Poker Players Alliance: Keep It Legal

GDC 2007 Recap

Serious Games Summit:
The SGS was very fruitful. Admittedly some of the sessions were less valuable than others, but on the whole the summit was very productive. The first session of the first day of the summit was by Ichiro Otobe of Square-Enix on their joint venture with Gekkan to form SG Labs, a serious game development studio. It was emphasized early in the talk that the purpose of the collaboration was to make money. This is important! Many people, especially those of us in academia, have a conception of serious games as non-marketable educational-based games. Games such as Brain Age for the Nintendo DS have started to change this. Square-Enix and Gekkan are working on a game to teach basic 2D game development and is run on the Nintendo DS. They have a website, http://sg-lab.net, but unfortunately is in Japanese only.

The “3-Up/3-Down” session was very enlightening. A variety of people heavily involved in the production and utilization of serious games in industry and education gave three positive things that happened last year in regards to serious games, and three negative things that happened. Here are the points made the speakers:

  • Richard Van Eck - Positives
    + Broad appeal of serious games and growing perspectives
    + Growing acceptance of serious games in academia
    + Reaching a critical mass in K-12 market. Textbooks are becoming interested in SG.
    - Multiple perspectives are not yet integrated
    - Still broad misconceptions of SG. Need solid data and methodologies for SG integration.
    - SG is still not a part of public education in any meaningful way.
  • Jesse Schell
    + Academic explosion of interest
    + The Wii! Whole Family Experience
    + Broadband penetration allows for broader access to and easier distribution of serious games.
    - No excuses any more! Tools are cheap as free and distribution is is available.
    - Still no clear guide and defined techniques for creating and using quality serious games.
    - Gatekeepers of education still do not believe in the usefulness of serious games.
  • Roger Smith
    + Use of serious game in the combat theater
    + Military funding of MMOG for insertion into the field
    + Recognizing the value of art assets
    - Limiting and constricting licensing agreements. Disallows flexibility.
    - DOD IT security policies. (for government computers and application)
    - Many tools and they’re all hammers. We must get away from the FPS genre.

A session that interested me greatly, although I’ve not found much useful information on the official website about it, was about the game Selene. Selene is what the presenters call a “theory-based” game. Players, 13-18 years of age, explore the Moon and learn about lunar geology. This appeals to my “learning science by doing science” philosophy, but I was unable to learn much about the game or even see a demo at their website http://selene.cet.edu/.

The two best sessions of the summit were on the last day. The first one was a demonstration of the massively-multiplayer game at Whyville.net. I don’t know if this is truly a game more than its a social interaction tool, but either way its a great vehicle for education. The players at Whyville, mostly 5-8 graders, have avatars that they create themselves and with them they can move around the Whyville environment and chat with other players. The moderators can at any time affect some change to the environment that confronts the players with significant challenges. The players are not told how to solve the problem directly, but they are given the tools needed to find the solution themselves. This again goes back to my philosophy of learning by doing.

The other remarkable session discussed the difference between what the presenter called a “big game” versus a “little game.” A little game is one that focuses only on mastering facts and skills for the purpose of passing a standardized exam. Math Blaster is a good example of a successful little game. Little games, however, don’t really build true understanding or long-term retention. A big game by contrast incorporates the learning of facts and skills with the values and culture of the discipline. The edu-speak term for this is epistemology. The example game demonstrated during the session was Digital Zoo in which students are given the task of designing a creature using point-masses and springs that exhibit a specific behavior or motion. During the process of building the creature, the students learn not only the facts and skills of a structural engineer, but they also learn about the process of engineering and how to think like an engineer. Once again, learning by doing was demonstrated to be a powerful educational tool. This is one aspect of science education that’s really deficient and serious games can provide an excellent vehicle for making science education more approachable, enjoyable, and productive.

Main Conference – Programming Track:
Throughout all of the programming track sessions I attended, the predominant theme was how to make your code more efficient and the real need to use as few CPU instructions as possible. The first session was a roundtable discussion about doing physics on the next-gen consoles. The presenter had a detailed agenda listed at the start of the session, but we only made it down to the second bullet, collision detection. The method by which collision detection is done can greatly affect the performance of a game. The keys to efficient collision detection code, in brief summary, are sticking with integer calculations where possible and use the parallel architecture of next-gen machines effectively.

Tien-Tsin Wong of the Chinese University of Hong Kong gave two great talks on efficient sphere mapping. The first was on using spherical radial basis functions rather than spherical harmonics to produce intensity maps. SRBFs are cheaper to execute, the mathematics are simple, provide nearly identical results, and are better at generating local lighting effects than spherical harmonics. His second talk was a comparison of various sphere maps, their advantages and their disadvantages, and which ones provided a more uniform solid angle for each sphere map element.

Keeping with the rendering theme, Eric Lengyel of Terathon Software gave a good talk on tips and tricks when using projection matrices. I’m not an animator myself, so many of the terms used during the talk were new to me, but the mathematics and the concepts presented were very interesting to me.

The most interesting presentation in the programming track set of sessions was Andy Thomason’s talk on compiler theory and techniques for streamlining the execution of your code. It was fascinating that with CPUs that can process 3×10^9 instructions per second, we’re still concerned about techniques for saving a dozen or so instructions in our code. Of course, what isn’t immediately obvious to a casual observer is that a writing a routine in such a way that you eliminate 10 instructions can really add up if you’re applying that routine to every one of the hundreds or even thousands of vertices. Now instead of saving 10 CPU cycles, you’re saving on the order of 10,000+ instructions. I have not yet been able to find the slides of Mr. Thomason’s presentation, nor his e-mail address. Now in retrospect, I wish I’d taken better notes.

Main Conference – Game Design Track:
I didn’t go to too many of the design track sessions since that’s not really my forte. One could make the argument that my inexperience in design is precisely why I should have gone to more design sessions. Perhaps next year. The one I did go to was absolutely one of the best sessions of the conference. Damion Schubert of Bioware Austin gave a tutorial on writing good design documents and the importance of good documentation. The slides are available on the WebCT page, and at Mr. Schubert’s blog site, http://www.zenofdesign.com/. Many of the things that he illustrates and stresses won’t really come as a surprise to many of you, but for people like me who have never had to write design documents before, it was extremely helpful, even if most of his points were merely an exercise in common sense.

Main Conference – Group Talks and Informal Discussions
There were two group-talks held in the IDGA booth on Friday morning. The first was a meeting of academians, both faculty and students, and the second was a meeting of bloggers. Unfortunately, I couldn’t hear much of the conversation during the academic talk because the meeting was in the middle of the main lobby where there was a lot of background noise. It was interesting that many of the people at the first talk stayed put for the second talk. Perhaps its the collegial mentality that leads many academians to also be bloggers. During the second session, I talked with several people about our own personal blogs, and how blogs are useful professional tools for conveying information to colleagues and students. I was also introduced to the IDGA Education Special Interest Group blog, http://www.igda.org/education/. They also have a wiki-site where one can upload syllabi to share with the larger game development educational community at http://www.igda.org/wiki/index.php/Game_Education_SIG. I’ve started the process of uploading the syllabus for the MATH/PHYS 191 course. So far, its the only game math and physics course listed.

The conversations about blogs and wiki, and their power in terms of information dissemination and encouraging collaboration and discussion made me wonder why we weren’t doing more here on our campus with blogs and wikis. In the past, the staff.jccc.edu server has been PHP-enabled, but there was no MySQL server. Most open-source blog and wiki software require both PHP and MySQL. I did find one open-source PHP-driven blog software, Pivot, that stores its data in flat files rather than a database. Unfortunately, after loading it up onto my folder on staff.jccc.edu, I discovered that the server is no longer PHP-enabled. This is seriously problematic. I’ll continue to work through the rest of the semester at convincing the Educational Technology Department that providing these services would be in the college’s best interest.

I talked with a lot of people in the expo hall during the conference, admittedly for the self-serving purpose of scoring more loot, but there was one conversation that turned out to be very rewarding. While drooling over the Playstation3 consoles, I noticed that they had a version of YellowDog Linux running on one of the PS3s. I asked about the price of the distribution, and the rep I was speaking with told me it was free and handed me the disc! This means, of course, that I’ll be buying a $600 game console after all. That got me thinking, though, why not have the college purchase a few of these for the game programming class? The PS3 is a potent computational platform. You can see the essential stats on Cnet’s site, http://news.com.com/PlayStation+3+stats+spotlighted/2100-1043_3-5709571.html. The cell-based architecture provides an platform for students to learn about parallel processing, and for faculty to have access to hardware capable of some very high-end computing. The cell processor alone has a floating-point performance of 218 Gflops! The machines can be purchased from YellowDog directly for $650 each. Compare this to the price of $1000 for a standard campus computer which will still need the software development environments installed on them. The PS3 with Linux comes with development environments for all major languages and for the cell processor specifically. It would be a good idea for us to apply for a technology grant to obtain at least one machine for trials and experimentation.

Summary:
This was an excellent experience for me, both personally and professionally. From a personal perspective, I got to see some of the real stars of the gaming community such as Shigeru Miyamoto. Professionally, many of the sessions spoke to several aspects of my responsibilities as a teacher and as a scientist. Here are some of the major take-away items for the conference:

  • The game programming world is really at the bleeding-edge of high-end computing.
  • There’s a large community of educators involved in game development programs, and the IDGA Education SIG is the main portal for communication among educators.
  • I have more to learn about design and programming that I realized!
  • Serious games can be extremely powerful educational tools when students are allowed to be immersed in an environment and given the freedom to learn by discovery.
  • There is a great need for the JCCC servers to support PHP and MySQL-based blog and wiki software for both staff and student websites. Dynamic pages with built-in communication tools are essential.
  • The PS3 is a far more potent and capable computing platform than I ever expected, and for half or even a third of the cost of most middle-range PCs, its a real bargin.

2 Comments

2 responses so far ↓

Leave a Comment