A Lecture for the 1997 Computer Game Developers Conference,
Copyright 1997, by Dani Bunten BerryI started out my career as speaker at this conference by delivering the banquet keynote address in 1990. I told everyone that if we want to reach the mass market in this industry we're going to have to become part of the main stream and stop being such nerds. I recommended that they go home, meet their neighbors, get married, have kids and to stop spending all their time alone in front of computers. I said something pompous like "Only when our products come out of a deep connection with real-life will they resonate with the mass market". I think that was when I coined the morbid quote that "No one on their death bed ever says 'I wish I'd spent more time alone with my computer'" to highlight the "people orientation" of the real world vs. the "thing orientation" of our business. Not mentioned in my bio is the fact that 4 years ago I changed pronouns. I tell you this partly so that those of you who "know" don't feel any anxiety about telling or not telling those of you who don't since I hate that kind of awkwardness. But a sex change also offered me enough humbling experiences to make me less willing to pontificate about what other people need to do. I switched instead to just sharing my insights on specific game areas. I've done round-tables and seminars on various aspects of designing and developing games. However, I've loved the big lectures given by my friends Chris Crawford, Brian Moriarty and a few others that eviscerate our industry, cajole us to think differently and galvanize us to try harder. For myself, I haven't felt driven enough by my beliefs, sure enough of my experience nor confident enough in my ability to climb back into that pulpit again -- till now.
What kicked me into motion was a conversation with a "marketing specialist" at a recent online game conference. He said something innocently grandiose about how great it is that the online service he worked for has got the full gamut of games to cover all possible demographic groups. In his view, as new products come up all he'll have to do is drop it in one of his genre slots and connect the dots between who's out there (audience) and the what they can play (content). I was dumbstruck. It's true that they had things from card games through "Warcraft" to "Doom" that featured different levels of intensity and abstraction. However, from my point of view all those games were holdovers from the pre-online era. I believed we had barely tapped the virgin territory of a whole new medium. Here was a 20 something whiz kid telling me it was all over but to stock those cyber-shelves and rake in the cash. Once I recognized this "new" point of view of his, I started seeing it in a lot of other places. Like the lack of desire by publishers and capitalists to underwrite even small experimental products. All they wanted to talk about were the fully fledged, competitively executed products that looked a lot like what was selling in the PC CD market. They were more willing to spend a million bucks on a "C&C" clone than under $100K on a promising concept for an original product. As the self-appointed life-long defender of people-oriented games in this business I had to speak up to save this potential new medium from being turned into another silicon valley meets Hollywood "sequel-itis" wasteland.
So that's why I'm motivated to preach. But am I qualified? In these days where spending six months with an internet service provider is enough to proclaim yourself an expert in the online world, it's a good question. I've done more original multi-player games than anybody else in the games business and the one thing we're sure about online games is that they will be multi-player. Of the dozen games that I've had published by Strategic Simulations, Electronic Arts and Microprose, 10 were multi-player. The first 7 of them used shared or simultaneous input (depending on the platform) and shared output. (In other words players were grouped around a computer with their own joysticks when possible or passing it when not). There were (and still are) numerous possibilities for social interaction and interesting play with the shared computer kind of design. However, there were a number of logistical issues related to getting groups of people playing games around a single computer (such as, it's not usually centrally located and people need to be "invited"). Online games "fix" these problems while still offering several of the social advantages of multi-player games. Thus, as soon as possible I switched to writing online games and my last 3 were of the type that each player had their own machine connected by modems. (Specifically, they were server-less, synchronous-state, real-time action-strategy games where all the code ran on the client machines). I did the first point-to-point game and the first four player network games published by major publishers. The last several years I've been a design consultant specializing in multi-player online games.
There are my motivations and my qualifications to speak. How about my confidence to deliver this tirade? On that I'm too stubborn not to try.
We're going to attempt to uncover why online games suck and what to do about it. First, I should share the good news about online games. A lot of people are having a lot of fun playing the online games they are being offered. "Quake", "Warcraft", "C&C", "Diablo" and their clones are doing an amazing job of convincing people that the age of online multi-player games is upon us. A whole sub-industry is being developed to bring more folks into that realm and to make money off them. Compared to this time last year, we now know that tens of thousands of people are willing to sign up and pay some amount just to play games online. Although, we still don't know the ideal system for financing online play we at least have some notions about mixing ads and box-office to fund the system.
What I mean in my title, "Imaginary Playmates in Real-time", is that for nearly all intents and purposes the current crop of games (and even the next crop that I'm aware of) have simply taken standard computer game genres from the pre-online era and replaced the AI opponents with humans. If you're playing one of those games, your interaction with those humans is at the same level as it was with the AI ones. What we're experiencing now is just the fact that people make better opponents. They will do more interesting things than any algorithm. Those of us who have been pushing multi-player games for years have known this part. It's just that this is such a tiny aspect of what having human playmates can mean. People can make you feel welcome and accepted. People can teach you and share with you. They can touch you emotionally. But in the current online games they are nothing but a few pixels on the screen and an occasional stream of text. So, I guess what I'm saying is that online games suck in comparison to what they can become. So, it appears I was exaggerating in my sub-title "or Why Online Games Suck". Guilty as charged. But, I got you here and paying attention and I promise to give you some suggestions on how we can create an online games medium that will make what's happening now look "sucky" by comparison.
Let me start by describing some of the institutional and structural problems of how computer games are currently being designed and built that we need to overcome in order to bring this new medium to life. The biggest and most threatening monster in our path is the distribution system that garners almost 80% of the revenue stream of games for virtually no value added. That's pretty harsh isn't it?!
What we have now is a multi-billion dollar computer game business. Lets look at where that money moves. Most of the accumulation of wealth occurs as a result of the sales and distribution of our designs. Retailers, who display our games to their audience, take the lions share of the revenue stream - between 50% and 60%. For this they present the edges of boxes on shelves under labels they consider descriptive. At this particular point in time those labels are thematic more often than not. You'll find our games categorized under Science Fiction, Sports, War, etc. Under these titles you will find everything mixed together. I've even seen "Command and Conquer", "Duke Nukem" and "Myst" under Science Fiction for God's sake! How do retailers not know that our audience fall into categories by whether they like strategy, action or adventure games not whether it deals with the future? I can't imagine how players find games they are likely to enjoy these days.
The next big player in terms of accumulation of wealth in our industry are the publisher/distributors. They take 35% to 40% of the pie. For this they put our software on CD's, put them in boxes and ship them to retailers. This is not to say they don't make a creative contribution. They turn their marketing gurus loose to invent the packaging that makes our games sound like something they want to sell. (Not that they constrain themselves to describing the actual game they are putting in the box.) The backs of boxes seem to be designed to make it harder rather than easier to figure out what the game is about or how it really plays. It's goal is to get that box off the shelf and whatever it takes to do that seems fine to the publisher. There are also the full-page, content-free ads that the marketing folks invent as another of the publisher's offerings to the success of our products. But by far the main contribution of the publishers is their input into our design process. They help us follow the lead of the last big seller. Without their help we might flounder around and make something original.
Thus, about 80% of the money goes to people who make their "votes" based on those superficial things that make the "sizzle" but not the "substance". Now, I don't want to imply that cool animated cut-scenes and appealing segues are not nice things to have in a product. They can help players get into the story-line and enjoy the rest of the product. It's just that in too many cases products are getting sold just for their icing regardless of whether there is cake underneath it or not. (A friend described that as "A big hole where the fun should be"). And I would contend it's the screwy distribution system that not only allows it but encourages it. In additions, corporate buyers, like all "market analysts" only know what sold last quarter and that's what they base their decisions on for this quarter. Hence, the chute is greased for sequels and "me too" games. Add to all this the data point that most computer software is bought when a computer is first purchased. I've heard numbers that say 80% of all software is sold to new users within a month of buying their computer. What that means is that sellers don't even have to worry about delivering trash and pissing off their buyers. They're not coming back and will be replaced by the next sucker in line anyway! A final new trend is that very few retailers will let you exchange anything but defective software in a "like for like" swap these days. I think you can see how the system has nailed the coffin shut on innovation. It has ensured both the continued success of "sizzle" and the fact that most "steak" is tough if not rank.
OK, maybe I'm exaggerating a bit. But none of these aspects of the current system favor success in the new online medium we would like to envision. The shelf-space online will be unlimited and deception will be punished by players not returning to your site. Originality will be required but "publishing" is unnecessary. Although sizzle won't hurt, substance in the form of addictive playability over time is essential. Instead of software that is primarily sold to new owners, the online environment will be peopled by savvy consumers who are wired into their own interest groups. Revenue accumulation with almost all expected financial models rewards repeat customers in the online world more than initial sales.
I'm aware of only one "multi-player only" game that was published by a major publisher - my own Robot Rascals by Electronic Arts. There's a reason. Solo sells. Or at least it has until now. All our metaphors, imagery and concepts as designers have gone to support the single player game as the prime feature of our products with multi-player as an occasional option. Solo games have a whole different style. Solo-games could concentrate on entertaining just one user so all the resources of the machine could be devoted to that end.
Brilliant graphics and sounds could be used to set the scene for players making their experience more compelling. Segues and cut-scenes are triggered as appropriate to that audience of one. No provision needed to be made as to what the other players would be doing while the awards ceremony visuals were running. Pandering to the only player didn't create any tension with the others who might not be doing as well. In solo-games they were all AI stand-ins who would mindlessly wait while you gorged on eye candy. And the pandering didn't stop with their ego's either, solo games needed to push the platform too since one of the biggest perks for many hard-core players was showing off their hardware.
The features of a solo-player's game are geared to take advantage of the "learning curve" involved in the process of mastering a new game and its environment. There can be whole groups of features that only show up in certain environments as the player advances and learns to play. Players of those designs not only accepted the idea that more was better, they demanded complexity for it's own sake or else they thought they were being cheated.
"Kitchen sink" design also has an advantage to solo-player games besides the titillation of the player. Designers could hide the limitations of the AI opponent behind the veil of added game elements that kept challenging the player when his opponent couldn't. Another aspect of the AI also influenced solo-game design. Although subtle nuances of pattern recognition might be trivial for a human player, the most blatant patterns could be a nightmare for AI. Hence, there's a tendency to make the externals of solo-games very conventional. Designers discard opportunities for interactions with audio-visual cues in favor of algorithmic and concrete presentations. Rather than allude to something with subtle patterns or sounds that humans excel at, designers used the same kind of logic in their representations as they used in their artificial opponent's analysis of the world. There aren't "maybes". There are only "zeros" and "ones". In addition, the need for competent AI required that the internal models be computational which computers "love" rather than heuristic which humans enjoy.
Taken together these design elements may make good solo games but they conspire to make games that are poorly suited to humans playing with humans. Rather than "over the top" production values, online games will reward small downloads, good multi-player balance and smooth play experiences. These will preclude pauses for "cut-scenes" and pandering to a single player's ego or hardware vanity. Since people derive much of their challenge while playing with each other from anticipating each other's actions, complex feature sets are unnecessary. Simple sets of rules, consistent over time make multi-player games more accessible. However, subtle nuances in audio/visual presentation of products make for much richer play experiences for human opponents. And, finally, heuristic (rules of thumb based) models are much more appealing to players than complex numeric systems and once again play to the strengths of our brains while not sacrificing playability.
Those are the two biggest obstacles to be overcome in order to bring online games up to their potential. Both are very daunting and will probably keep a lot of good talent from coming to this area or being effective here. It's even possible that the distribution monster will attempt to sabotage this medium to protect its turf. Just the simple fact that their financial model sells the CD but gives away the "play time" makes "pay as you play" and "advertiser supported play" less appealing. However, I believe once we have some content designed specifically for online multi-player play, it will be no contest which kind of product people want to play. The concept of continuing to tack the online option to basically solo CD games will look pitiful by comparison. Even if the online media creators have to move to some kind of sales of their software, without middlemen, manufacturing and shipping we could be pretty darn competitive. I used to tell folks who bought my games that by the time EA got done with it all, I only got about $2 a copy so if they ever wanted to clear their conscience for pirating "M.U.L.E." they could just send the $2 straight to me. The shareware distribution folks seem to make much better than that per copy.
One of my favorite game designers is Sid Meier. When I met him in '89 and was going to do a new game for Microprose I told him I was torn between doing one of two games. They were both board games whose design while solid in their own medium would be very interesting to work out in the computer game field. I wanted to do either "Axis and Allies" or "Civilization". Sid talked me into doing the WWII game (Command HQ it was called) and he did Civilization. Not that I would have done the amazing job that Sid did with that game but it does give one pause to consider the ways fate works out. Well, anyway, Sid just did a "Soapbox" piece for the magazine "Game Developer" where he applauded the current situation that is allowing small creative shops to take a prominent place in the design of innovative products. I assume he's alluding to groups like Blizzard, Westwood, ID and his own Firaxis and to a certain degree I agree with him. To whatever extent the big boys (EA et al) will farm out money to these groups to do original stuff, it is more likely to succeed than trying to push a new idea through their in-house development system. However, the budgets (and hence the resources) that are required to build products for the CD market are so large that risk must be reduced and thus wild impulses constrained.
I think a much better case can be made that small creative shops will be the province of the new online medium where so much is unknown about what will work and hunches, instincts and wild impulses can and must be followed. The budgets need not be the astronomical numbers ($1million plus) that are talked about for CD products. Online games need to be small (under 10 meg) so the download won't be prohibitive. Since they are multi-player they need to be simple to play, without complex expanding features that make testing and debugging a nightmare during development. There's no place for fancy cinematic segues and cut scenes so the art requirement is just what's needed to support game play. And, finally since rev's are no problem when you don't have to master a CD and ship it to several thousand retailers in time for the ads to hit, you can make the whole development process iterative. The game project can evolve and be financed in stages and still be productive. Taken together, I think these items create a wonderful opportunity for innovative products financed inventively. To me, it's a situation very much reminiscent of when Trip Hawkins was founding Electronic Arts. He approached several game designers (including myself) in the early '80's to empower us to do the creative things that he believed the new medium (at that time floppy-based-games) could accomplish. (The "Can a computer make you cry?" ad was the hallmark of that era). A lot of neat stuff was done before the business was taken over by it's distribution system. We need another "Trip" to seed this next growth spurt. If the new "Trip" is in the audience, would he or she see me after this talk. Have I got a deal for you?!
Another opportunity that I believe the online medium offers is a new demographic landscape. Although to look at the "successful" games online at this point you'd think we were stuck with the 18 to 35 year old male audience that populates the CD games world, it ain't so. The demographic of web-browsing (see "Online Magazine") has an average age of 33 and is 31% female. They come primarily from educational and computer-related occupations. Almost 60% have a college degree or better. They are information consumers and have an average household income of $59,000 (these two numbers make advertisers excited). They have grown 10 fold in the last year and Nick Donatello of Odyssey Research predicts online entertainment will exceed $1 billion in revenue by 2000. Those are some impressive numbers! Designing games to reach them will be an amazing challenge. We will need to think creatively about what a game is. Unfortunately, I have no idea what might be. I can tell you some of the pieces that make for good multi-player games but my guess is that the big breakthrough will be as like our current games as "Doom" is like "Pong".
Here comes my annual punch list of things to consider when designing multi-player games
updated and expanded from last year based on what we've learned: