in Conferences and events

My Take on GDC 2004

The days immediately following GDC are always decidedly hectic. Not only do you need to play catch up with the life you left behind for a week, but you also try to get back all the hours of lost sleep between parties and red-eye flights, email all the people that you met (or missed seeing) at the conference, prepare roundtable summaries or put up session slides, and deal with all the emergencies that happened at the office while you were gone. With all that now safely behind me, it’s time to look back on this year’s GDC. What was hot and what was not? What were the underlying themes and trends of the conference?

gdc2004The full-day tutorials were a bit on the weak side this time around, but the “Test Automation in Game Development” done by a group of Microsoft test engineers more than made up for it. I found it to be a very well-paced session, covering things in enough (but not excruciating) detail and with lots of very solid information. The topics towards the end were particularly exciting, dealing with unifying all the testing strategies under one system, and doing data mining effectively with the unwieldy amount of results generated by so many automated tests.

The sessions in the last three days were all very strong. I felt there were more quality sessions in the topics that interested me than in previous years. Interestingly enough, I didn’t go to many programming talks this year; for some reason, I ended up going mostly to game design and production talks, and enjoying most of them a great deal. The award for the most redundant talk of the show goes to the ones covering some form of real time global illumination. Maybe it was just me, but it seemed the same talk was given by a different person with a slightly different title several times each day.

The highlight of the show without a trace of doubt was Will Wright‘s talk “Triangulation: A Schizophrenic Approach to Game Design”. After attending six GDCs, I have learned that there are some people that you simply have to go see. It doesn’t matter whether they decide to give a talk about how their baby girl is learning to walk; you’re guaranteed to have a great time and walk out of there with a lot of insights. Will’s talks are like that. This time he described the process he uses for designing a game, how he approaches the problem from multiple angles, and how he treats it as an infinite tree of possibilities that needs to be culled through experimentation and instinct. Even his digression into the history of the Russian space program was highly entertaining and bizarrely relevant. But please, when you bring Will back next year, give him the Auditorium. There was no excuse to put him in a medium-sized room that was already overflowing 10 minutes before the talk.

Soren Johnson’s talk from Firaxis on “The Civilization Series: How to Maintain a Successful Franchise” probably gets the award for the best unexpected talk. I might have missed this session except for the fact that I met Soren the day before and there was nothing much going on at that same time slot (except for the fourth rehash of the global illumination talk), so I decided to give it a try. Boy, was I impressed! Soren had some very good insights on game franchises and how to think about them in terms of technological/gameplay innovation and release frequency. He then analyzed in detail two very successful franchises (Age of Empires and Warcraft) and compared the decisions made for each of them. The talk finished up by presenting the Civilization franchise in terms of everything he described before, which wrapped things up in a very effective way. I can’t wait for Civilization IV!

A very simple, yet effective and informative talk was Brian Sharp‘s session on “The Physics-Sound System of Deus Ex: Invisible War and Thief 3” (whoo-hoo, I made it to a programming talk for a change!). I really liked how he decided to approach the talk: Present an apparently simple problem (what sound do we play when two things collide), then start with simple solution and build on it as its deficiencies became apparent. He wrapped up the talk with a nice demo level with objects making all sorts of noise when they were being dropped on each other.

will I had high hopes for John Carmack‘s keynote. I met John a few years ago at the first GDC Hardcore Technical Seminars and I knew he was a reasonably good speaker and, most importantly, had great insights into the industry and future directions of the technology. Unfortunately I think this might have been a case where I was expecting too much and the keynote fell a bit flat in the end. He echoed a lot of the same things he mentioned back in 1999, such as the fact that graphics are getting better and better and other things need to catch up soon (especially physics and animation). He also expressed his concern about how more and more people are required every day to make a game, and he encouraged people to find alternate genres and different ideas that might be accomplished by a small team. I found it very funny when he lamented how Doom 3 was the first project he ever worked on where he didn’t have full control over the source code and there were even a few source files he hadn’t even opened. Welcome to our world, John!

A notable mention goes to the talk given by Toru Iwatani, the creator of Pac-Man, “The Secret of Pac-Man’s Success: Making Fun First”. It was very interesting learning what he was thinking when he created Pac-Man, and to get an insight into Japanese culture. The session was simultaneously translated and worked remarkably well. Besides, I even caught a few words in Japanese with what I’ve been learning from Shogun :-).

Another interesting talk was the “IGDA Quality of Life White Paper Unveiling” by Scott Bonds, Dustin Clingman, Hank Howie, François Dominic Laramée, and Greg LoPiccolo. They are bringing strong proof about something I’ve always felt very strongly about: The industry is really hurting because of the poor, non-sustainable work conditions. When people are working 60-70 hour weeks on a regular basis, they burn out. In the short term they won’t do their best work; in the long term, they’ll leave the company and eventually, the industry. The results from a recent Game Developer Magazine survey were surprising: Over 60% of the people in the games industry have less than 5 years of experience. But when combined with the results of the IGDA survey, they’re positively spine-chilling: 34% of the people are planning on leaving the industry in 5 years or less, and over 50% of them are planning on leaving in 10 years or less. Those are really scary numbers, and unless something is done about it, the industry is doomed to continue being stuck in its adolescent phase forever.

The best-talk-I-didn’t-go-to award goes to “Entertainment Experience First, Videogame Second: The Making of The Return of the King” by Neil Young. I don’t know what I was thinking when I chose to go to another (rather uninspiring) talk instead. To add insult to injury, I was actually in the room next door and I could hear the audience roaring with delight through the wall. I don’t think the audio proceedings will cut it either considering it was the Visual Arts keynote. I’m still kicking myself over that, so don’t even think of mentioning it to me again.

Stepping back and looking at the sessions I attended, there were a couple very strong underlying themes. The first one is that the teams necessary to create a top game continue to get larger and larger. People are finally starting to realize that the development methods traditionally used throughout most of the industry are falling short and something needs to be done about it. It’s probably no coincidence that the number of attendees to my software engineering roundtables has been increasing every year. Also, middleware is making its way into more of the talks, often as a side or secondary theme.

The other sub-theme apparent in many of the talks was a lament (or alarm, depending of the talk) about the lack of innovation in game design. Speakers were urging developers to think of new approaches and not try to compete with the big players in well-established genres. The only hope of spark was the downloadable game IGDA competition, where innovation is still alive and well. Hopefully we’ll see more activity along those lines next year.

I’ve never found the GDC expo floor to be particularly interesting. It’s a great way to meet face to face with tool creators and middleware providers but not much else. It’s not like I’m going to purchase Renderware Studio on an impulse buy, and it’s not like I haven’t read all the specs and seen demos of the tools I’m interested in. Still, a few times you might come across some small company that had gone under your radar before. I didn’t spend much time in the expo this year, but I got the impression that it was mostly the same faces from last year, no surprises there. However, the expo was always a chance to renew your t-shirt wardrobe. In years past, you would be literally bombarded by t-shirts being thrown at you by the vendors as you walked past their booth. You get a cool t-shirt (hopefully not in fat game developer standard xx-large size), and they get advertisement for a year. Not this time. Maybe the economy is down, maybe the industry is getting ready for another change with the next generation consoles looming on the horizon, but t-shirts were more scarce than hens’ teeth this time. Either things change next year, or I’ll have to actually buy some t-shirts for the first time in years! 🙂

Next year GDC will break with tradition and be hosted in San Francisco. I admit I was finally becoming fond of San Jose, but changing venues will have its own share of advantages. It would be fantastic if they decided to change locations every year like SIGGRAPH and we ended up with a GDC on the East Coast once in a while. In the meantime, I can look forward to dinner at the Stinking Rose every night!

Images from