Editor's note: This week's Spectrum edition is a special report from last week's Game Developers Conference in San Francisco. We'll return next week with news from the show floor and elsewhere. ___________________________________
Today's Headlines (details below)
GDC 2005 Report
by David Duberman
I've been going to the Game Developers Conference (GDC) for over 10 years now, starting when it was the Computer Game Developers Conference, held in Santa Clara (but I'm still not a game developer--go figure). Were you to ask me which show was my favorite, I'd probably say it was the one I attended most recently. But I had a particularly great time at the GDC that took place last week in San Francisco, not least because I didn't have to schlep down to San Jose through horrible traffic at an ungodly hour every morning for a week straight. More important was that the show seemed to partake of S.F.'s fantastic energy; as one developer I overheard discussing the new location for the event said (and I paraphrase), "It's nice to go outside and see people other than those attending the conference." (The S.J. conference center is located in a relatively isolated part of a city that doesn't have much energy to begin with.)
If you took any notice of this year's conference, you probably heard about how Microsoft gave away 1,000 HDTVs (23-inchers) at J Allard's keynote. I was one of the lucky winners, which was nice, but that was far from the high point of the show for me. Much better was getting a chance to hear Will Wright describe his new game, Spore. In a talk that by all rights should have been a keynote (the line to get in wound around an entire floor of the Moscone West conference center), which was greeted with wild enthusiasm and a small amount of kowtowing by the SRO audience, the King of All Sims showed that his creative fecundity hasn't slowed down over the years; if anything, it's become more intense. Ironically, Wright ended his talk by telling us that Spore was based an idea he'd first had a number of years ago, but had determined that it was technically unfeasible. His point was that other designers shouldn't give up on ideas just because they don't seem possible at the time.
Wright's talk was titled The Future of Content. Although he apologized up front because he'd had to provide the name months earlier, and the talk had evolved since then, he faithfully began by describing how, with the recent ramping up of technological capabilities, providing high-grade content in adequate quantity is becoming very difficult for game developers. In response, he created a game in which players provide most of the content, which they love to do, going by Wright's experience with The Sims.
If you're not familiar with Wright's work, you should know that he specializes in what's called "sandbox" games; he provides the sandbox, and the consumer plays in it, making whatever he or she can dream up given the game's context. If you let players create content, they can then share it with each other and create stories about it, which provides them with ownership and in turn gives them more meaningful experiences with the game. Wright talked about how most games begin with a sandbox-like tutorial where you can just play around, and then take you into the goal-oriented game proper, but Spore turns the tables by starting you out with a goal-oriented period followed by the rest of the game, which is a vast sandbox (essentially, the universe).
On to the game itself, which I'll describe sinfully briefly: Spore takes in a lot of gameplay territory, to put it mildly, and as such is difficult to describe. Essentially, you start with a small creature in the water that evolves and becomes more powerful, eventually climbing up onto land and forming civilizations. Wright describes this phase as a simplified version of Sim City. Different civilizations can wage war on each other, and eventually the player gets a UFO that he can use to take the best parts of one planet to cultivate life on another. The asynchronous multiplayer aspect is where the game gets really interesting; players can swap content and take off on each other's ideas. Now that the cat's out of the bag (this was the first I'd heard of the title, and judging from the audience reaction, I wasn't alone), you'll be hearing a lot more about Spore, at least after E3, which is the next time Wright will discuss it. When will it ship? He didn't say.
The first two days of GDC are dedicated to one- and two-day tutorials, of which I attended a fascinating one on Monday. Noah Falstein, the speaker, has been creating games for longer than many gamers have been alive. His class, titled "The Working Game Designer," met its goal of describing the practical aspects of being a game designer admirably well. Falstein started by telling what game designers are like: They come in all shapes, sizes, and sexes (including trans), and are typically obsessive players, moviegoers, readers, etc. Falstein went on to describe a number of other game-designer characteristics, but one in particular that stuck with me, which he mentioned in response to an attendee question, is that a true game designer has many more ideas for games than could ever be brought to fruition in a single lifetime. So it's important to be well rounded, but if you don't eat, sleep, and dream game design (typically starting from an early age), you might want to seek out some other line of work.
But what does a game designer actually do? Falstein described this in terms of a typical project, in which a producer runs a core team of a lead designer, a lead programmer, and a lead artist. Early in the project the designer has no direct reports, but later on he might manage two level designers and work with QA as well as an external writer. Of course, the designer must come up with a concept for the game; ideally, she originates the idea, writes the initial concept, and teams with production to sell it to management. In the real world, however, the designer gets handed a license and struggles for ownership of the idea while management dithers and then finally green-lights the project two months after it was supposed to start.
Falstein also discussed the all-important design document, lamenting the fact that few actual examples exist on the Web, unlike the case with movie scripts, which are readily available for most productions. He has a number of them in his archives, of course, and is considering writing a book showing how to create one. During game production the designer maintains, expands and interprets the design document, which often changes radically throughout the project. The designer's responsibility is to make sure the game is fun while completed on time and budget, and might manage the creative tasks of others, such as writers and artists. When ship time approaches, the designer might shift to working with playtesters and QA to fine-tune the game, or possibly begin work on an expansion or sequel.
There was much more in Falstein's tutorial that I don't have room for here, so I'll conclude by mentioning a couple of highlights. Falstein referred regularly to shining examples of game design, rightly so, and one of his current favorites is last year's sleeper hit Katamari Damacy for PS2 (only $20 and lots of fun; get it now!). After lunch he brought in his teenage daughter Mara to demo the game while he described what he liked about the design, such as its "desperately simple" gameplay (quote ascribed to Brian Moriarity), which appeals not necessarily to women but to non-hardcore gamers.
Lastly, Falstein provided attendees with an invaluable handout, the result of his polling a large sample of game designers on six questions chosen to advise and inspire people interested in becoming game designers. The information in the handout is proprietary, but I can probably get away with paraphrasing, without attribution, a few of the responses. A beginning designer can learn by: designing board games for friends on his/her own time; testing a combat system by creating a playing-card version; playing lots of games and trying to figure them out rather than just enjoying them. Advice to would-be designers: don't; get a good education; build a complete game using an existing tool such as Visual Basic; play traditional games such as cards and Scrabble.
Falstein is a real treasure in the game-development community; someone who's been down in the trenches since the start, but still loves his work and the idea that games can enrich lives, and is working to spread the gospel. His Website URL is http://www.theinspiracy.com. If you visit, be sure to read the Natural Funativity article.
The rest of the conference mostly took the form of all-too-short one-hour sessions on Wed.-Fri., the same days as the exhibition. One interesting session in the new Vision track was led by Peter Molyneux, the British designer best known for inventing, or at least popularizing, the god game genre, of which Populous is perhaps the best-known example. Molyneux is certainly a visionary, but he's also a master salesman, which role he played to the hilt in a Wednesday lecture entitled Gameplay Moves Forward into the 21st Century. He began by listing what games need to appeal to the mass market, such as a clear concept (e.g., in GTA, be a gangster); greater accessibility (i.e., no need for instructions); deeper interaction; and the ability for the player to experiment.
Molyneux devoted the remainder of the session to demos of two games and an unusual experiment. The two games, due out this fall, are no secret: The Movies is a simulation of a movie studio, complete with the ability to let you create your own 3D movies, and Black & White 2 is a follow-up to Molyneux's, latest, somewhat unsuccessful god game. Both look to be a lot of fun, and will probably ship this fall.
But of greater interest to those seeking innovative design was The Room, which Molyneux created in response to the challenge, in another session (which I unfortunately had to miss), to make a game based on the poems of Emily Dickinson. Molyneux described the evocative 3D environment as "more about the emotion of where you are and what you're doing," and as an attempt to play around with the senses. Despite the "hyper-realistic detail" (you could zoom in very close to objects and keep seeing more), the "room" was quite dreamlike, and you could move through portals and send objects through, which would cause them to change in some way. For example, he showed moving a ball through a portal into a larger space, whereupon it scaled up to retain its relative size, but then upon moving back into the original space it was much bigger. I might have that wrong, because by that point I had been lulled into something of a trance by the unreal nature of the demo. Molyneux first said it would never be released, but then relented to say that it might come out as a demo. He concluded by saying he's looking for game designers, so if you want to move to England and work for a genius, now's your chance.
Ernest Adams, an American academic and designer who resides in England, gave a largely theoretical talk on interactive narrative. He began by referring back to a talk he gave 10 years ago, during the heyday of the adventure game, in which he concluded that interactive narratives don't actually exist, due to three key difficulties. First, it's impossible for the designer to ensure internal consistency and coherence due to the fact that the player is out of the designer's control. The second problem revolves around narrative flow: how to make sure the player is prepared for a dramatic climax when it arrives. And the third concerns itself with amnesia: the fact that the characters understand the world they live in, but the player is new to it.
Since the original lecture, with the changing gaming landscape, Adams has concluded that some of these aren't necessarily insurmountable. Adventure gaming as such survives (cf Broken Sword), but the genre isn't very strong; "It's easier to entertain people through adrenaline," quips Adams. Nonetheless, storytelling has crept into other genres and has improved; we now find good stories in RPGs such as Final Fantasy and Planescape, in FPSs such as Metal Gear Solid, and in the relatively new action-adventure genre. As an example of the latter, Adams cites the relatively ancient title Indiana Jones and the Infernal Machine, but there must be more recent instances. Nowadays, story is used to make games feel richer, but it takes a back seat to gameplay.
The typical solution with regard to internal consistency is to place the gamer in an important situation and allow him/her some (but not complete) freedom to choose his or her actions. The problem of narrative flow is resolved by creating more linear storylines; it's not clear that players want or need fully branching narratives. And so on … Adams concluded, somewhat inconclusively, by figuratively pounding a few stakes in the ground with regard to quality in game narrative. For example, an NPC (non-player character) should never repeat itself unless the player requests it. Introducing the story through exposition is inferior. And no story event should ever happen the same way twice.
And that brings us to the keynotes. On Wednesday, Microsoft Xbox honcho J Allard got the crowd all revved up with razzle-dazzle and hype about the forthcoming "HD (high-definition) Era," but in the end offered a paltry few particulars about the forthcoming Xbox Next/360/whatever, the next-generation console rumored to hit retail later this year. (That's right; they didn't buy *my* loyalty with a free HDTV!) He also delivered some barely concealed barbs to Sony's forthcoming (2006, probably) PlayStation 3, referring to a "science-fair project."
The next day's keynote featured Nintendo exec Satoru Iwata, co-founder of Japanese game developer HAL (named after Kubrick's rebellious CPU), which gave the world the nebulous character Kirby. Iwata proclaimed himself a gamer at heart, gave a fairly entertaining talk, showed some footage from the upcoming Zelda title for Gamecube (looks great), and demonstrated some innovative new software for the DS handheld console. First was Nintendogs, the latest project from in-house game god Shigeru Miyamoto, which features a cute, realistic animated puppy that you can pet with the touchscreen/stylus, feed, play catch with, take for walks (while cleaning up its messes), and train to follow voice commands. The "game" will feature 12 different breeds, each with real-world traits. I tell you, between this and Sony's Aibo, it seems that the Japanese are trying to put real dogs out of business. Iwata also showed Electro Plankton, which also seems to be more a software toy than a game. The demonstrator created a simple, repeating tune by touching bubbles in an underwater seascape. He also recorded his voice as part of the tune using the DS's built-in mic.
The show floor had little in the way of true innovation, except for a new application from tiny developer Darkling Simulations, creator of the nifty DarkTree procedural-texture-creation software. Darkling describes Tactile Ink, slated for release this summer, as a paint program that lets you draw with realistic procedural materials in a real-time shaded environment. Designed specifically for game developers, Tactile Ink automatically generates tileable channel maps for detailed game environments. The version I saw was in an early (pre-beta) state of development, but I can tell you that it has two types of paint: cover and modifier. Painting with cover paint replaces any existing paint, while a modifier paint functions as an overlay, adding features such as a cracked surface. It also recognizes dimensionality, so that you can, for example, paint a water layer that fills deeper areas but doesn't affect higher area. Also, the deeper the water, the greater the distortion of the underlying surface. Spectrum will feature more info on Tactile Ink as it becomes available.
Aging wunderkind Cliff Bleszinski (he just turned 30!) of Epic Games/Unreal fame gave an entertaining talk with the somewhat dry title Dissecting Interactive Design. Most of the entertainment value came not from the talk itself, but from Cliffy B's asides, some of which were a bit risqué. To boil it down to bare essentials, the lecture was basically about Bleszinski's concept of the loop, which he defines as a series of actions visited and revisited by the player to accomplish a goal. A game can have short-term loops, such as collecting and shooting; mid-term loops such as navigating and earning money; and long-term loops such as leveling up and spending money. Likewise, real life features short-term loops like morning rituals ("eating Cheerios"), mid-term loops such as establishing relationships, and long-term loops such as getting married and having kids. Bleszinski discussed loops as implemented in a variety of games, including his current fave, Resident Evil 4 (he loves the dude with the blue flames who sells you stuff).
Once you understand how loops work, you can use them to improve games by, for instance, tightening them, using parallel loops (e.g., lockpicking and computer hacking), and transplanting loops, such as combining a fighting game with a role-playing game. Some of the reasoning/logic seemed a bit spurious, but in the end Bleszinski won over the audience by sheer force of personality and some well-honed cautionary conclusions: Focus on the experience over the list (of features); fight to keep the player from getting frustrated, which leads to boredom, which leads to your game ending up on the shelf; and maintain an air of mystery, which can involve lying to the player. But that's okay, because people sometimes prefer the illusion of freedom to real freedom. If your game resonates and is fun, gamers are willing to forgive a lot, such as the repetition in games like Diablo, Halo, and Final Fantasy.
Finally, we come to one of the closing sessions, a star-studded panel that filled the too-small room to capacity, with attendees covering almost every square inch of floor space. Entitled Burning Down the House, the session was an opportunity for developers to bite the hands that feed them, with relish. Revered developer Warren Spector, who recently quit Eidos to start his own studio, kicked it off with the unequivocal statement, "This business is hopelessly broken; we're really doing everything wrong." His ultimate point was that game-development funding needs to come from multiple sources, not just money-hungry publishers as it does now. Following Spector was IGDA head Jason Della Rocca, who complained mildly about apathetic game developers.
Next came a fiery, vitriolic rant from Greg Costikyan, who started working in game development in the paper-gaming days of 1976. Says Greg, game development was once about creativity, but that's over. The Microsoft keynote made his flesh crawl; there are big bucks to be made, but not by you and me. Iwata-san says he has the heart of a gamer; what poor bastard's chest did he carve it from? To explain the latter, Costikyan is critical of Nintendo for establishing the industry-wide practice, on the console side at least, of the hardware marker charging developers royalties based on manufacturing, not sales, as well as reserving the right to decide which titles will get published. Costikyan ended with a call for revolution (i.,e., gamers should get their own funding for innovative games and not knuckle under to publishers), and then fellow old-timer Brenda Laurel, a true radical, followed in a similar vein but somewhat less coherently by railing against "the spectacle," saying, "We are contributing to the damage that spectacle does to human beings by suggesting that the joystick provides real agency." Yeah, I didn't get her point either. Perhaps more to the point, she queried a bunch of L.A.-area teenage boys about Grand Theft Auto: San Andreas, and the feature they were most interested in was being able to see their own houses in the game. "Where are the games about geopolitics?" she wonders. I guess Laurel wants more-relevant games, but there's no guarantee they'd be commercially successful.
Concluding the panel on a more technical note, indie developer Chris Hecker bemoaned the upcoming consoles from Microsoft and Sony, whose CPUs he says will use in-order execution, which "sucks for gameplay code." Most modern CPUs use out-of-order execution, which Intel introduced shortly after the Pentium launch, and optimized "average" code (that is, software written by non-assembly programmers); that is, it lets bad code run fast. The return to in-order execution could result in a third to a tenth the performance at the same clock speed for average code. The only thing we can do about it is be prepared or hope that Nintendo uses an out-of-order CPU for its next-generation console. Or maybe do PC games.
For a remarkably faithful transcript of the above session and several others, plus some interesting opinions and worthwhile links, I recommend a visit to the excellent Wonderland blog at http://crystaltips.typepad.com/wonderland/. Blogger Alice is to be highly commended for providing this service to those unable to attend the show or certain sessions.
GDC and game development in general has traditionally been a male-dominated arena, so it was good to see a hearty attendance of women at the show. In fact, Brenda Laurel crowed that she'd been delighted to experience a crowded ladies room. Let's hope this trend continues.
I understand the GDC organizers didn't manage to procure Moscone Center for the 2006 show, so it'll be back in San Jose for at least one more year. Let's hope it returns to San Francisco for every year following, but wherever the show takes place, I'll be an eager attendee; if you have any interest in the future of gaming, I suggest you do likewise. There's nothing like GDC. Thanks to Chris Crawford for founding it, and thanks to CMP for keeping it going, and thanks to everybody else who makes it possible for an amazing show.
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: firstname.lastname@example.org.
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: email@example.com. - David Duberman
Spectrum founding editor: Doug Millison (www.Online-Journalist.com)
©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.