extended the set for Santa Maria della Vittoria here and
the basilica interior,
in the image on the previous page, and composited fire into the image
the fire sequence,
above using photographed elements and CG fire.
and the interior of
Santa Maria della Vittoria and Santa Maria Cataclysmic Effects
del Popolo. MPC worked on shots in Piazza At Double Negative, Ryan Cook led the
Navona, inside CERN, and the antimatter visual effects team, which created approxisequences. And, the Senate handled shots mately 200 shots and graded another 70.
for the Pantheon and inside an underground They had the biggest environment and
the biggest workload, Bickerton says. To
Vatican vault.
In addition to the four post houses, build St. Peters Square in the Vatican for
Bickerton worked during production with that sequence, dNeg modelers working in
compositor Duncan Kinnaird. Wed send Autodesks Maya and Pixologics ZBrush
him shots from LA, and hed comp them created the 284 columns in Berninis monuovernight with (Adobe) After Effects, using mental colonnade topped with 140 statues
backgrounds we shot or temporary stills, of saints that encircle the huge, elliptical piazza, two fountains, the 83-foot-tall obelisk,
Bickerton says.
Even though Kinnaird used 2D stills for the statues of St. Peter and St. Paul, and the
the composites rather than the 3D back- exterior of St. Peters Basilica. (CIS-Vancougrounds the effects houses would eventual- ver built the basilica interiors.) In the film,
ly create, Bickerton found them useful. It thousands of people swarm into St. Peters
was the first chance to see how the sequence Square to wait for the white smoke that anworked and get responses from Ron [How- nounces the election of a new pope.
We photo-modeled what we could from
ard], he explains. And the rough composour photographs using dnPhotofit, our
ites were perfect for Howard.
We were quite a low-tech cutting room, photogrammetric software, and we relied
cutting at NTSC resolution in the Avid on tourist photographs and aerial surveys,
because thats the way Ron and the editors says Jack. For textures, we have software
June 2009

39

n n n n

Digital SetsVFX

Double Negative extended the set built for

St. Peters Square (top) using 3D models
and projected textures. Artists then filled the
set with crowds they created and animated
using two particle-based techniques for the
shots in the final sequences (bottom).

originally built for Batman that automatically stitches the bracketed exposures and
tiles that we shot with long lenses into panoramas. When we didnt have enough detail
for the baroque architecture and statues, we
relied on our texture painters and sculptors.
The bigger problem was lighting the
square for day and night. At night, lights
point up to the top of the colonnade, floodlights point down, and numerous lights illuminate the basilica faade. It was the most
complex lighting weve done, says Jack.
DNeg uses Pixars RenderMan, and to
help optimize the shadow generation for
the hundreds of light sources, the team rendered multiple passes using RenderMans
point-cloud occlusion. We rendered the
floor, the colonnade, and the basilica into
different point clouds, Cook explains.
And then to calculate the ambient occlusion and color bleeding, we combined
all the point clouds into one using a little
command-line tool that read in the multiple passes after we baked them. So, if the
lights had to change in one area, we needed
to re-calculate only that layer.
40

June 2009

To create the crowds in the square, the Shake, Cook says. Our fluid solver allows
team applied two techniques. Particle-based you to compress or expand, so we could suck
sprites created with Maya and RenderMan the explosion back to a point.
put extras, shot on a greenscreen stage,
into the background. Artists placed these Path of Illumination
sprite-based crowds by painting areas on Except at the very beginning, the entire
the ground plane. When people needed to movie takes place one night in Rome, with
move in particular waysespecially when Langdon trying to deduce where the Ilthe shock wave from the explosions hits luminati would execute four kidnapped
the crowdthe crew used a solution built cardinals, one after another. Each locaaround Side Effects Houdinis particle sys- tion represented a primordial element, and
CIS-Vancouver helped create the interiors
tem and its channel operators (CHOPs).
Every particle has a metaball associ- of two: The Chigi Chapel in the church of
ated with it, Cook says. They move based Santa Maria del Popolo, where Langdon
mainly on how crowded the area is around arrived too late to save a cardinal branded
a particle; the system can find the nearest with an Earth symbol from death, and the
particle and plot a course to avoid other church of Santa Maria della Vittoria, where
particles. Because the system knows
the position of all the particles, it
works out, based on velocity, how
to blend motions and trigger actions. It matches the size of a step to
the distance traveled, and blends between walk speeds. Motion capture
provided the animation clips, and a
RenderMan skinning tool provided
bodies for people.
The explosion is a combination
The cataclysmic and climactic explosion starts with
pyro footage, changes to a fluid-dynamics simulation
of practical effects and a fluid-dythat contracts to a point and then, still using fluid
namics simulation created at dNeg.
dynamics, expands back out.
We wanted a believable explosion
in a style that echoes things people have he arrived too late to save another cardiseen, but something that no one has ever nal from immolation. Actors were filmed
seen before, Bickerton says.
inside detailed sets for both these church
At first, the dNeg team created nuclear scenes. CIS extended the sets and created
explosions using Squirt, the studios in- the fire that burned the second cardinal,
house fluid-dynamics solver, but Howard using a mix of practical and CG elements.
felt that was too cataclysmic. We changed The majority of CISs work, though, cengears to develop something more astro- tered on the enormous St. Peters Basilica.
nomical and bathed in light, Jack says.
CIS used NewTeks LightWave to exThe resulting explosion starts with pyro tend the Santa Maria del Popolo set, but
footage, then becomes a fluid-dynamics built everything else, including St. Peters,
simulation that contracts to a point and with Maya. They rendered the models with
expands back out.
Mental Images Mental Ray, and composTo do the final effect, we used a combi- ited the shots with Shake. Visual effects sunation of fluids rendered in our volume ren- pervisor Mark Breakspear led that studios
derer, procedural effects, and matte-painting work, as he had for The Da Vinci Code.
elements, all composited together in (Apples)
To prepare for the studios work, Break-

Digital SetsVFX

spear and his team took nearly 10,000 photos in Rome using Canon EOS 22-megapixel cameras equipped with 8g Flash cards.
We used those to build our CG models,
he says. They became the basis for previs,
and we improved and improved them until
we could use them in the movie.
The previs helped Howard with pacing, and DP Salvatore Totino with framing for shots in the huge expanse. On set
the filmmakers had surrounded a replica of
the base of Berninis baldacchino, the 90foot-tall bronze canopy located beneath
St. Peters dome, and a section of the floor
with greenscreen. CIS built the rest of the
baldacchino, the 138-foot-diameter dome
(which rises 452 feet above the street), and
the entire 694-foot-long interior in CG.
The building, designed in large part by
Michelangelo, covers 5.7 acres.
The previs renders were primitive, of
course, says Karen Ansel, CG supervisor. But, our model duplicated the entire
length, so it helped Sal [Totino] know what

CIS-Vancouver rendered the interior of St.

Peters Basilica using as many as 20 layers
to create and more easily change any mood
without weighty re-rendering, whether in
daylight (top) or at night. At bottom, the floor
is a 3D model with detailed shaders and
texture maps.

n n n n

chapel the actors might

be going past.
Once the crew got
rough cuts back from
shots for the basilica sequences, they knew what
theyd need to render
and what textures theyd
need. We went back to
Rome knowing the camera angles, Breakspear
says. We could stand
On set was a replica of the base of Berninis bronze baldacchino that
rises
90 feet from the floor, and a section of the floor. CIS-Vancouver
almost in the same spot
built the rest of the baldacchino and the entire 694-foot-long interior
and take all the pictures
of St. Peters Basilica using CG models and projected textures.
we needed, so it was a
very efficient process. Our technique is that texture maps to create the aged tile inlays
you work out where the camera is and take and to reflect the baldacchino, itself a fully
pictures from there, from the end position detailed and shaded 3D model. And, light
and all the way along the in-between bits so beams created with volumetric lighting
the texture can adjust for different camera and dust motes added to the atmosphere.
Technically and aesthetically, we wanted
movements.
Back at CIS, artists readied the photo- to support the suspense in the story, Ansel
graphs for projection onto low-res geometry says. And, we wanted to keep the basilica
by painting out the tourists and bringing the as inspiring in the film as it is in real life.
Similarly, MPC used projected textures
images back to neutral lighting. When you
take HDRIs, if you have a shot with strong to create shots of the actors in the Pasetto,
sunlight hitting a column, you can bring a secret passageway disguised as a Roman
back the detail using one of the other ex- aqueduct. We were allowed on the Paposures, Breakspear says. Its a painstaking setto, so we photographed it using a moprocess. But, the meat of our work is in refin- tion-control tessellator head and four digiing those textures to be beautiful and crisp. tal still cameras on tripods, Bickerton says.
The lighters used three setups in Maya MPC built limited geometry to project
and Mental Ray to render for daytime se- that imagery onto. And, the Senate used
quences, dark and somber night scenes, and photo-tiling for the Pantheon, which has
nighttime action sequences. In addition some fully CG shots.
to the regular ambient occlusion, shadow
Hopefully, Bickerton says, the effects
passes, depth maps, and specularity layers, in this film are as invisible as possible. My
we rendered passes with 3D lighting that aim was to be photographically real.
worked with the projection maps to creAnd it is. After all, even though they had
ate lighting effects, Ansel says. That meant to film many of the scenes on soundstages in
the lighters could create areas of dark and LA, they did shoot the film in Rome in a way,
light, cool and warm, and the falloff with- albeit with still photos turned into photoreal
out weighty re-rendering. And, if the look locations, thanks to the science of computer
changed, the crew could re-render any of graphics and the skills of CG artists. n
the 16 to 20 layers sent to compositing.
In addition to the projected textures, Barbara Robertson is an award-winning writer and a
CIS artists modeled the basilicas marble contributing editor for Computer Graphics World. She
floor in 3D using detailed shaders and can be reached at [email protected].
June 2009

41

Broadcast

A new breed of superheroes

fights evil with a digital twist
By Debra Kaufman

Aside from unique characters, which were created in CG, The Middleman contained a number of
scenes that called for actors to be filmed on greenscreen.

or those who like quirks and irony

with their superheroes, The Middleman television series was a bracing tonic. Based on the graphic novels by
Javier Grillo-Marxuach and Les McClaine,
the episodes incorporated some unusual
digital imagery into the story lines, which
followed Wendy Watson (played by Natalie Morales), a struggling artist working at a
temp agency. Recruited by the Middleman
(played by Matt Keeslar), she began training
so she could eventually assume the hero role.
The two crime-fighters get their orders from
Ida, an android disguised as a librarian, who
42

Grillo-Marxuach, who was The Middlemans show-runner and executive producer/writer, created the concept while
he was working as a writer on the
television series Charmed. He turned it
into a graphic novel (in July 2005) that
acquired a cult status among fans,
explains Lebed. Through ABC Family
Channel, he was able to realize it into a TV
series, albeit one that was short lived.
Charmed producer John Pare acted as
a middleman of sorts between Mechnology and Grillo-Marxuach. Our first show
was Charmed, and because of that [work],
we incorporated the company about four
years ago, says Lebed, who has worked as
a model builder and digital VFX creator
for Image G, Digital Magic, and Encore.
When The Middleman came up, [Pare]
called me and asked if I was available, and
he got us involved.

June 2009

represented a secret organization.

In the course of their adventures, which
were told in 12 episodes for ABC Family
Channel, Wendy and the Middleman battled
a strange group of villains, including a terracotta warrior and a winged Peruvian pike
fish. Of course, they always saved the day.
It was a cool show aimed at the ComicCon crowd and was tailored towards those
who like action films and comic books, says
Stephen Lebed, visual effects supervisor at
Mechnology, which handled the VFX. Lebed, along with Chip Potter, co-founded the
Burbank, California, studio a few years ago.

Mech Tools
Comic books and graphic novels are full of
visual effectsas is evident from the spate
of comic-book properties that have been
transformed into popular feature-film fodder. And despite its debut on the small
screen versus the silver screen, The Middleman was no exception. Although several
of the 12 episodes had 30 to 40 visual effectsa standard number for a show like
this onethere were a couple of episodes
that reached 80 visual effects.
With this show, we were limited by
budget and time constraints in terms of

Broadcast

how much we could do, says Lebed.

With 80 effects, we tried to keep them as
simple as possible so we could turn them
around quickly. Whats more, The Middleman was shot in high definition.
Mechnology has a core of eight artists
but ramped up to as many as 20 to handle
the shows quick turnaround. The studios
Mac-based pipeline consists of Autodesks
3ds Max, Maya, Flame, and, occasionally,
Smoke; Apples Shake; and Adobes After Effects and Photoshop.
The renderfarm, meanwhile, is PCbased, with dual AMD boxes, with Nvidias Mental Ray software used for rendering
and Autodesks Back Burn as the render
controller. That allowed us to change the
priority of the renders, says Derek Zavada,
VFX producer, about Back Burner. We
had several jobs on the farm, but sometimes something would happen to change
the priority, and we would go in there and
easily change it.
Mechnologys main file server is an 8tb
Dell running Linux; the VFX company also
uses a proprietary asset management system
(a Web-based setup that uses Mayas QL
and Adobes ColdFusion, formerly from
Dreamweaver) for logging in/out and tracking footage. It has a color-coding system to
let us know the status of each shot, says
Zavada. Every artist can see what shots
theyre assigned to.
Another tool that helped the Mechnology group throughout production was
Iridas FrameCycler. This was the easiest
way to view the frames and check the final
output, says Zavada. Sometimes you get

n n n n

The scene depicted above was filmed at a power plant. Left shows the film plate, while right
contains a 3D monorail composited at the top of the shot, added by Mechnology.

a bad render, and this was a good way to

check for that.

Adding Effects
For Lebed and Potter, each episode began
by perusing the script. The fun part was
breaking down the script and figuring out
how to create the VFX, says Lebed. They
knew how to challenge us. We had five
days or less to turn around each episode. It
was a wonderful experience. Grillo-Marxuach, who is knowledgeable about the
canon of sci-fi movies and TV, paid homage to the genre in each episode to a scene
or themesmall inside jokes for TV-savvy
geeks and sci-fi fans who watch the show.
For example, We got to create a 1950sstyle flying saucer, says Potter. They shot
it out at Vasquez Rocks (near Los Angeles),
where Star Trek and The Twilight Zone has
filmed. Theres a specific rock formation
that you see in every sci-fi show, and, of
course, thats where our CG saucer lands.
The CG door opened, and the aliens
walked out on a little plywood ramp that
we later turned into a ramp that came out
of the saucer.
Another action-packed scene had Wendy
ejected from a jet in midair. They shot
Wendy greenscreen in an ejector seat, and
we created the 3D jet that left the frame as
she is being ejected, Potter continues. That
was homage to Bruce Willis Die Hard.
Any superhero has his or her supertoys,
and The Middleman was no exception.
He has guns, his secret office, his Middle-

man car, and other devices, points out

Potter. Its Men in Black meets Inspector
Gadget meets Get Smart. There were a lot
of things going on. The Middle-mobile
was based on a Smart car, which, of course,
had special powers created by Mechnology.
When he pushed a button, a jet engine
bursts out, and the vehicle goes down the
road a billion miles an hour, says Potter.
In another episode, an alien posing as a
precocious girl rode a bicycle that also transitioned to warp speed when she rang the
bikes bell.
Some of the effects helped out a typically crunched TV production budget. For
example, the Middlemans high-tech headquarters featured a bank of 20 monitors that
played back everything from surveillance
video to newsreels. The production couldnt
afford to use real monitors, so they created
plastic ones instead, and relied on Mechnology to shoot the footage that played in the
monitors and composite it into the monitors blank screens. Doing it that way also
made the production much more flexible,
says Lebed. Ordinarily youd have to shoot
the footage that played in the monitors
before you shot the actual scene. This way,
we were able to shoot that footage whenever we needed to and add it in later. For a
seven-day schedule with so many locations
and stunts, this helped a lot.
Not too long ago, all-digital characters
were too complex and expensive to create
for anything but high-end feature films.
Now, even with the tight turnaround,
June 2009

43

n n n n

Broadcast

smaller-budget TV shows, like The Middleman, can feature digital creatures galore.
We had a terra-cotta warrior who came to
life when a terra-cotta roof broke and the
pieces formed into the warrior, who then
transitioned into a live actor, explains
Lebed, who says the effect was created with
Autodesk 3ds Maxs Particle Flow.
That was challenging mainly because all
the pieces had to swim around in a way that
allowed us to control them and tweak them
to suit the needs of production, Lebed says.
Once they formed into the warrior, there
was a moment where the digital warrior
opened his arms and struck a pose.
Another CG character was the toothed
Peruvian flying pike, a fish with wings that
chased after the protagonists. The original
idea was to use a practical fish. An effects company built a fish puppet, with the idea that a
puppeteer would move the fish throughout
the scene and Mechnology would paint out
the stick. It didnt quite work out the way
they wanted, says Lebed. We had already
done a couple of CG shots of the fish, and
we ended up doing nearly all the shots. Only
one shot ended up being the puppet.
According to Lebed, the challenge was to
get the CG fish to match the environment.
One shot had the fish under a shade tree
with mottled lighting, he explains. We
shot a lot of lighting reference on set and
spent a lot of time with lighting and shaders so it had the same reflective and translucent qualities of the puppet.
Ida, the cranky librarian, is played by actress Mary Pat Gleason, but according to the
story, is supposed to be an android, leaving
44

June 2009

Working on a tight TV schedule, Mechnology had to analyze the shots and come up with the most
effective solutions to make them work. Sometimes that meant using a prop augmented with CG.

Mechnology to create her robotic abilities.

Mechnology also helped build the kind
of immense sets ordinarily found in blockbuster feature films. On stage, the production created the bare bones of a Mayan
temple. They had a big chair where the
Mayan king sat, and [it was] built up with
sand, recalls Potter. The camera started at
the top of the stage, then came down and
did a slow reveal to get our audience into
the temple. We built the rest of the temple
in 3D and tracked it in there.
It wouldnt have been a sci-fi show without the parallel-universe episode, and The
Middleman put effort into theirs. We have
two Wendys, the good one and the evil one,
says Potter. They met up and exchanged
dialog, so there was quite a bit of split screen
and greenscreen so they could do that. The
scenes were shot on location in an immense
power plant. Mechnology turned that into an
Orwellian futuristic world, adding, among
other touches, a 3D monorail system.

Its a Wrap
At the end of the 12 episodes, says Zavada, the crew worked on 2.2tb of data,
for a total of 33 minutes of visual effects.
But it wasnt the quantity of the effects that
proved to be the biggest challenge. It was
the schedule.
In any episode, we had a 10-day turnaround, so the challenge was to build a
concept that we could execute in those 10
days and that, obviously, the production
company, show-runners, and network all

liked, says Potter. If we were going to

build a monorail, a fish, a gun ray, or spaceship, we had to come up with those designs
and have them approved quickly. But we
often didnt have much lead time. We knew
that wed be building a fish for Episode 4,
for example, but we were so busy building
Episode 3 that we didnt have the time.
The approval process became more
streamlined as the first episodes were
turned around. But the writers would see
what we were doing and up the ante the
next week, says Potter. Theyd say, These
guys did eight shots of a fish, so next week
they can do a 3D car with a jet engine on
top of it and three shots of an explosion.
Nevertheless, both Lebed and Potter enthuse over working with Grillo-Marxuach.
Hed come up with a concept in the script,
and if he liked the direction we were going,
hed just let us keep going, explains Potter.
Adds Lebed, This was the most fun Ive
had while working on a show. There was a
great crew, the producers were wonderful,
and every script was so well written that I
couldnt wait to grab the next one and dive
right into it.
With such enthusiasm, its no wonder
that Mechnology successfully completed
its weekly missions on The Middleman.
Unfortunately, this unique series was not
super enough for the network, which
cancelled it after its initial season. n
Debra Kaufman is a freelance writer in the entertainment
industry. She can be reached at [email protected].

Careers

nimators are continually striving for creative excellence as

they explore various story narratives, ideas, and assignments,
and observe the world around them. In order to stay fresh,
inspired, and motivated, they approach each new challenge as an
opportunity to improve their art and learn and grow.
That is the beauty and challenge of becoming an animator and
what excites and motivates most of the professional animators surveyed for the white paper Behind the Animator (www.animationmentor.com/report) from Animation Mentor, an online animation
school. But how can animators grow as artists in this exciting profession? Here, two highly skilled professional animators draw on their
years of industry experience to answer that question.
According to Rob Coleman, animation director on
the last three Star Wars prequels at Industrial Light &
Magic (ILM), the real art of animation is the performance of the characters. Animators are actors, and
they are most successful when they can use great acting
to tell us stories we care about in their animation, says
Coleman, who also directed five epi- sodes and was
an animation consultant for the
first 15 episodes of Star Wars:
The Clone Wars TV series for
Lucasfilm Animation. Currently, he
is in development on the animated feature The
Fourth Magi at Lightstream Pictures.
A lot of people think the Holy Grail of animation is creating
Beowulf
a photorealistic digital human (like the characters in Beowulf),
continues Coleman. However, I think the real goal of animation is creating a fantastic reality that touches on real life
but integrates characters, settings, and action that you
could never create in a live-action film. People want to
be swept away and entertained.
So how do you create an extraordinary world and
draw people into a fictionalized reality where they
can suspend their disbelief and experience
something entirely new? Coleman suggests
becoming a better observer of the world.
Characters are exaggerated versions
of humans. Learn to watch and

understand emotions, he advises. See how someones face looks

and what their eyes do when they say one thing but mean another.
Figure out what that person is thinking based on their facial expression and body language, and then use that in your animations.
Video reference is great way to study acting, but only if you are
watching good actors. If you do video reference of yourself and
your friends, and none of you are good actors, you tend to get
vaudeville animation, whereby everything is badly acted and over
the top, says Coleman. That doesnt make for good animation.
He also cautions animators not to just copy from video reference,
but to use it as a starting point to figure out weight, timing, and
key poses, and then exaggerate it ever so slightly for realistic animation or more purposefully for cartoony
animations.
Animation consultant Frank Gladstone,
whose resume includes A-list animation
facilities such as DreamWorks, Disney,
and Warner Bros., works with studios
and animation schools to train animators
in the artistic fundamentals behind animation.
His recommendation: reading. This will help animators better understand storytelling. He also
suggests studying visual artists, such as painters
and photographers. Watching great films helps
us understand narrative structure and cinematic
storytelling, he adds.
Become a lifelong learner, Gladstone further
advises. Although having a college education is
not absolutely required in order to get a job in this
field, a degree does show employers that you have
the experience, curiosity, and drive necessary to expand
your skills and the discipline to follow through on the
things you learn. To become a skilled animator, you
are going to have to keep learningnot only about
new technologies, but also about art, literature,
music, movies, books, and ideas, he says.
An industry professional advises: If you use video references, be sure that the acting and motions are good.
June 2009

45

n n n n

Careers

That is what will inspire you to create extremely good work and to keep growing as
an artist, Gladstone notes.
According to Coleman, the great Disney
animators figured out a long time ago that
acting and storytelling were the most important parts of animation. Weve kind of
lost touch with that and have a generation
of animators who went right into learning computer technology and never really
learned the fundamental principles of animation that are so well spelled out in Disney Animation: The Illusion of Life, a character animation book by two of Disneys
Nine Old Men.
Gladstone agrees. There is a bit of a production mentality in schools and studios,
where they are churning out people who
are technically proficient, but not artistically proficient, and that is limiting to both
the individual and the art form itself.
At ILM, Coleman would pull the animators away from their computers to act

out scenes instead of thumbnailing them.

If you have a solid 2D background,
thumbnailing is great, but not so much if
you dont have that kind of background,
he says. In my opinion, there is no better
way to figure out a scene than to stand up
and live it with other characters. Youll feel
whats right when you move your body and
shift your weight, and the poses you create
will be right.
To this end, Coleman highly recommends that animators take acting classes,
improv classes, or clowning classes to get
used to performing.
Because animation is a collaborative
endeavor, we are often working on someone elses vision, not our own, comments
Gladstone. We need to find that nugget
that inspires and motivates us to give it the
best that weve got. Sometimes it will be
the story, or the art direction, or the director, or the technology, or the adventure of
working in a new place with new people.

Windows-32/ Windows-64/OS X-32 / OSX-64

As artists, we need to find something positive to connect with and keep us passionate
about the project, he adds.
Indeed, not every project will be the
next blockbuster feature film. However,
even when a project may be less exciting,
it can be a positive experience if you learn
something from it. Nevertheless, Gladstone warns that it is important to trust
your judgment when accepting work. If a
project goes against your personal integrity,
its important to turn it down, he advises.
As a storyteller, its important to remember that people will see what you create and
be influenced by it, Gladstone continues.
If its in poor taste, or will communicate
something you dont agree with, dont do
it. As an artist, its important to feel good
about the work you do. n
Kris Larson writes for AnimationMentor.com, an online
education and mentorship program focused on character
animation.

32-bit only$399

HighPerformance
Camera Tracking

Use SynthEyes for animated critter insertion,xing shaky shots,

virtual set extensions, making 3D movies, architectural previews,
accident reconstruction, virtual product placement,
face and body capture, and more.

See the website for details

of the latest version!
46

June 2009

For additional product news and information, visit CGW.com

focal press
b o o k s

. c

By Chris Georgenes
ISBN: 9780240521312
$39.95

How to Cheat in Adobe Flash CS4

MAC

WIN

Mod Tool
Autodesk has announced Autodesk Softimage Mod Tool 7.5 software, formerly the
Softimage Mod Tool. The 3D modeling and
animation package designed for aspiring

l
a

modeling - AnimAtion

Focal Press helps

youmaintain your
competitive edge.
Buy online
or in your favorite
bookstore today!

Blackmagic Design;
www.blackmagic-design.com

In Studio 5
Toon Boom Animation has released Studio
5, a new version that adds stop-motion
animation to the softwares existing animation capabilities. Positioned as multi-technique animation software, Studio 5 is
designed to enable users to master all
animation techniques. Studio 5, well suited
to educators and students, helps users not
only to become familiar with stop-motion,
traditional, digital, cut-out, and rotoscoping
animation methods, but also to combine

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.

AnimAtion

Converter Utility 1.0

Blackmagic Design is offering its new
Converter Utility 1.0 software for free
download from the companys Web site.
Converter Utility 1.0 enables all Blackmagic Design Mini Converter customers to
upgrade their converter software with new
features. For Windows and Mac OS X platforms, Version 1.0 delivers a high-quality
HD down converter in the Mini Converter
SDI-to-Analog model. The new hardware
down converter enables the conversion of
HD to SD in letterbox, anamorphic 16:9,
and center cut 4:3 modes. At the same
time, Converter Utility adds support for
1080p/50, 1080p/59.94, and 1080p/60
progressive video formats via 3g b/sec
SDI on the Mini Converter SDI-to-HDMI
model. Converter Utility 1.0 also takes
advantage of Mini Converters built-in USB
connection, enabling all video and audio
level adjustments to be made using a software interface and automatically checking
the availability of firmware updates. Users
can set analog component video output
to output HD directly or to down-convert
to SD component output by selecting the
bypass switch.

LINUX

Video

w w .

WIN

MAC

By Markus Montola, Jaakko Stenros, Annika Waern

ISBN: 9780123748539
$49.95

Massive Software;
www.massivesoftware.com

WIN

Nash.

Toon Boom Animation;

www.toonboom.com

Pervasive Games

WIN

Insightful Engineering
Massive Software has unveiled Massive
Insight, which is designed to deliver accurate, real-world visualization and simulation
to architects, planners, civil engineers, and
research scientists. The new software solution is well suited to applications involving
life safety, life sciences, pedestrian planning, transportation
simulation, architectural design optimization, disaster
prevention and
recovery, environmental impact studies, and
consumer behavior research.
Massive Insight employs Artificial Life technology, which draws from
natural processes rather than traditional
simulation methodologies. Its artificial
intelligent-driven agents use simulated
natural senses, such as vision and hearing, to accurately simulate real-world
behavior. The software can be used to
simulate virtually any object or creature,
ranging from planes, trains, and automobiles, to pedestrians, insects, and
microbes. Massive Insight includes a
variety of behaviors, such as urban traffic
patterns, office evacuations, and animal
migrations.

Editor: Andrew Gahan

ISBN: 9780240811475
$69.95

Vis - sim

F O R C R E A T I V E S,
BY CREATIVES

Game Art Complete

SOFTWARE

them for increased creativity. The latest

version incorporates an enhanced user
interface, annotation layers for collaboration, the ability to export animations directly
to YouTube and Facebook, live view/image
capture, Onion Skinning for precise positioning, and optimized image playback.
Toon Boom Studio 5 is now available at a
cost of $330, or $80 as an upgrade.

aZVgcbVhiZgXgZViZ

June 2009

47

Courting Controversy
continued from page 2
violence and conflict have been the
cornerstones of dramatic art, as much as art has been a harmless outlet for violent tendencies. If we were to ban MadWorld,
making it a scapegoat for violent adolescent behavior or inept
parenting, wed have to ban Shakespeares Titus Andronicus,
John Miltons Paradise Lost, Frank Millers Sin City, and nearly
Stephen Kings entire canon of work.
To deny adults or, more importantly, kids that outlet would
be to assume theyre incapable of differentiating fantasy from
reality. A wise seventh-grade teacher once told me, you cant
lock children up in the castle tower; you have to send them out
into the world with a suit of armor. Indeed, with a good suit
of armor, they should have no problem enjoying Inabas mad
MadWorld. n
To comment on this piece or to post your opinion, go to the
CGW forum on www.cgw.com.

NUX

07 QUANTEL eQ
VIDEO EDITING
SOFTWARE
Unexpected lease return Immediately available!
Components include: eQ System,
eQ Control Panel
(including mouse and keyboard)
and 3 Extension Boxes.
Please contact Marlene for details
[email protected]
on this award winning movie, film,
TV show editing software.
Located in: Redondo Beach, CA

800-686-8885

game artists is based on the Autodesk Softimage tool set. Version

7.5 boasts Noesis Interactive learning materials that demonstrate
how to create a character and export it to Microsoft XNA Games
Studio 3.0. Users can download add-ons for content export to
Crytek CryEngine 2, Epic Unreal Engine 3, and Valve Source
game engines. Softimage Mod Tool 7.5 also offers support for the
Windows Vista Business operating system and Realtime Shader
API Version 3.0. Autodesk Softimage Mod Tool 7.5 is free to download for non-commercial game creation and modding. Autodesk
Softimage Mod Tool Pro 7.5, for commercial use, delivers all the
functionality of the Mod Tool, and also enables users to produce
assets for games sold via Microsofts Xbox Live Community
Games. Autodesk Softimage Mod Tool Pro 7.5 is available only
with Microsofts XNA Creators Club Premium membership.

Autodesk; www.autodesk.com

WIN

Plug-in
Continuum Complete 6 AE
Boris FX is shipping Boris Continuum Complete 6 AE (BCC 6
AE), with nearly 200 filters for Adobes After Effects and Premiere
Pro. The latest edition sports 13 new filters that enable 3D vector
graphics extrusion, organic 3D deformation, painting, and image
restoration. The new 3D Objects category includes Extruded Text,
Extruded Spline, Type-On Text, and Layer Deformer filters. New
image restoration filters include DV Fixer, Smooth Tone, and Pixel
Fixer, whereas new image painting filters are Charcoal Sketch,
Pencil Sketch,
Water Color, and
Cartoon Look. BCC
6 AE also delivers
support for After
Effects Camera and
Lighting System, a
single-click custom
preset navigation
tool, and the ability
to import After Effects mask paths, to save and load XML-based
animated presets, and to import/export motion-tracking data. BCC
6 AE is now available for $995. Owners of previous versions of
Boris Continuum Complete AE can upgrade for $295. A 14-day
trial version is available for free download from the companys
Web site.

Boris FX; www.borisfx.com

WIN

MAC

June 2009, Volume 32, Number 6: COMPUTER GRAPHICS WORLD (USPS 665-250) (ISSN-0271-4159) is published monthly (12 issues) by COP Communications, Inc.
Corporate offices: 620 West Elk Avenue, Glendale, CA 91204, Tel: 818-291-1100; FAX: 818-291-1190; Web Address: [email protected]. Periodicals postage paid at
Glendale, CA, 91205 & additional mailing offices. 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 specific 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.

48

June 2009

www.aja.com/kiprotour

Ki Pro Worldwide Roadshow

From Lens to Post.

Ki
AJA Video just announced the revolutionary Ki Pro at NAB 2009,
and now were taking it on the road so you can see it for yourself!
Finally, shoot on the same codec as you edit with, full raster 10-bit Apple
ProRes 422, built natively into Ki Pros stand-alone, portable hardware.

Experience Ki Pro for yourself at one of our US roadshows

Chicago - June 10th

Miami - June 25th

Toronto - June 16th

Vancouver - July 14th

Los Angeles - June 17th

Dallas - July 16th

New York - June 18th

Mexico City - July 21st

Atlanta - June 23rd

Sao Paulo - July 23rd

K i

P r o .

B e c a u s e

i t

m a t t e r s .

Find a city near you,

and register today!
Register at:

www.aja.com/kiprotour

T:8 in
S:7 in

HP recommends Windows Vista Business.

Only in theaters

powered by the brilliant performance of the Intel Xeon processor, Linux

operating system and the billion-color HP DreamColor display.
Learn more about the next-generation design, extreme speed, massive
expandability and the new Intel Xeon processor 5500 series of the
HP Z800 Workstation at hp.com/zworkstations

DreamWorks used the HP xw8600 Workstation in Monsters vs. Aliens. Monsters vs. Aliens & 2009 DreamWorks Animation LLC. All Rights Reserved. Certain
Windows Vista product features require advanced or additional hardware. See http://www.microsoft.com/windowsvista/getready/hardwarereqs.mspx and
http://www.microsoft.com/windowsvista/getready/capable.mspx for details. Windows Vista Upgrade Advisor can help you determine which features of Windows Vista will
run on your computer. To download the tool, visit www.windowsvista.com/upgradeadvisor. Intel, the Intel logo, Xeon, and Xeon Inside are trademarks or registered
trademarks of Intel Corporation in the U.S. and other countries. Copyright 2009 Hewlett-Packard Development Company, L.P. The information contained herein is
subject to change without notice. Simulated images.

T:10.75 in

S:10 in

To create Monsters vs. Aliens, DreamWorks turned to the HP Workstation,