GDC

I attended GDC 2007 the week of March 5-9; it was my first time there and was pretty excited about it.  I tried to attend a variety of sessions, from the very technical (some of which were way over my head) to the general (those that were part of the game design track).  I also spent some time in the expo halls - one focussed on large companies such as Sony and Autodesk; the other on smaller companies and independent games.

 

Overall, here are the big ideas that I walked away with:

  • GDC sessions start ON TIME and sometimes fill up; being late means you'll either miss something or everything.
  • Wii and other alternative inputs are big for attracting new customers (e.g. Buzz: the Music Quiz)
  • The Serious Games sector appears fractured and disorganized
  • Physics engines and other tools out there are big and powerful.  Breaking stuff in a virtual world is fun for some reason.
  • The gaming industry has iterated a lot on graphics and physics over the last 10 years.  It seems like the next frontier might be the social/psychological realm and doing real research on what makes people enjoy games and how games affect them..
  • Independent games are an interesting space but aren't necessarily as interesting as independent films.
  • There is a lot of stuff I know very little about.
  • XNA is very fun and easy to get started with.
  • San Francsico is a real city with real public transportation that enjoys sunny weather every day and has great bridges, architects and museums.

Would I want to go back?  Definitely.  Being away from Seattle and work and everything for a whole week (6 days in total, actually) was a bit much, perhaps.  Three days (and three nights) might be ideal. 

Proceedings are available here, but sadly it appears that far less than half of the presentations have actually been uploaded. Intel's presentations are on Intel's GDC presentation page.

Rundown of sessions I attended:

Monday: Serious Games Summit

  • Keynote: demo of DS game that supposedly teaches game development.  No programming, but some art development and the adjusting of sliders, etc.
  • Games where there's no difference between the game and the task.  I thought this was interesting: you can get people to do work in unexpected or surprising ways.  e.g. Using porn-surfers to pass the Turing test for website account creation.  Another interesting idea: making call centers like a game: "If ou can get to level 5, we start paying you.  If you get to level 10, you can start interacting with customers."  In short, all of life can be looked at with an eye towards "how can this be fun and challenging?" much the way we look at games.  Another intersting not-quite-a-game-but-kinda-because-it-resembles-monopoly is "Seriosity" - a plugin to Outlook that limits the amount of e-mail you can send based on how much play "money" you have.
  • A Fundamental Challenge: Pacing - "Warfare is long periods of boredom punctuated by brief periods of terror."  In other words, sometimes it's necessary or at least a good idea to alter the rate of time passage based on what's going on in the game.
  • Taking advantage of what people would be doing anyway: people out with GPS units geocaching - get them to spend a few minutes collecting environmental data while they're doing it.
  • Interesting idea: surgeons "warming up" their hands before performing surgery - by playing off-the-shelf video games.
  • Labyrinth: the education arcade.  A math and literacy game, funded by the department of education and created by grad students at MIT.
  • 3 Up, 3 Down: 4 panelists (Jesse Shell, Richard Van Eck, Roger Smith, and Doug Whatley each present the good and bad things that have happened in the Serious Game arena in the last year. 
    • Some bad things: standardized testing making educators reluctant to use games in the classroom.  No good guides for serious game creation.  The best and brightest work on entertainment titles instead of serious games.  Art asset management across the industry is poor, so there is duplication of work all over the place.
    • Some good things: universities are taking games more seriously, offering degrees in game development, etc.  The Wii is making a positive impact on the "gatekeepers" of education.
  • How to use consoles for your own "serious" purposes: for example, installing Linux on a PS3.  The army is controlling robots with a 360 controller!  The Wiimote is being used to replace other devices that cost > $1000.
  • Reliving the Revolution: a project by an MIT grad student.  A game that can be played at Lexington... kind of like Carmen Sandiego: you walk around to different points, and based on your GPS coordinates, you can "talk" to different people and gather information to help you decide who fired the first shot in the American Revolution (no one really knows).  I found it interesting that the presenter said that she "wasn't interested in the factual knowledge that was gained" by the students.  I can see teachers and parents having issues with this... she instead emphasized the "soft skills" that were gained in the exercise.
  • Next-gen conversational characters for serious games.   Presented by the creators of Facade . Key takeaway: this stuff is complicated as hell and Facade only gets things approximately right 80% of the time, even for a seemingly small world of possibilities like chatting with two friends in their apartment.  (Not even trying to use voice recognition or NLG - natural language generation).  Some of the choices that need to be made:
    • Natural language vs multiple choice
    • Turn-taking vs realtime
    • Static images vs smooth animations.
  • Some examples of serious games that need to make the above decisions: Training doctors to interact with patients (e.g. how to tell someone they have cancer); training managers to deal with employees; training soldiers to deal with foreign civilians.
  • What We Know: Eliminating Noise With Simple Truths  This guy (Howard Phillips) seriously had his shit together and offered tables of information comparing and contrasting cognitive science and computer games.  

------------------

Tuesday: Physics for game programmers - full day tutorial.  Also attended by NickCo and Alan Muller (DirectX team).  A lot of talk about collision detection.  I have printouts of all the slides, but I don't see them online.

 

------------------

Wednesday:

  • Women in gaming meet-up: a networking opportunity.  I was disappointed that out of the 10 or so women that I met, none of them were devs.
  • Power consumption for games on laptops.  Talked about the TDK; more information here. To quote from that page:
    • Deliver code that responds to platform state changes such as power source changes, battery power levels and WiFi signal strength variations.
    • Improve gameplay by monitoring platform characteristics and taking appropriate action in response to changes due to mobile usage.
    • Provide for better processor utilization in multi-core platforms using threaded API calls.
  • D3D10 - new features (arrived a bit late)  Lots of stuff.  Way over my head.
  • Composing Character Performances with the Havok Behavior Tool:  Allows you to create complex state machines describing all of the animations that a character uses and the transitions between them.  A cool viewport.  Can export directly from Max into the behavior tool.  The guy next to me in this one kept falling asleep, but I thought it was interesting. We don't exactly have tools like this for Flight Sim. :)
  • Women and Games in the Future.  The one session that had an audience of at least 50% women.  Takeaways: Sheri "Did you read my book?" Graner Ray, head of WIGI, came across as a stubborn, argumentative, and not especially bright person. Energetic Carri Heater did just the opposite, as did the sage Don Daglow.
    • My favorite quote from Don Daglow was something like "If you're not happy, you can't make fun games."  His point was that sustainable workstyle is important in the industry.
    • 70% of online casual gamers are female
    • One of the attendees of this session looked just like Abe Lincoln. 
    • The percentage of game designers that are female is even smaller than the percentage of game programmers that are female.
    • Book mentioned that sounds interesting: From Barbie to Mortal Kombat: Gender in Computer Games
  • Making games for the other 90% (Buzz: the Singing Quiz). This one was attended by a full house.  Before its RTM, this game had some definite ups and downs.  It got an icy reception at E3.  But in the end it did quite well in the market (although it hasn't shipped to the US for reasons that he wasn't clear on).  David Amor was a fun speaker to listen to, even if he was weak on his stats and data backing up his statements.  His main ideas for appealing to a broad audience were:
    • Keep it simple
    • Use ideas that people are familiar with already (e.g. gameshow)
    • Make the game approachable (sometimes realism is intimidating)
    • A lot of the entertainment is off-screen (people laughing at each other trying to sing, for example, or coming up with strategies for gameplay)
    • Interface is "really, really" important.  Buzz shipped with 4 big red buzzers that people could use instead of the PS controller and was a hit.  It was also used in bundles for console sales.
  • Viva Pinata graphics engine (presented by the faster talker I've ever seen)  Unfortunately I mostly just heard lots of intriguing terms I didn't understand here, such as:
    • Predicated tiling pass
    • Opaque alpha-tested rendering
    • "alpha-to coverage"
    • CIE clear sky
    • MSAA (Multi-Sample AntiAliasing)
    • Phong lighting

-------------------

Thursday:

  • Keynote - Miyamoto (famous for Zelda and Super Mario) talked about Nintendo and the evolution of games.  In general throughout GDC it sounded like games had a relatively broad appeal across genders at the very beginning (Pong, e.g. being gender neutral), but soon focussed on the young male consumer and are now turning the corner with consoles such as the Wii.  Like David Amor he emphasized the importance of simplicity, prioritization, and making games not intimidating but fun.
  • 10 video games you need to play.  The panelists chose (in no particular order):
    • Doom
    • Sim City
    • Sensible World of Soccer
    • Tetris
    • Star Raiders
    • Spacewar!
    • Zork 1
    • Super Mario 3
    • Civilization
    • Warcraft I/II/III
  • They didn't mention King's Quest or Myst or anything made in the last 10-15 years, which I thought was odd. They did talk about the need to preserve old games, which is something that isn't being done well currently. 
  • The Making of the Godfather - This is a violent game that at first blush has nothing to do with Flight Sim.  But my ears perked up when I heard the term "Living World" being used repeatedly.  They have some relatively simple AI (like the Sims; one of the panelists had worked on the sims previously) for all characters in the game, which allows them to wander all over the world.  So sometimes it's hard to find the butcher...  The panelists emphasized the importance of iteration: they said that the game development naturally "followed the fun" which I thought was an interesting and wise statement.
  • Round table - Interacting with the fan community.  It was interesting to talk to other people in the industry and hear about their experiences.  And how some companies filter interaction between developers and the community.  For example, at EA developers aren't allowed to post.  Everquest has 20 rules for posting. Also interesting: for some MMORPGS, devs have distinguishing marks so that players can bother them with questions.  Some games and some game characters have MySpace pages.

-------------------

Friday:

  • Hands-on workshop: Profiling tools for XBox and Windows (covered PIX for windows and VTune from Intel
  • Applied Physics in Motorstorm.   Holy cow.  This stuff was very very cool.  I want to buy a PS3.  (And a Wii.  Hmm.)  They show cars driving through all sorts of things causing damange not only to the car, but also dismantling buildings and knocking over poles, etc., etc. Unfortunately I haven't yet found a cool video clip online of this stuff, although YouTube has a lot of videos that people captured illegally at GDC.  The mud/dirt deformation is cool, too. They talked about their multiplayer data stream: critical data (player position, etc.) comprises 40 bytes and is sent at 60 ms intervals, and lower-priority data (taunts, damage, boost) comprises 88 bytes, sent at 200 ms intervals.  Damage in the world is not synchronized.
  • Hands-on workshop: XNAXNA might be the coolest game I played during the week. I want to go home and play. Now.  In two hours, Charles Cox (from Microsoft) walked us through every line of code we would need to write to create a 3D shoot-down-space-invaders-with-a-cannon game.  OK, there was some a camera library and terrain library that were given to us (and I'd like to get my hands on)... and the 3D models of the cannon and the space invaders were given to us as well.  But STILL.  We were paired up on machines, and my partner and were able to tweak some of the physics and space-invader-behavior as we were going along.  Good stuff.

--------------------

 

Notes from the expos:

  • The Novint Falcon is a very cool input (and output) device - for only $200 it gives you the ability to give 3D input AND get force feedback orders of magnitude more interesting than the rumble of a game pad: it simulates the feel of a rough surface or pulling your hand through molasses or playing with a ball at the end of a string. 
  • Finally tried TrackIR for the first time. I still managed to crash the helicopter in whatever game I was playing.  If I wasn't so distracted by trying not to crash, it might have been a fairly compelling device.
  • Finally tried the Wii for the first time, although not in a sports game (which I think would be most interesting).  It was easy to use as a pointing device.
  • Also tried out the DS
  • Crysis engine = cool.
  • Independent games:
    • Armadillo Run was the one I played the most; it allows you to build mechanical systems with the goal of getting a ball to roll into a slot, etc.  Similar, apparently, to The Incredible Machine, which I never played back in the mid-90's .  
    • Another game (can't remember the name, but vaguely resembling space invaders) generated the soundtrack based on the current gameplay.