Spectrum: Interactive Media & Online Developer News

29 August 2005
Reported, written and edited by David Duberman
For editorial/subscription inquiries, send spectrum_news [at] yahoo.com
Search the Spectrum archives at http://www.3dlinks.com/spectrum


Editor's notes: In this edition of Spectrum we're pleased to bring you an excellent report on the recent Siggraph conference from longtime contributor Jeffrey Abouaf, plus several news items.

We're taking next week off for the Labor Day holiday, and will return with more news September 12.


Today's Headlines (details below)


THE DIALS & LEVERS OF POWER --Independent Games Festival Seeks Submissions --IGDA Forms Sex SIG

GAMES PEOPLE PLAY --Sony Releases EyeToy: Play 2

HAPPENINGS --IGC ´05 Speakers and Sessions Announced

F.Y.I. --About Spectrum



SIGGRAPH 2005: Do We Get CEUs For This?
By Jeffrey Abouaf

How close did this year's SIGGRAPH conference at the Los Angeles Convention Center come to achieving an exchange of the latest in CGI research, educational developments, trends and products? While veterans today bemoan how intellectual cards are held more closely to proprietary vests (there's plenty of that), I was impressed with how much normally secret information was exposed for comment and evaluation. Perhaps there's less commercial advantage to keeping production secrets nowadays than revealing them to a workforce reaching toward a higher bar. Recruiters continue to complain of a shortage of qualified applicants. Perhaps the CGI revolution is over, prior technology is supplanted, and the open issues are narrowed to pipeline optimization and tips & tricks -- harder to sell than groundbreaking techniques/technologies. Maybe the latest hardware and software news is commonplace, with all drama of a word processor press release. Most likely: If it helps me work better, faster, more easily, I care; if not, I smile and wait for the next revolutionary train. SIGGRAPH '93's offerings are today pretty much commodities. This year's show served mostly the entertainment and educational industries, with special sessions directed at enhancing professional skills.

George Lucas Gave the Keynote

George Lucas addressed the crowd of thousands who had waited over an hour for his presentation. This talk reflected our times, both for the speaker's celebrity (likely a major to this year's conference) and what he had to say. I find a strange bracketing between James Cameron's keynote at my first SIGGRAPH in '93, (where he spoke with convincing optimism of a wide-open potential and creative freedom in the undeveloped digital landscape), and this year's, where digital filmmaking is a done deal.

From the middle of the crowd, I could barely make out the stage, which was small and close to the ground. From ceilings on each half of the split auditorium hung three large screens to complement large ones up front. On each, in bold white letters on black, was the directive “no audio or video recording permitted.” Watching this “Keep-Off-The-Grass," “No-Trespassing” sign for an hour, I pondered how this proprietary prohibition serves SIGGRAPH's nature, i.e. free and open idea exchange among peers. What shattering closely-guarded secrets awaited us? Lucas delivered his talk in a scripted talk-show format, responding unsurprised and directly to the interviewer's questions. There were no questions from the floor. Yet, I enjoyed his observations and am interested in how they will play out.

George Lucas described himself as a storyteller who, in an effort to create on film Kurosawa's “Immaculate Reality," hit a technological barrier and committed his substantial resources to breaking it. He said “I'm not a computer person, I'm a storyteller… who puts his resources into pushing the envelope” and who looks to staff “to do what they do to get it done." Massive understatement aside, I truly wanted to hear what the artist had to say. Two observations: (1) digital cinema is in place and has replaced the old technology. Lucas intends to produce a new TV series with off-the-shelf Sony HD cams and Powerbooks; and (2) for this artist, games will have “arrived” when he can talk to them and they and they can talk back. Lines in the sand.

I attended two exceptional courses that day: a four-hour “Acting and Movement for Animators: Student. Teachers and Professionals” class taught by Ed Hooks, and Tim McLaughlin's “Taxonomy of Digital Creatures: Interpreting Character Design as Computer Graphics Techniques.” Hooks is an exceptional speaker: a performer as teacher and vice-versa. To the SRO crowd he lectured the first two hours on acting basics, and illustrated many in a second-half workshop with help from eager audience volunteers. You can likely find most of the material in his books and online newsletters (http://www.edhooks.com/), but he's best in the actor's medium. Attend his classes if you ever get a chance.

Hooks presented these “basics” as a series of rules, with examples of how they can make or destroy suspension of disbelief. For example, “Thinking tends to lead to conclusions; emotion tends to lead to action.” “An emotion is an automatic value response,” i.e., a thought you have elevated in value so much it is associated with an emotional trigger. Usually the association came from an “adrenal moment” - one intensely pertaining to survival or well being. For instance, hearing male footsteps behind you in an urban alley late at night triggers fear, with the result you quicken your pace. You don't think about what the footsteps mean; you feel fear and act on it. Hooks suggests that an actor's connection with the audience derives from empathy, an emotional connection with the character, as opposed to sympathy, feeling for the character. Connecting the dots, the animated actor must connect emotionally, by tapping into those elevated values. Of course this varies across cultures, genders, ages, experiences, etc. Actors are trained to elicit empathy - animators face the same task.

Next, Hooks distinguished “theatrical reality” from “normal reality” in a definition by Stanislavski: theatrical reality has form as “(1) action in pursuit of (2) an objective while (3) overcoming an obstacle." A scene is a “negotiation." Obstacles, or conflicts, derive from (a) conflict with yourself; (b) conflict with the situation; or (c) conflict with another character. They can overlap. The key is portraying them seamlessly. Facetiously, he illustrated with a first-date scene: boy puts his hand on girl's knee. She removes it. Lastly, Hooks suggested that since a scene presumes action preceding and post, and since theatrical reality is compressed time, you should start your scene somewhere in the middle, not at the beginning.

In the movement portion he posed the idea of “power center." Our personal power is located in and project from a power center (for most of us, at the “Hara," above the pelvis). By repositioning it we can change the character we project: anxiety shows a higher, heady power center; machismo reflects a lower positioning. Compare Woody Allen with Bruce Willis. When two characters meet, a status transaction takes place. Relocating power centers can alter that negotiation. Example: An employer/employee encounter will play differently depending on whose power center is positioned like Woody Allen or Bruce Willis. But it's trickier than that - assuming the encounter was by chance, each character's destination takes precedence over power-center issues, i.e., the purpose of their motions overrides the character form. He then played scene selections from Michael Keaton's 1996 “Multiplicity” (http://www.imdb.com/title/tt0117108/) where Keaton plays himself and three clones, each with different power centers. I need to re-watch the film, and of course, follow Ed Hooks a little further down the rabbit hole.

In “Taxonomy of Digital Creatures….," Tim McLaughlin presented an outline for character design he developed at ILM's creature shop. He presented a decision-tree of design and performance issues to focus creatives on which problems are presented/eliminated as you go down the tree. Not only does this map focus the issues, it offers a breadcrumb trail for the inevitable (dreaded) overnight character redesign. This approach contemplates design in terms of CG techniques at each step, looking first to the aesthetic, then to implementation. The goal is to construct articulated surfaces that articulate and perform together with minimum complexity. A given surface type must deform in a particular way, influence and be influenced by other objects, (e.g., clothing, props), and animate procedurally and dynamically. He discussed the role of concept art: perspective and action poses as well as orthographic ones. Reference materials that suggest surfaces, realistic details, range of motion, weight shifts, scientific acceptability, etc. Sculpted maquettes. The elements were industry standards. But his sequence for addressing these issues is a valuable offering.

The basic questions don't change: “What does it look like?” and “What does it have to do?”. The first subdivides into “Is the style photo real, cartoon, or a hybrid?” and “What's the body type? The surface type?” The performance subdivides to: “Is it live-action? Are there digital sets? Is there interaction with the set?” And so on. I hope his materials are available, and the class transcribed. Or maybe he could just market lunch bags and T-Shirts sporting this outline, as mandatory apparel for digital creature creators.

The Educator's Program

The Educator's Program included panels, papers and demos acknowledging that the technologies have been implemented in many schools, and we have a track record to provide lessons and suggest directions. I attended the panel, “Lessons Learned from Games for Educators.” On the panel were Yasmin Kafai, UCLA, Brian Slater, North Dakota State University, Carrie Heeter, Michigan State University, and Idit Caperton, Mamamedia, Inc. These researchers described their projects where game design served learning in both training and educational scenarios. They explored question of “What features promote learning?” “How does non-game content incorporate into games?” (This turns out to be a moving target.) “How does game-based learning transfer to the real world?”

Idit Caperton discussed the Tactical Language Training System his firm developed for the military under a DARPA grant. There were three game structures: a mission game where the goal was to contact the local leader in a foreign culture to arrange to repair a school; an arcade game to build speaking skills; and a general skill builder game. They tried to build a game that encourages authentic practices without penalty, maintains the right level of challenge, highlights issues often overlooked when soldiers enter a different culture (i.e. local politics), and afterward, getting students to talk with each other in furtherance of an assigned task. This example came closest to a training simulator, where students repeat the game to build proficiency. He stated those who experienced the Arabic languages version substantially shortened their training times before deployment to the Middle East. He presented no data on how well these skills translated to field use.

Until this panel I understood “Serious Games” to refer to using game design in simulation to train professionals, i.e., first-responders in a disaster situation. Half these projects served a broader educational goal, e.g., a game to teach prehistoric history; a game-based course in freshman archeology. These “games” were designed to be played only once--one for a 40-minute class period, the other over an entire semester--and like other texts, not used again by that student. I wonder whether differentiating Serious Games for training versus general education amounts to whether the game is designed for a one-time play experience.

Carrie Heeter found a substantial performance difference across genders when game success was measured by score. By using a strong story, cut-scene overviews, and changing the reward system from points to advancing to the next level, girls engaged and performed at the same level as boys. This came from substantial play-testing.

Yasmin Kafai's research was conducted within http://www.whyville.net, an online community populated daily by around 25,000 boys and girls ages 8-15 who access the site during non-school hours. The members interact through customized avatars, to “explore, express, and exchange." There's even a market for buying and selling custom characters and accessories. To explore community response, the researchers introduced a “Pox Virus” to infect the avatars, slowly and at different rates. In other words, the avatar would exhibit spots, lose speech, and then go down. Then they set up a virtual CDC (Centers for Disease Control) to encourage reporting and follow the epidemic. The pre- and post-virus survey of 400 participants showed they enjoyed being research subjects, and that the virus invasion inspired lots of activity in the chat rooms. Most telling, perhaps, was two weeks following the researcher offering a “virtual vaccine” for sale, a counterfeit vaccine appeared in the marketplace!

Brian Slater described NDSU's Geology Explorer game, a semester course for university freshman. Modeled on a research field camp set on another planet, this is a multi-user virtual environment for education where students learn by doing: first learning to identify rocks, then exploring terrain, and finally creating a geologic representation of the region.

The panel was strong on the success of the learning effects of game play and how the learning can be as effective as in the non-game environment, but called for further study as to how these skills translate to the real world: not skeptical, just no measurable data yet.

Autodesk Master Classes

Autodesk conducted 12 Master Class sessions following its rollout of 3ds Max 8 and new-technology demos. Unfortunately these professional-level sessions conflicted with presentations at the conference proper, including AutoDesk's sponsored sessions on advanced MaxScript and pipeline workflow, which I'd have attended. But the eight classes I attended Tuesday and Wednesday were of great value, and skill-wise, a step above last year's offerings.

Artist and trainer Mike McCarthy offered tips and tricks for Max 7.5's Cloth and Hair add-ons, applying apparel and hair/mustache to a character, skipping the basics to focus the on pitfalls that slow production and suggesting settings (some undocumented) to enhance control and reduce render times.

I was most impressed with Chris Harvey's three classes: Advanced Facial Rigging, Vehicle Rigging, and Bringing a Sea Monster to Life. Harvey, currently at Frantic Films, has a long pedigree from film and game studios, and is not too proud to reflect his influences in this field: Bay Raitt (WETA, http://cube.phlatt.net/home/spiraloid/), Paul Neale (http://www.paulneale.com/), and Jason Osipa (“Stop Staring: Facial Modeling and Animation Done Right” (Sybex, 2003, ISBN: 0782141293).

Harvey's facial rig was the most complex I've seen. Photoreal character faces often look stiff or rubbery, but not lifelike; the solution is to build the structure beneath the skin, but do it simply and in a way that facilitates procedural animation. Harvey modeled large bones for the skull and jaw; used smaller bones that follow certain edge loops, such as eyebrows and mouth, and added another layer of bone-based separated “facial muscles." He incorporated Raitt's “combination-sculpting” principles to capture poses not of phonemes, but of “visemes": expressions based on accurate muscle flexion and extension. To save out poses, he created layers within Max's List Controller. With 50 or so poses captured as layers, he can animate the controller weight values to create, blend, and change expressions. The advantage is low overhead.

He then used Max's Wire Parameters feature to connect these states to a series of spline-based character control objects. He next showed how to make a virtual joystick (a spline control made up of a small circle moveable within a larger rectangle) to which you might wire four face poses at a time. Moving the circle blends between wired expressions. But wait; there's more! He suggested applying another layer of realism using hi-res Normal and Displacement Maps, applying them to the rig with the Morpher material and Morpher modifier, and wiring in these attributes in as well. Now the List Controller blends between realistic expressions and morphs produce wrinkles and facial creases. Most of this animates procedurally with very real facial results.

The Mechanical Rigging class offered similar depth. Here the problem was to build a rig for a car that reads collision between each wheel and the road, incorporates springs and suspension, provides for vehicle roll and wheel slippage, gives tires a “bulge” at the bottom where they contact the ground, and performs all this procedurally. This evolved from a presentation he gave last year.

Harvey's third class, “Bringing a Sea Monster to Life," showed how he (in California) and another artist (on the East Coast) created a film-res detailed sea monster from concept art to final performing character, all in two days. The workflow moved among Photoshop, 3ds Max, and ZBrush, with Normal and Displacement maps applied through Max's Morpher material. The final character was color-adjusted for underwater shots, and featured a combination of rigging types with falloff and blending to generate mostly procedural animation on a minimum of keyframes. Most attendees were reaching to grasp the nuances in Harvey's classes, and at Frantic Films, much of his work is generated/held together with scripts he's written to automate much of these processes. The best news is that Autodesk taped all the Master Classes and will make them available, together with the exercise files in a few weeks.

In MAXScript Secrets Revealed, Borislav "Bobo" Petrov demonstrated how a simple 16-line script could be adapted to display onscreen most properties of any Max object, or to add additional functionality to the interface. Reminds us how much of the tool we non-scripters miss. I particularly enjoyed Bobo's script for generating dust procedurally from a collapsing rock wall. This script created a single Particle Flow system that would dynamically generate particle emitters at the surface points where and as objects collide. Deceptively simple!

Dagan Potter, CG supervisor at The Orphanage, gave a stellar presentation on Photorealistic Environments and Techniques. This class covered generating HDR (High Dynamic Range) images for use as reflection maps in Max and using Camera Mapping and complex projection mapping to create photoreal environments and image-based lighting. If you knew nothing about this subject, you would have been hard-pressed to follow him. But for the initiated, it was one pearl after another, and audience members I spoke to afterward appreciated the specificity, depth, and experience-based insight. All of us were there to learn secrets behind the curtain at top studios, and hear explanations of these techniques as used in impressive commercials for Motorola (where the room collapses into the phone) and BMW (where a photoreal car folds apart). My thanks to Dagan and all the presenters for their generosity with what could be secreted as proprietary. This is why I still come to SIGGRAPH after 12 years.

Brandon Davis, a valued Vf/x artist in the Max community, disappeared from the online forums this past year for military service in Iraq. He returned only a month before SIGGRAPH, and put together his session, Combat Effects Animation; what he referred to as a therapeutic transition back to the industry. He showed a collection of reference material: DV footage of M16's, AK47's, rocket-propelled grenades, and rockets, to demonstrate that the real thing doesn't look the way it's portrayed in the movies. He then showed techniques for using Max to make realistic special effects on that live footage. What struck me most was the stark reality behind the content, what this artist had seen versus what he was sharing. It left a lingering sobriety in sharp contrast to the SIGGRAPH fantasy.

Of course, these case studies were explained in greater depth; I've got a full notebook and the assurance of the DVD record and course files to look forward to. But this year I feel the Autodesk organizers--Natalie Nedjeru, Scott Laughlin, Maureen Higgins, and Kevin Clark--raised the bar, reflecting industry advancements as well as the needs of pros willing to pay and take time away from competing SIGGRAPH presentations. I noticed attendees were more skilled than in prior years, often faced with the same problems as the presenters. The educational staff at Autodesk is on the right track, and I urge them to continue raising the level in future classes.

I've evaluated this SIGGRAPH less as a journalist than as an artist, student and teacher. I appreciated the wealth of information available. The parties were plentiful and the show floor busy, but I found most value in the presentations.

(Editor's note: In case you're wondering what CEU in the title stands for, it's "continuing education unit.")

Jeffrey Abouaf is an artist, teacher, author and analyst, and can be reached via his Website at http://www.ogle.com .



Independent Games Festival Seeks Submissions

The Independent Games Festival (IGF) is now accepting submissions for the 2006 competition, which honors innovation in videogames created by independent game developers and students. This year's festival will feature a $20,000 Grand Prize for the best game, a new modding competition, and the student showcase competition. The deadline for the main IGF Competition is Sept. 6, 2005; the modding competition submissions are due Oct. 10, 2005; and student showcase submissions are due Nov. 15, 2005.

The increasing prevalence of videogame modifications has stimulated the introduction of a mod category in this year's competition, and the chosen modifiable games have now been picked with the help of an audience vote. Therefore, mods can now be submitted for Valve's Half Life 2, BioWare's Neverwinter Nights, Epic Games' Unreal Tournament 2004, and id Software's Doom 3, with a total of $10,000 in prize money for the best original modifications.

"Modding has long been a creative source of new game experiences, and it highlights the vibrant collaboration between leading developers and the communities that build around their games," said Jamil Moledina, director, Game Developers Conference. "The new modding competition is our way of recognizing the importance of modding not only as a gateway to professional game development, but also to stimulate the potential of the accepted modding community."

More than $45,000 in cash prizes will be awarded to IGF competition winners this year for innovation in five major areas: Visual Arts, Audio, Game Design, Technical Excellence, and Best Web Browser Game. Main competition finalists will be eligible for the Seumas McNally Grand Prize for Independent Game of the Year and the Audience Award. The IGF will also be selecting 10 games this year for Student Showcase, including a new category for games created using middleware.

The IGF was established in 1998 to encourage innovation in game development and to recognize the best independent game developers, similar to the way that the Sundance Film Festival has benefited independent filmmakers. The IGF has given away hundreds of thousands of dollars in prizes and major recognition to innovative, independently created videogames from all over the world, from early pioneering titles such as Tread Marks and Shattered Galaxy, to recent stand-outs such as Oasis, Alien Hominid, and Gish.

The IGF takes place annually during the Game Developers Conference (GDC), scheduled for March 20-24, 2006 in San Jose, Calif. The GDC is the largest gathering of videogame creators worldwide. Developers interested in submitting a game can visit www.igf.com for official rules, deadlines, and entry forms.


IGDA Forms Sex SIG

A Sex Special Interest Group (Sex SIG) has formed within the International Game Developers Association (IGDA) to address the challenges and possibilities for adult content within games. The Sex SIG, which has been in the making since the 2005 Game Developers Conference last March, will provide an outlet where information can be shared among developers and those who wish to enter the field.

"The Sex SIG will serve as an arena for developers to address the issues and challenges facing the use of adult sexual content in video games," said Brenda Brathwaite, Sex SIG Chair/Founder and Lead Designer on Playboy: The Mansion from Cyberlore Studios. "Our main objective is to encourage responsible development."

The Sex SIG will serve as a source for related industry news and will provide an online discussion forum and mailing list to promote developer interaction. The group is also working on several initiatives to fortify adult content representation, including conference lectures and white papers outlining responsible development practices and how to promote appropriate access to content.

"Sexual content in video games is not a new phenomenon," said Jason Della Rocca, Executive Director, IGDA. "Recent events are merely intensifying the attention to the topic, further validating the need for developers to connect on these issues."

For more information on the IGDA's Sex SIG, visit: http://www.igda.org/sex.



Sony Releases EyeToy: Play 2

Sony Computer Entertainment America announced today the release of EyeToy: Play 2, developed for PlayStation2 by Sony Computer Entertainment Europe. The collection of 12 new games lets players catch intruders in the act with the new SpyToy mode or create interactive 3D models of their own face and head with EyeToy: Cameo.

Since its launch last year, more than 7 million EyeToy USB cameras can now be found in homes worldwide. The camera plugs into one of the two USB ports on the front of a PlayStation 2 system while the camera sits on top of or below the television.

Combining various game genres, players can challenge their friends as they become a master chef, hit homeruns, battle it out for the table tennis crown, and fight off kung-fu warriors, ninjas and ancient monsters. For a more competitive atmosphere, EyeToy: Play 2 enables up to four players to play together in multiplayer modes for a heightened sense of fun and laughter.

Additional new EyeToy technology featured in EyeToy: Play 2 includes sound recognition, background subtraction, SpyToy and EyeToy: Cameo. Using the included SpyToy mode, players can use their EyeToy USB camera to guard their room, ward off unsuspecting intruders, and discreetly catch trespassers by tracking their physical motion, as well as audible noises, with their EyeToy USB camera. While playing detective, players can set up their SpyToy to capture secret photos every five seconds, record video of unwanted visitors or sound an alarm when the camera tracks any physical movement. With EyeToy: Cameo, gamers can create interactive and lifelike 3D models of their own head which can then be used in EyeToy: Play 2 as well as other EyeToy enabled games such as World Tour Soccer 2006.




IGC ´05 Speakers and Sessions Announced

GarageGames has announced the session highlights and speakers for the fourth annual Independent Games Conference. IndieGamesCon '05 (IGC) runs October 7-9, 2005. This year's conference is an informal gathering of independent game developers featuring a keynote address, "The State of the Revolution", by Mark Frohnmayer.

IGC's aim is to provide connections and content for programmers, artists and indie entrepreneurs. Speakers include Melv May on 2D game design, Ill Clan on machinima, John Welsh and Dave Nixon on the casual gaming industry, Thomas Buscaglia on legal issues, Ageia on the future of game physics, Ryan Gordon on Mac OS X game development, Microsoft on Xbox development, and more.

IGC 2005 presents a platform for the independent game development community to show off their creations, while learning about next-generation technologies and helpful topics for game developers spread out over five different break-out tracks. The tracks cover topics such as animation and lighting for artists, marketing and legal issues for the business side of gaming, and development techniques from seasoned game programmers.

Early registration is available online and costs $195 for a three-day pass, including sessions, t-shirt, food, LAN parties and beverage of choice ("beer") for those registering before September 16. Space is limited, so reservations will be on a first-come, first-serve basis. Registration after September 16th will be $250.




About Spectrum

Spectrum is an independent news service published every Monday for the interactive media professional community by Motion Blur Media. Spectrum covers the tools and technologies used to create interactive multimedia applications and infrastructure for business, education, and entertainment; and the interactive media industry scene. We love to receive interactive media/online-development tools and end product for review.

Send your interactive multimedia business, product, people, event, or technology news by email only to: spectrum_news [at] yahoo.com

If you contact companies or organizations mentioned here, please tell them you saw the news in Spectrum. Thanks.


Please send address changes (with old and new addresses), subscribe and unsubscribe requests etc. to the above address. If you use the Reply function, please do _not_ echo an entire issue of Spectrum with your message.


Publisher's note: We are now accepting limited advertising. If you'd like to offer your company's products or services to Spectrum's elite audience of Internet and multimedia professionals, send an email query to mailto: spectrum_news [at] yahoo.com.
- David Duberman


©Copyright 2005 Motion Blur Media. All rights reserved. No reproduction in any for-profit or revenue-generating venue in any form without written permission from the publisher.