The Future of Games

MAN: Before the panel starts I want to acknowledge that an event like this doesn't come off without an enormous amount of effort by the staff, the students here at MIT, the folks at the interactive digital software association and their collaborators. So I just wanted to take a moment to acknowledge all the various contributions that went into putting this event. [APPLAUSE] That said we're going to plunge now into the future of games which is our final panel. I hope it will not be the end of the conversations we've started here. I think there's a lot of exciting opportunities for the kinds of exchanges we've been involved with. And these kind of exchanges are central to the comparative media studies program here at MIT. And we will be having future events of various kinds. And hopefully people will stay abreast of the media in transition web site where we publicize the events that are forthcoming. And again, watch for transcripts of this which will be up as soon as we can get them up. Our first speaker is Peter Molyneux. Molyneux co-founded Bullfrog Products in 1987 and created a new genre of computer games, the god game, with the release of Populous. Since then he has been responsible for a string of massive selling games, including Powermonger, Theme Park, Magic Carpet and Dungeon Keeper. In 1997 he left Bullfrog Products to form a new game development company, Lionhead Studios and is currently hard at work on the company's first game, Black and White. HE's spoken on the subject of development of computer games at the America Museum of the Moving Image, the British Film Institute, ICA in London and the Dortmund Museum of History and Culture.

MOLYNEUX: Thank you very much indeed for inviting me over. What I'm going to do today is there's not an awful lot of time so I'm just going to very briefly touch on what my thoughts are for what the future of gaming is. It's a huge subject matter and we could talk about this for the next two years. So what I'm going to do is talk briefly about the thoughts going through my mind when I designed the latest game that I'm doing called Black and White. And show a short videotape of the current state of play of the game.

MAN: All right.

MOLYNEUX: Which should be quite cool. [LAUGHTER] I haven't really seen the videotape and you must forgive it because it was just plugged in the back of my machine and so there are some bits that don't work and stuff like that. So what was running through my mind, firstly the future of games, what it should be and what it must be and what it has to be is mass market. We have to realize that we are making games to entertain people and as many people as we possibly can. That means that there has be games that are invented that just simply do not exist today. And they have to do one thing and the most important thing they have to do is entertain people. In whatever way that is. And not just hobbyist people. Not just dedicated gamers but a whole mass of people. And how do you do that? You've got to make games incredibly, incredibly deep. You've got to use technology that would even make something like [UNINTELLIGIBLE] creak at the seams. And you've got to marry that with accessibility. You've got to make games so accessible that people can go home, sit down in front of their TV and say shall I watch a video, shall I surf through the channels on cable or shall I play a computer game? And if we can do that then computer games can really make a huge difference. They must be social and they must be personal. And what I mean by that is why should computer games be the same for everybody? Why shouldn't we create a game that allows the game to look at the person who's playing it and change itself to cater to that one single person? And that is a pretty high concept in itself. What we need to achieve all that are some amazing visuals. We need movie quality visuals. It's no good to have these iconical figures any more. We've got have things that people recognize, we've got to have scenes that look like they're out of a movie. It's just an excuse as far as consumers are concerned. They don't really understand whether or not we use 120 polys or 10,000 polys to create a character. They just want to see a real face with real tears and real sweat and real hair. We must have games with compulsive narratives, compulsive stories, as compulsive as any soap opera, as any film. And they must be of such sophistication that they can monitor what the player is doing. And we must have AI characters within virtual worlds that are as realistic as possible. And we've got to have incredible AI engines that allow us to change characters as people play them. We've got to have interfaces that are seamless. And all these ideas are bubbling around in my head when I was thinking of games. And quite frankly, it scared the crap out of me. I can't think of any other way of saying it. Because all of that is a huge amount of technology. What I'd like to do now is just introduce you to Black and White and talk about how we at Lionhead have approached some of those problems and how we're hopefully on the road to solving maybe a few of them. So what is Black and White? Black and White is a role playing game. A role playing story based game where you play the role of a god. And you are presented in the story with a world. At the very start it's a very small world. And you're told that you can do anything you like in this world. You can be unspeakably cruel and mean to all the little people that you see. Or you can be wonderfully kind and benevolent. And what the game doesn't do is say you must be good to win or you must be evil to win. It says you can whatever you like. And within that story, very, very early on, you'll discover that you have the ability to bring up a creature. And this creature unconditionally loves you. And you can teach it anything you like. And eventually you give it a personality. Whatever personality you like. You can make the mass murdering killing machine that will go out and rampage and destroy everything in its path. Or you can make the wonderfully kind, benevolent, caring character. It's totally up to you. Again, the game doesn't say thou shalt do this. It says do what you want but you have to progress through the story. So if we run the tape, if that's possible. And I'll carry on talking explaining. There's no audio in the tape really to speak of. I'll just talk while the tape is running. That's [UNINTELLIGIBLE] [LAUGHTER] It's called Black and White because you can be anything from black to white. So the first thing to notice is, I apologize for the no audio, is that there is no interface. There are no icons. There is no [UNINTELLIGIBLE] control mechanism. The only interface you have into the world is this hand. And with this hand you can do amazing things. You can pick up things, you can throw things, you can cast spells, you can cast magic. And the first objective really was to create this amazing wonderful world full of all these little people and all these little characters. And allow you to do anything you want in that world. This is the creature I was talking about earlier. This is how you tell him off. You hit him and slap him with your hand to say that things are bad or good. This is the citadel. It's just running through a whole sequence of scenes. This is the inside of the citadel where there's certain rooms. One of those rooms is actually where your creature. This is when your creature meets another creature. It's running through the visual side. So the idea is to construct this world and allow people to do anything they want in it. This is how you cast magic, how you throw spells. We have this thing called gesture recognition which allows you to gesture your spells rather than clicking on different keys. And all of this is done with just two [UNINTELLIGIBLE] and a mouse. No keyboard, just two mice button and a mouse. And these are the sort of beautiful scenes that you can get. All the weather is totally realistic, a total complete weather sim. Full physics engine in there as well. Meaning that you can manipulate and move anything you like in the world, [LAUGHTER] even trees. It's quite cool. [LAUGHTER] You can do unspeakably cruel things to the little people in the world. And all of this runs on your average home PC, amazingly. Just waiting for it to get to [UNINTELLIGIBLE]. This is a multi player game in progress. Some of the natural weather conditions that we have within the world. I think it's looking incredible. I think it's looking really nice. The main thing, though, the main trick with the game, the most amazing thing is that creature that you see there. The idea was to give you a creature, the player a creature, at the very start. He's always showing me up. Give you a creature, a little tiny young creature, and allow that creature to grow up as you give him a personality. And as you do he changes physically to represent the personality you're giving him. That means you get this incredible thing within Black and White. You get the ability to give a real personality to a character and that personality to be as unique as you are. And then you can take that creature online and you can show him off to other people. You can communicate with him to other people, you can trade things to other people, trade skills to other people. It is really incredible, the things that the creature can do. So while this video is going on has anybody got any questions about Black and White or about [OVERLAPPING VOICES].

MAN: We're going to go to questions now anyways. Could you go up to the mike? I don't mean to run you around. And give us your name please.

MAN: My name is Rick Goodman. And my question is, do you think that in the game of Black and White that a victory condition is important or is it not important?

MOLYNEUX: It's incredibly important actually. I mean one of the reasons it's so important is that in all my previous games the question that I hated the most from the press, that I absolutely loathed, was what's the point of the game. And I never had a real compulsive answer to that. I did god games before this where you're regarding the world and you had to dominate and you just played through one level or another. And what it missed was a story. Was an unfolding story that gave people objectives. Because it's all very well giving people untold powers but unless they have a focus for those powers it soon becomes very boring. So it's really putting a story within a god game, a role playing game within a god game, that was, I think, the key thing. Which has turned Black and White into [UNINTELLIGIBLE]. You can't really appreciate it without me giving you a full demo.

WOMAN: I'm Christina Moriarity. My concern is that it requires a two button mouse. And I won't give up Mac for anything. [LAUGHTER]

MOLYNEUX: A very good point.

WOMAN: Up until now game developers don't care about us. But with iMac you're going to have to start caring. How are you going to make a game for me?

MOLYNEUX: We are actually doing a Mac version of Black and White. And it is a real challenge to try and put all the functionality on one mouse button. But we're going to do it. As it turns out all it means is touching one key on the keyboard. I mean I believe in the Mac. I believe it's a great platform that should be supported by more games companies.

WOMAN: Thank you very much. [LAUGHTER]

KEONG: My name is Douglas Keong. I'm with Harvard's technology and education program. I'm curious with this game and with Dungeon Keeper, these games have shown us it's OK to be bad. And I'm wondering here where you give people the choice. Do you have any expectations about how many people will choose to play the evil side of things and how many people will play the good?

MOLYNEUX: The interesting thing is you can see where this came from is that a lot of people, after me doing Dungeon Keeper, said you're making games encouraging people to be bad. So I thought, right, this time I'm going to make a game where you can choose whoever you want to be. And you show this to any 15 year old kid and they'll say cool, I'm going to make the most despicable horrible world with a really mean cruel nasty creature. But you know, when they actually sit down and play it they just can't do it. They can't bring themselves to be that cruel and that mean and that horrible, especially to their own creatures. It's a fascinating insight into human psychology. In that what I've found is that for most people it's as hard to be truly evil as it is to be truly good. And you can see people's worlds. What I didn't explain is that everybody in this world changes to reflect the sort of player you are. If you're cruel and mean and nasty then your world will look dark and scary, the trees will have no leaves on them, the weather will be all horrible. If you're really good then it'll be beautiful blue skies the whole time, the trees will blossom. Maybe everything will be a little bit too bright and nice. Maybe a little bit too squeaky clean. But it's that reflection of the gamer that is the fascinating thing. It is the first game that I know of where it really truly reflects what you're like as a person.

MAN: I'm James from the MIT media lab. We've been mentioning [UNINTELLIGIBLE] throughout the conference more in terms of story and structure. But in a recent [UNINTELLIGIBLE] interview you made some interesting comments about borrowing techniques in terms of cinematography and editing. And how, for example, when you have these 3D environments and you're placing the camera angles, the need to balance certain cinematic techniques to achieve particular purposes. And yet at the same time you have to accommodate the user's interactivity, not just cut randomly to particular angles. Do you feel like you've succeeded with Black and White? What were some problems and issues or [OVERLAPPING VOICES].

MOLYNEUX: That is a very big challenge. Because if you try and predict what the person wants to see you're going to fail every time. I mean you may succeed 50% of the time but the 50% you don't succeed is what will stick in people's minds. So my feeling is that the solution to this particular problem is to allow people the flexibility to choose their own camera angles but to sort of suggest them through an avenue. And there are occasions in this where you see that actually happening. You can show them something. You've got to give them a choice of whether they want to see it at this particular point in time. And what they want to see in the particular point in time is then revealed in all its glory. But allow the person to come out of it and view it in any angle he likes. So the best camera man, in my mind, is the player. He's the best camera man because he's the one who knows what he wants to see. You just have to make sure that you show him the thing that he should be interested in. But it is a very, very tough problem. There are people in Hollywood who spend their whole time deciding on camera angles. And we're making games of such flexibility that we should be taking that much dedication to it.

MAN: My name is Scott Dalton. And I've played Dungeon Keeper, I've played Populous, Magic Carpet. All fine games. And that looks cool. [LAUGHTER]. I've read about Black and White that your creature can take on your personality. And if you're generally good he'll generally be good to the people in the game. How difficult was it to get that incorporated into the game? Or what particular challenge did it pose to you to get that to happen inside the game?

MOLYNEUX: To get this working we needed a state of the art AI engine. We first started looking at a hybrid version of neural nets. That wasn't good enough. And we ended up really choosing our own technology to do the learning of the creature. [LAUGHTER] Sorry. What he has to do, this amazing thing that could seem like he could learn anything. And you can teach your creature to do anything. You can teach him to build things by showing him examples. You can teach him to be aggressive or passive. Any of those things. But it was a massive challenge. And it's only down to the brilliance person at Lionhead called Richard Evans who came up with the final model. But we've had some really funny times with it. When we first got the creature working the first rule, the first thing that we got him to learn, was think what to eat. And the first time we really got him visually up he was standing there and there was things all around him and he was standing there looking down at himself and he was just reaching down, this is absolutely true, I swear, he was just reaching down. We couldn't understand what he was doing. It was only after we debugged it we found out that sure enough, he looked around in the world to see something to eat and the most tasty thing that he could find to eat was his own leg. [OVERLAPPING VOICES] you can eat anything but don't eat yourself. Tons of really, really funny things that's happened in the game room.

JAZZIERI: Darius Jazzieri [SP?], student here at MIT. I guess we saw some of the ways that you can be mean to your creature but how can you be nice to him? Is it just you're nice by not doing bad things to him or there are actually constructive good things you can do?

MOLYNEUX: One of the nice things that you do to him, stroke him. You can slap him, you can stroke him as well. You can give him things. You pick something up in your hand and you can actually give it to him and he'll take it out of your hand. He loves playing. When he's really young he loves playing with the little villages here. If you're bringing up a kind creature. The only unfortunate thing is when he gets older he gets bigger and he gets taller. Until in the end he's about 200 foot tall. And if you've got a kind creature then he gingerly steps between the little people not wanting to hurt them. So there's lots of ways to be kind and cool to your creature. He loves to see new things. If you've got a creature that's outgoing. But the most amazing thing is that he is really reflective of people's personalities. There's people at the office who are timid and introverted and their creatures tend to be timid and introverted. There are people that are outgoing and extroverted and the same with their creatures. So yeah, you have ways of loving your creature. In fact, the linchpin of this game is the relationship that you build up with your creature. If you don't love your creature, if you don't feel like he's yours, then I've failed [UNINTELLIGIBLE].

MAN: Peter Walsh. I wondered, did you do marketing research in developing this game? Or was it just an inspiration? And the reason I ask is because it strikes me as a classic fantasy for non game players dealing with computers and the Internet, which is to take on personalities that are not their own. To become something other than they are in everyday life.

MOLYNEUX: I did the most intensive marketing focus group I possibly could on three children in my life. They told me everything I needed to put in this game. And just by watching them grow up and learn. And as you say, occasional children want to be other people, we all want to be other people. And just watching that was absolutely my inspiration. And knowing when they sat down and said that's boring, I knew that was wrong. And when they said that's cool I knew that was more right. And on a personal level that is much better than 100,000 focus groups, I think.

MAN: My name is David Anderson from Boston University. I've played a game called Creatures Two which is really similar to this where you develop a character. And I got really frustrated in that game. How does your game help the user actually develop a character in a friendly way?

MOLYNEUX: One of the problems with Creatures Two, and I'm not knocking because it's a great application of AI, the real problem was in the interface. Is that there was a million things for him to play around with and you didn't know how important a rule was or how important it wasn't. There was no story to flow through, there was no objectives for the main character. Why were you teaching him to do these things? Why did you want him to have a creature? They just posed the question because you wanted a cool creature. That's why the story is really important in this. It gives you objectives of why you want your creature to be bigger. And also the interface is incredibly simple. The teaching mechanism is incredibly simple. There's two basic ways this creature learns. The first one, you've seen it, where you slap him and you stroke him. And that's saying he's just done something, he may have a kicked a little person or picked up a little person and you may not want him to do that, that's saying bad, don't do that. That's one way he learns. Second way he learns is by watching you play the game. If you go out and do unspeakably cruel mean things, just like any kid he will then think that's cool to do that. It's cool to be cruel and mean, it's cool to be nice, it's cool to be kind. So he learns by watching you play, by being inspired by you, and he learns by your direct interactions. So he's part of this thing, not the whole of it. And so I hope that we've shortcutted that. I mean I kind of know we have because there's people's creatures that are many, many years old. Game years old.

MAN: I'm afraid that I have to be the mean god now. It doesn't seem fair to make our next speaker compete with the images so if we could shut the video down and we'll move to the next speaker. Who is David Perry. Perry started designing and programming video games in 1981. After authoring numerous books Perry moved to England to develop professional games for Mikro-Gen, Elite Systems and Probe Software. In 1991 he moved to the United States to work for Richard Branson's new company, Virgin Games. There he headed up the team responsible for hundreds of millions of dollars in retail sales through award winning games. In October 1993 Perry formed Shiny Entertainment which so far has generated six highly acclaimed titles. Shiny now focuses on PC based games and this year will release the highly anticipated games Messiah and Sacrifice.

PERRY: Basically what I want to do is slightly different angle. I actually wanted to cover a whole set of different topics but do it reasonably rapid fire. Really just to create a discussion. Because this is an interesting forum. We give speeches every year. We're part of the game developers conference. And we tend not to have people like you guys sitting around that won't have heard these speeches before. So I'm really interested to hear what you have to say about this stuff. Because it makes us think. And Peter and I are always listening for that little thing that makes us think harder and better. I think we learned a lot from that. So let me just fly through it. I've got most of this noted out. Over the next five years video games will have survived a full family life cycle. Starting around the mid 1970s games have been around now for about 25 years. New parents today with children themselves have actually grown up in the video game era. And these gamer parents are very, very likely to actually endorse games or let games come into the home. They're not going to see them as the enemy. Basically I think parents of today actually understand what the video game thing is all about. And I expect in the long term, and I'm looking forward to this date, to commonly see parents actually playing the games with their children. Instead of it just being a clean buy for the child I'm hoping to see parents, I see it with like my father will look over my shoulder sometimes and say that sounds just like a real soccer game. And you look at the screen, it looks just like a real soccer game. And I see this guy who's not interested in video games getting drawn into it. So I hope to see more of that in the future. But basically what's happened is we're getting lots of new hardware. And with all of this fantastic new hardware, which we all love, we get incredibly better visuals. Today all 3D games are still made of polygons. The more polygons, the more incredible the image is that we see. The problem is that on PC games there are endless configurations to the video cards and the processors and the sign cards all around the world. So basically we actually as game designers have no idea. Llike Peter doesn't know what this guy in Greece is going to be trying to play Black and White on. So it ends up kind of limiting your game design. At my company, Shiny, we've been working on it for the last three years and heavily invested into a way to basically deal with today but also to look forward to tomorrow. So what we've done is we've started planning for that super high polygon future. And what we've done is we've created technology which we call real time tessellation [SP?] and deformation. There's plenty of research papers on it but to actually make it work in the video game world is a lot more rare. There's only a few different companies working on this now. So basically what it does is this engine, while the game is running, it's constantly monitoring the performance of your hardware. It's actually stress testing the hardware while you're actually playing the game. So basically what the game is doing is when it sees your machine start to choke then it eases back the pressure that's putting on the hardware by removing polygons from the scene in real time. And it's actually deciding polygons that you're not really going to notice. Then if you were to drop a really complex character into the world or you caused a massive explosion it automatically will steal polygons from all over from other characters and basically let that happen seamlessly so the machine basically won't suddenly turn into a tortoise, it won't come absolutely to a standstill. And basically this makes it for us a lot more exciting because our engine effectively is pushing whatever machine, and we don't care what you have, we know that your machine is going to run at 100% of its potential 100% of the time. And that's actually quite a nice idea. So basically you are going to see the richest display that your machine can handle. And that's important because a lot of our competitors are always trying to guess that average par when their game is released but they don't know when their game is going to be released. And so on some machines they're going to find their game actually does end up running slowly or else they're going to find out that they're not really pushing the hardware at all. So business wise it means that our games are designed happily to support Moore's law. So by the time the game finally hits the shelves what's actually going to happen is the person who finally buys the game, if he's got some brand new hardware, he's going to see visuals in the game that we as developers never saw. So we effectively put more work into the game knowing that at some point somebody is going to see this stuff but on our development systems in the office we never got to see it. And that's actually quite a nice thing as well and it works. That's the best stuff, by the way, the stuff that works. [LAUGHTER] It's actually causing quite a stir in the hardware community right now. We're getting lots and lots of attention because of this. My favorite kinds of technology are actually the ones that don't just improve the visual or audio quality but that actually give us a chance as designers to design better games. So if we have to always worry about this theoretical limit of what somebody's machine can handle then we have to obviously start limiting our designs. And that means we're constrained before we even begin and that's bad. So basically the normal solution is to use a thing called level of detail. Very simply it's you have multiple versions of the same character. It's actually a lot more work. You'll redraw a character multiple times. One big, the medium one, the small one and distance. And that effectively means that you can reduce the load on the machine. Basically using our scalable technology it actually gives us infinite levels of detail. So as we slowly move characters closer or further away you'll see the polygons [UNINTELLIGIBLE] quite correspondingly. And this lifts a giant design shackle, certainly off our teams. Because we're able to now put 100 or 200 characters standing all around. So for example, if you look at Black and White, our next game Sacrifice looks just like this with the open terrain. But I have can 100 people, big people, standing all around me. And that obviously helps a lot.

MAN: That's my phone. Just ignore it.

PERRY: That's obviously really good if you're designing war games or something like that. Another interesting part of the technology is that by recreating the polygonal character in real time we can also add influences without really very much penalty at all. So what we've done is we've added subtle volume effort lighting on each polygon. And we've also added skin deformation to show the bones pushing through the skin and skin stretching as the character is actually moving. It also allows us to add random artifacts, which is something that Peter has actually got the best demo of right now, which is if you've seen games, a lot of war games or like Command and Conquer and things like that, all the characters in the game of a similar class or the same class are identical. So basically one archer looks like all of the archers in the game. The reason you do that is just simply because it's a lot cheaper way to store stuff. But with 3D, especially if we're processing everything in real time, we can actually add all those subtle artifacts to make one creature have a longer neck or a bigger head or a big fat tummy or whatever. And it actually will create that nice variance that makes people look like an audience, like you guys. Anyway, let's keep moving ahead. I think Peter's actually doing a really good demo of that. Some of my friends right now at Cal Tech are working on a development. They're developing this system which will actually take motion capture data. You saw how the characters there will run around and do things. One minute they're happy, next minute they're sad. But what they've been developing is a system for blending lots of different styles of a walk, for example. So they record somebody walking normally, then somebody walking all happily, then somebody walking a bit sad. And they can actually blend from one to another. So you can have a guy happily stepping over a dead body or he can be sad as he steps over the dead body. So effectively that's another nice way to add more character or personality to the characters without actually having to record a billion combinations. The extra quality in game displays and the constantly improving processor par also help us move forward the immersive experience. In the old Pacman days very crude characters kept, I think adults were thinking, certainly the conversations I had adults tend to think of video games as being some childish toy. This silly space invaders thing. And to be honest, a lot of adults still think they are. They haven't seen what we're really doing and playing so they still think we're in that weird Pacman-y, Space Invaders thing. But nowadays we can cyberware scan bodies. We can put a laser on somebody's body and scan it in absolutely perfect, every little detail. And we scan the texture so you'll get the texture of their skin and everything else as well. And then we can use motion capture to get the subtle nuances which is part of real acting. There's a big difference between our programmers doing the motions and then getting a real actor to do certainly good motions that actually will progress a story. I think going forward once we start to get these photorealistic actors –

PERRY: I think it will actually help make them appeal, again, to a wider audience. It will actually help the mass market. If anything I hope it actually starts to attract the other 50% of the world that tends not to play these games, girls. That would be great. The problem, however, is that better characters, or the better they look and sound, then the production costs are really going to skyrocket. So you're going to see an awful lot of video games go bust. They're going to try to do it but they're not going to be able to deliver. So my prediction is you're going to see a lot of them go bust over the next decade. And you'll see a lot of symbiotic relationships happening between Hollywood, special effects companies, artists that leave the best companies forming artist guilds that then freelance out for software development companies. But you're going to see a lot of moving around and action just based on the fact that games production is going to get extremely expensive. And some people, like Peter, is very lucky to be able to actually, he's given the chance to really make what he wants and he's also given the resources to hire the people that he needs, whoever they are. But that's not as common as you'd like to think. Basically these relationships, if we get them in the future, I think will actually help us reach the goal. To touch again on this emotion thing, because I'm concerned about it too as a designer, we do strive to keep creating new emotional states. We do do it. It's not like we're ignoring it. Basically our goal, ultimately I'd love to be able to make you scream at any time that I chose or make you cry any time that I chose. If we could get that control that would be something that would be certainly another nice tool in our designer's tool box. [LAUGHTER] Basically I think, just my personal opinion is that certainly there are people that have felt sad with some games today. But generally not the mass amount of people. Because partly it's our fault. We really don't try to really make you care. We just throw people out there and let you shoot them. But the fact is if we do try I think it is more rarely the case that you really do really sincerely care about the character. And that's because my belief is that a character that you care about is somebody that you like and you want to be with or you feel they're fun to be with. So if you have a sidekick in the game and he gets killed generally you think damn, I've lost some of the firepower that I had with me. But the fact is if this guy was really cool to be with and the things that he said and the way that he reacted to things was funny or entertaining you'll miss him. And right now the missing part, I think, is missing. So that's something I'd like to keep pushing for.

MAN: Two more minutes.

PERRY: So this is where we can learn from movies, books and psychologists and that kind of stuff. Quickly I want to touch on AI. Personally I still think AI is appalling. 90% of video game AI really is pretty damn bad. I think that's actually why it's so much fun to shoot things. Because the AI is so bad and the characters are so annoying. [LAUGHTER] So basically the reason for that is a lot of the time we try to just code the intelligence and you can't do that. Real intelligence is them actually thinking for themselves. There's no easy algorithms that we can copy out of magazines or download from the Internet that will do that for a video game today. And so you're seeing lots of people touching on it. But still if you look at culture, and culture, when people who recognize each other in any country in the world meet, they'll nod to each other and that's a perfectly, it crosses cultures, it's real thing that you would do. In video games you'll see people, something that's not normal to do would be, for example, if I went over and started walking into the wall. But in video games characters don't nod but they do just walk into walls and look completely stupid. Another limitation of animation which makes characters look stupid would be say there was a beam in my way, in a video game if I'm walking along, or in real life I would just duck under, but in a video game my face will hit this beam and I'll just keep walking with my face rubbing against the beam. And that's just absolutely standard. And there's a lot of people just accept it now because that's the way it's always been. So I think you'll start to see people start to care more and make people use their eyes and their ears. I think they'll actually try to make characters that are actually thinking and using their senses the way we would. And that would make a huge difference. To wrap up I don't know how many of you people here are people who play video games but no matter who you are or where you are we're going to get you through cell phones or we're going to get you through cable boxes. You're going to be playing video games whether you like it or not. [APPLAUSE]

MAN: Now it's your chance to get him if there are questions from the audience.

WOBBLE: Hi, my name is Brad Wobble [SP?]. Bitching about AI is a very common theme when talking about games. And I'm wondering what do you think is the root cause of the current lack of AI? Is it a lack of understanding of what constitutes intelligence? Or lack of skill in transcribing that understanding into the game? Or just lack of horsepower? I mean obviously it's some of all three.

PERRY: I believe at MIT your robotics department, it's basically very similar to video games because they have these characters or this robot that they want to walk around and sense and explore the world. And these robots, to teach them how to interact with the real world and not eat their leg you have to physically teach them common sense. And the problem is that when you're making a video game we only fix what we see. So we see the character try to eat his leg, we put a thing in don't eat your leg, and we control that. That's common sense. Common sense has millions of parameters, just basic common sense. Don't fall down the stairs. Don't try to walk up the banister. It would have got you there but it's a pretty dumb way to go. And so effectively just by listing all that stuff out it's going to be a long time before we get characters that are very believably dealing with the world, that you can walk up and say let's build a house. I saw a video from an MIT robotics guy that was saying, just teaching robots to build things from the top down. For example, they'll think I'll start with the roof but they don't realize that now that roof won't support itself. And so they have to teach it just every single little thing. There's so many parameters it's going to be a long time before you really believe that this thing is alive.

MAN: My name is Ian Vogel. I'm from Irrational Games here in Boston. When we did our last game, System Shock Two, we had severe 3D restrictions, portals and polys. So I'd be a big fan of seeing real time tessellation and deformation. I'd like to open this up to both Peter and Mark also, though. Concerning a more stable type of a platform, such as the Playstation Two and maybe down the road Dolphin, and how do you see that impacting PC culture down the road?

MOLYNEUX: It's a big problem. Because one of the great things about level of detail, as you said, David, allows us to say to an artist build what you want. And that's exactly what we said with Black and White. Build what you want, it doesn't matter, the engine will handle it. But actually what it turned out was they couldn't really build whatever they wanted because it's all very well having level of detail but if you start getting what's called popping in the foreground it just looks bad. So there's got to be a marriage between that. More Playstation Two and Dolphin and whatever [UNINTELLIGIBLE] are offering is untold power. But there is a price of that power. And the price of that power is it's going to take a lot to get it right. It is like starting with a blank sheet with these new [UNINTELLIGIBLE] machines. Totally new processor. We've all been weaned on PCs, we've all been closeted by open GL and by direct takes. And that's going to go away if you're developing on the PC. So it's going to take a while for the Playstation Two and the Dolphin really to show off its performance. Because as you've properly found, the development community, it takes us two to three years just to get used a machine really. I mean Playstation Two is a fantastic machine, don't get me wrong. But it's not quite as fantastic as perhaps we'd all hope it to be. And what that means is we don't need to have any of these tricks there. We could just throw an infinite number of polygons at it. I mean and that eventually is the best thing.

MAN: The problem we have actually is also for a good business model for a company it's best if you can plan to know that you will be converting that game across onto other formats. And so the way we do it, the standard way people do it is they make their game, time has moved on, so time has moved on two years, this hot new console is out and they're trying to convert up to it. Which means for a lot of PC games that are around today, they actually have to redraw a lot of the graphics to step up. So what we do, [UNINTELLIGIBLE] is probably about 1,000 polygons. A typical character in our next game is 250,000 polygons. We just went nuts. We said OK, there's no machine on the planet is going to push this many polys so if we draw it at that detail we only draw one model once. It will work on any conversions and anything we do. If somebody says I want a cover for a magazine, pick that model. It's high enough for a magazine cover, a poster, whatever. So effectively we skin all those cats and we plan for the future as well. And so I think business wise that's the smart way to do it.

MOLYNEUX: The only thing is, of course, especially on the PC, the hardware manufacturers are constantly coming up with new tricks and new things and new techniques which you can't plan for. Lighting models. What's the most important thing? It's not poly count in my mind. It's lighting. We don't get lighting right in computer games. And we've since focused more time on it.

MAN: I have one comment to make which is that I think in five years time there won't be PCs used as gaming platforms in the home. I'm pretty sure this is Sony's strategy and if Sony executes the PS2 strategy correctly the PC is simply going to be wiped out of the home, it will not exist. So in that sense it will solve your platform problem. Because there will only be one platform.

MAN: I was intrigued by the comments about families, about computer games existing for one full family cycle. I assume when I become a parent I will want to be able to play games with my kid. But I will think that if I was playing with my kid I still want to be sort of a parent and I'd like to have a little bit more power over this game than the kid does. Which doesn't happen with like current games. Kids can go [UNINTELLIGIBLE]. [LAUGHTER]

MAN: Meaning [UNINTELLIGIBLE].

MAN: Other than that result is it possible to maybe start thinking about games that parents can play with kids and still be a parent?

PERRY: I am shocked there's nobody doing that right now because that is the most obvious, for the companies, they make children's software, I'm shocked that nobody's got some big marketing campaign out there right now that's just pointing a big finger at that new market that's being created. And that market is only going to get huge. That's the only way it's going to go. And so the problem is there's all sorts of directions to always go in this business and you have to choose them. But it's something I think you'll see a lot of.

MOLYNEUX: I think it's a great idea. If anybody wants inspiration for a great idea.

PERRY: Start a company today quickly before somebody else does. [LAUGHTER]

MAN: [UNINTELLIGIBLE]

MAN: So Gabe just said that there's a lot of parent child playing in the online world which is one of the primary places where you see that sort of activity.

MAN: Something just in general. We do find, you're right, absolutely, I get letters from people saying for god's sake would you listen to us, we've written you so many times, please add a feature in your games, and you might want to do this too in [UNINTELLIGIBLE] the father and the son both play Half-Life, please put an easy setting in so that when I play he doesn't screw up all my keys and settings. They're basically sharing the same PC. That actually happens quite a lot. And so it's something that we again tend not to think about and so we don't design for it. We don't make their lives easier with our interfaces. But they would love to have custom set-ups for the different people in the family that share the computer. As you know, it's a day's work.

WOMAN: My name is [UNINTELLIGIBLE] and I have a question for David. I'm curious what exactly is Shiny Entertainment going to do to my cell phone? [LAUGHTER]

PERRY: Basically what we're going to do is personally I don't like where cell phones are right now because they're kind of in a big change. There's going to be a big change until you actually get proper color screens and little video cameras and all the rest in it. But right now it's going through this wireless application protocol which is basically a very simple little web page system that's being built into cell phones. But it actually does allow, and something which I find very interesting, is multiplay gaming on cell phones. So you can be sitting in an airport playing somebody else in another airport somewhere else, a very basic game. And the only reason the game is basic is because we still need to get people interested so then they'll start buying the more expensive gamer phones and then they'll get 3D chips in the phone and all that kind of stuff. But right now the idea is you could play somebody Tetrus, they could be somewhere else in the world just on your cell phone. Or you could just log in using the [UNINTELLIGIBLE] protocol, find somebody that wants to play you that's already on the server. So it's a very simple but very, I think it's going to be very far reaching with all the wireless games in the future. Are you thinking of tying it in with your existing games as well?

PERRY: The games we're doing right now are like super, super high tech. So no, they couldn't run on cell phones for a very long time. But the fact is I would like personally to code up some real simple games and try and help get this started. Because I believe, like my girlfriend doesn't really like video games but whenever she's sitting in the movie theater just waiting for the movie to come on I see her playing Snake on her Nokia. So the fact is that if the game is right there and it's a relief from being bored it's an excellent thing. And the more of that in the future, I think, you'll see more people reaching for it.

MEAD: Hi, Gary Mead from Sloan School of Management. With the success of Japanese imports game like Final Fantasy or Metal Gear Solid which, relatively speaking, stresses much more on kind of the plot and on what Mr. Perry mentioned earlier, the emotional element of the game, do you think US companies or game developers in the US are trying to adapt and trying to change and stress more on the emotional element? Or do you think that users will have to wait for the next Japanese import?

PERRY: It's a good question. Emotions are an expensive thing. Whenever you say the word emotion it means money. Somebody's having to spend money doing research. It's not something that they can just use the Quake engine and do a copy of Quake. The fact is, very, very simply as a programmer, the programmer can see you through his, or [UNINTELLIGIBLE] your machine. He's actually watching you and seeing what you're doing. He knows that you're frustrated because you're pressing the keys rapidly. If you're trying to open a door and damn, it's locked. We can start to sense that stuff. We know if there's something really important and you just keep walking around the room because you seem lost. We can start to sense that you may be lost and we'll keep an eye on you a bit longer and then give you some help. And effectively, as we were starting to sense what's going on, in fact, we were talking about actually sensing stress in people's voices. That's going to be definitely another part of voice recognition. It's not just to hear what they say but to understand how they feel about this is very important. And I think the more research that, we're probably, like Peter is doing the stress stuff right now, we're definitely going to be looking a lot more at creating situations where we know how you're feeling right now, we can force situations upon you. There was a great game years ago called commando. I'll give you a better example than this actually. A pinball machine I bought recently. The pinball machine is called getaway. And the machine would suddenly start giving you lots of balls. It actually had a bug in the software. You just go nuts trying to deal with all the balls. And it puts you under so much pressure for about, I think it's two minutes 45 seconds of insane pressure. But you score a huge score, it's very exciting and you're literally sweating. I mean my employees play it and they walk away sweating. But that's a bug. [LAUGHTER] So they actually fixed it and took that out of the game. It's by far the highlight of the game. It's the most exciting thing for you to do. And so if we know that we can get you sweating and get you all excited and play around with you like that, I think there's many different sides to the emotion thing.

MAN: You said before, what does emotion have to do with doing research?

PERRY: Basically the research side of it is, for example, if I'm added speech recognition in the game right now I tend to think damn, I've got very limited resources here, I can't store, I can't understand your language, especially when you start adding accents and all the rest of it. So what I'll do is I'll create a clever way to fool you into thinking that I'm being smart. So I will make you say words like lock on target. And then I'll say fire air to air missile. And I can spot the difference between those two very easily. But if I say go back or attack then those are very different things that I don't want the computer to make a mistake on. Do you see my point? So basically what we're looking for, it's funny because we're actually doing this all the time. We're actually watching for academics to come up with something which we go you know something, they've done the hard work here, we can use that and actually make an improvement in games from it. I think [UNINTELLIGIBLE] is a good example of that.

MOLYNEUX: I think your question was how can you research emotion. And it is a tough thing. I mean for me portraying emotion in a computer game is all about looking at something and asking yourself whether or not --

MAN: That's not quite what I mean. I mean what does emotion have to do with research? I mean we all have emotions. Why can't we just get a talented writer to write an emotionally compelling story and then put it in the game?

MOLYNEUX: Because there's a lot more than just putting it in the game.

MAN: Are you sure? [OVERLAPPING VOICES]

MOLYNEUX: And I'll tell you why. Movies have it so much easier in way, for me. Even in the black and white era. Even when movies just first came out, even silent movies could make you cry. The first movies could make you cry. Computer games, I mean we're saying that you cry and that's a broad thing we're saying. But computer games find it a lot harder. Because firstly it's not a medium where you sit back and look and you absorb everything. It's something you're interactive with. And so it's very, very different. It's more like me coming to your house, sitting down and telling you a story. And if I tell you a good story maybe you'll cry. Maybe you won't.

MAN: Maybe some stories are just that good. And maybe you guys need to invest more time and money in making sure your stories are that good.

PERRY: I think you're right. You hit on this point before. You are actually right. The fact is a lot of video game developers are just that, they're just video game developers. They're programmers or artists. And at some point they say I'm going to write this compelling story. The fact is they're not a writer, they don't have the ability to do that. To get the very, very, very top best story writers, the ones that really, really know how to --

MAN: Maybe they just need to get better. [OVERLAPPING VOICES]

PERRY: It's not something that we can just get better by reading a book or something.

MOLYNEUX: It's not easy, I promise you. I mean you can tell it's not easy because we've been doing for over a decade and [OVERLAPPING VOICES].

MAN: To make us get better why don't we turn to Mark Pesce? Mark Pesce has been exploring the frontiers of the future for nearly two decades. The author of three books, Pesce is widely respected as a technologist who possesses vision in equal measure to his technical prowess, paired with a unique ability to translate abstract concepts into concrete explanations. In September 1998 Pesce was appointed chair of the interactive media program at the University of Southern California's world renowned school of cinema and television. His mandate, to bring cinema and broadcast television into the interactive era, led him to create a program that encourages creative vision and is producing a generation of entertainment professionals shaping the media of the next century.

PESCE: Thank you, Henry. So hopefully that means that one of the things that the cinema school stresses, which is writing ability and the ability to write a compelling story is going to become one of the elements of interactive media in the context of the program. I have a couple of comments, they're basically arranged in three categories. And since I'm not a professional in the game industry I'm trying to present a bit of an overview. I'll toss out some ideas for the future and some observations of things that are going on now. Last night when I got back to my hotel room I flipped on CNN. And there was a segment on headline news which was reviewing the latest PC gaming title from Electronic Arts, Ted Nugent's Wild Hunting Adventure. [LAUGHTER] With expert commentary and exhortations from perhaps the most washed up of the Dixiebilly [SP?] rock and roll stars. You learn to lift your rifle, load it, camouflage yourself and wait for your prey. The goal is to fire upon the endangered species of North America. Grizzly bears, gray wolves, bighorns, boars and so forth. Of course, this is an instructional title. So you get cutaway views of the animals in flight from your bullets, their bodies rendered semitransparently so you can see their vital organs to get a better sense of where the fatal shot must land. [LAUGHTER] After you've taken down your beautiful white tailed deer Ted Nugent pronounces, now you're an American. [LAUGHTER]. We can easily see that play has become a major component in our culture, that games have become continuous, evolutionary, a part of life, although perhaps an alternate life. With our lives increasingly in cyberspace and the near constant communion with the disembodied souls out there on the net. This is not really anything new. The financial markets have lived this way for over 100 years. And although many people may not agree that financial trading is a game John Von Noyman [SP?] and the theory of games and economic behavior made this case clearly a half century ago. Today we get this sense with people playing the markets. Running the CNBC ticker in the background all day long, bringing up their portfolio on My Yahoo, etc, it's already a game. We just need to extend the possible range of forms. One of my interactive television students at USC is hard at work on developing an interface which will allow you to let the stock ticker follow you anywhere. As long as you're in front of a television you'll know where you stand in the game of capitalism. Michael Benedict, in an MIT press title called Cyberspace First Steps pointed out that absence from cyberspace has a cost. This is becoming increasingly apparent. In the continuous online life we're living in the world of financing gaming being offline, even for a few minutes, can have enormous consequences. This is the harbinger of the gaming environments to come. Here's my palm five. It's one of the millions of PDAs that have been sold. I had ordered, and it was supposed to have arrived by now so I could do a good demo, something called an omnisky which is a wireless modem that fits onto the back of this. And when I add it I'll then be connected continuously to the global networks. So it'll become my portal to anywhere in cyberspace. And it will become the platform for a wide range of games which will involve us continuously, be they strategy or role playing with a cast of thousands or simply a tool that will mesmerize us to the fluctuations of our portfolios. It's not that this would be using fancy graphics right now but it would create a world that could pass the touring test. And that's what counts as immersion. It's not how many polys per second. It's whether it's engaging you as a human activity. So gaming is about to extend into everyday life continuously and deeply. Each of us will find the kind of challenge that suits us best. And we will engage a far flung but ever present community of players. We might live our lives in the drab Februarys of the real world but like so many Walter Mittys we will possess a rich life of internal twists and turns unseen since the time of the Medicis. It will become the lifeline to the imagination, the sandbox in my hand. Much has been made about the danger of the game industry which is facing an enormous onslaught from copyright violation which it regards as the outright theft of intellectual property. My own position, which was published yesterday in Feed magazine, is that the net regards copyright as a form of censorship and routes around it. This does not bode well for the $7 billion game industry. This of course, the gaming industry contends that without access to an ever increasing stream of revenues it cannot continue to exist. But in fact, it costs only one tenth as much to make a game as it does to make a film. So they can probably survive a pretty healthy dent in their revenues. Not that I'm suggesting this but I have a feeling that's the wave of the future. Even if the worst happens and the increasingly available supply of illegally acquired software drowns the gaming industry, which is similar to what happened in the early 1980s, gaming will not go away. Far from it. Not very long ago Quake was released as open source. Now people are using it to create other forms of entertainment, racing games, chess and probably a few hunting titles. In the ever more prescient words of William Gibson, the street finds its own use for things, uses its maker never intended. What we've seen, particularly in the case of Linux, is that enough caffeine and enough programmers intent on a task, nearly anything is possible. Gaming will not go away. In fact, as the platforms approach near infinite power it becomes easier and easier for a novice to design a compelling game. I cannot imagine that we'll progress much further into the 21st century without a legion of open source multiplay gaming projects similar to Ultima Online but without the usurious fees. And they'll begin to take shape within, I'd say, the next five years. These projects will probably begin at universities like MIT and USC where kids with the right tools and enough time on their hands will begin to assert their own right to design the worlds they play with. Because these worlds will be designed as open from the very beginning they will present creative opportunities that the commercial titles simply cannot offer and they'll quickly supplant them. This is not to say that gaming will cease to exist as an industry. As in the case with the Linux marketplace, the lead designers will form companies, likely IPO for billions of dollars and draw a revenue stream based in product placement advertising, consulting and event management. The counter to a world without copyright is a world where event and performance are the commanding values, things people will pay to see. Last August the University of Southern California announced a partnership with the US army to create the institute for creative technology. Headed up by Richard Lindheim, a former vice president of Paramount Television and the man formally charged with the management of the Star Trek properties, the unstated goal of the institute is to create the real holodeck. While the entertainment industry sees the holodeck as an essential armament in the war for location based entertainment, which is another sure sign that in the age after copyright event and performance are the commanding values, the army wants to use the holodeck to train soldiers in a realistic, synthetic environment creating the war games necessary in an age of peace keeping. It should be noted that we have only the faintest idea of how to create a holodeck. But the idea is compelling enough that it drew $100 million of tax money to fund the institute. The army believes that its prowess and simulation combined with Hollywood's ability to create compelling stories and characters will be the one two punch that solves this nearly intractable problem. However, there's a very natural dissonance between the stated aims of the two organizations. Hollywood wants to produce the machinery of fantasy while the army seeks the perfect of synesthetic reality. Soldiers could be trained in an environment of a simulated Somalia or Bosnia or Haiti, learn the local landscape and customs as to be better conditioned to the actualities of the ground situation. Of course, this logic applies only if the simulation represents ground truth. In fact, every simulation always reflects the biases of its creators. And soldiers trained in an environment would be reacting not to actualities but to somebody's prejudices about realities which may simply be well rendered fictions. But regardless, the project is moving forward and will result in some very expensive high tech toys for the creation of synthetic environments. In the search for verisimilitude, that is emotional faithfulness, the designers will begin to add consequence, the missing element in so many of these more violent video game titles, to the environments. As the high technologies of simulation invade the skin, as nanotechnology produces rapidly reconfigurable sensory environments we should be able to directly create the sensation of being hit by gunfire and probably even the lingering sensation of death, at least until the reset button is pressed. In all of this the army will have forgotten an important lesson that Hollywood learned during the second world war. It's a Wonderful Life director Frank Capra produced a series of movies called Why We Fight which, because they so faithfully depicted the horrors of war and of the enemy menace, actually drove enlistments down. It seems that in war, as in a good game, too much truth ruins the play. But half truths presented to the soldiers as reality would not be training. It would be brainwashing pure and simple. You can imagine the generals saluting the soldiers as they leave the holodeck with a hearty, now you're an American. [APPLAUSE]

MAN: So are there any Americans out there ready to fire some questions? Looks like somebody's coming.

MAN: I'm curious, since this is a discussion of future games, what role you see in game advertising having in the near and long term future. And there have already been games that have had clothing tie-ins. Do you see that trend continuing? Is it reversible?

PESCE: Is what reversible?

MAN: Having ads in games? Will we ever be able to go back?

PESCE: I don't think it will ever be reversible because there's actually companies forming that are agents for putting advertising into video games. So that they're going to make it their full time, all day long, goal to get advertising in the games. I was meeting with the folks at Idoss [SP?] last year and they were excited about the fact that Laura Croft could put her label on a brand of combat boots. And they really saw this as something that was very exciting to them.

MOLYNEUX: We've got a health care provider in the United Kingdom called Bupar [SP?] who have approached us and wanted to set up an online site where people could take their sick creatures to be made well again. [LAUGHTER]

MAN: And you have to wonder because, I don't know if it's been mentioned already at the conference but somebody auctioned off their Ultima character on Ebay a few months ago for about $5,000. So there is, in fact, a real creation of value outside of the cost of a cartridge. Apparently the collective value of the community, the time invested in the creation of a character, actually presents real world wealth.

MOLYNEUX: It's inevitable, I think, and you're going to see more of it. I just hope it's done tastefully. [LAUGHTER]

BECKLAND: I'm Christian Beckland, a student here at MIT. I have one question that has two parts. You just talked about copyright issues. And something I'm interested in is, and this applies to the entire panel, entire conference, is most of the games people have been talking about at this conference seem to be increasingly larger budgets. And relating to movies almost the A list of movies, the A movies. And I'm wondering will be [UNINTELLIGIBLE] space. And I think you just addressed this somewhat with say for example like open source games or games which are created in an open community for the low budget. And again making a movie reference, say like the clerks [UNINTELLIGIBLE] of games. Similarly, will there be a room for, seems like again, as I was saying, most of these games seem to be the A list of games. Will there be a room for the games that aren't meant for everybody? Not the action drama comedy that is True Lies but again, the art house cinema of games. Will there still exist, because of this increase in money demands and technological demands, say the indie game. The game that is meant for, by gamers for gamers.

PESCE: I'm pretty sure that because the next generation platforms, and even the generation after that, the Playstation Three is probably, what, 2005, 2006 and it's supposed to be 1,000 times faster than the Playstation Two, at least that's what they're projecting. That at that point you won't need to be a game designer per se in order to get the kind of performance to be able to get something that's effective. So in fact, it could be somebody like a cinema director who's just making his small 15 minute film who could turn out something like an art house project with a relatively small team and a relatively small budget. Because they aren't trying to crack new ground in terms of the technology, they're trying to crack new ground in terms of story. And I mean that's the essential part of what Blair Witch was. It was a story. It was a story told well as opposed to a $70 million epic.

PERRY: The problem actually, there's a lot of people who want to do that. But the problem is you'll get this really great artist and this really great programmer who's very lost in the world. He knows he wants to be a video game programmer. He's willing to develop a lot of time to it. Maybe he's got some great creative ideas. But they just can't find each other. And that's always been a problem. They're not looking to work for a video game company. They've got a job. They just want to do some particular thing. And these no web sites where they actually meet and discuss and finally you are the musician that we want and they basically join together. One of the sites is developerscorner.com. And they, I think that is going to create the most interesting and bizarre games. There was another solution which was Sony created a thing called the [UNINTELLIGIBLE] which was, it was a black Playstation that you could buy. It was a little bit more expensive than a normal one but they give you all of the tools you needed to make games on it. And they had a web site with plenty of people there that want to help you. I think that's one of the best things about the web. We generally don't want to have to deal with all of the general public. We just can't. There's just too many people. I can't have 300,000 people e-mail us in a day asking us questions. But what happens is if you create a community on the web no matter how technical it gets, like they're stuck trying to make this game, there's always somebody in there who has the answer as long as you can get the right people. And that guy loves the fact that he helped. You'll find you get these evangelist guys that live on their helping everybody. And so if you can create that community I think you'll see a lot more of those titles.

MOLYNEUX: I think what your question actually says is will it ever be possible to create games of the same quality that come out day after day and year after year. And yeah, I think that will be the case. I mean you can go online now and you can download shareware 3D engines and shareware 2D engines today. And there are loads of graphics, free shareware graphics online. One of the biggest barriers to making a game is it takes a heck of a long time to do so. And if you are there with a camera you can go out and shoot a film in a month and edit it in a month. Can't do that with a computer game.

PESCE: In point of fact, though, any reasonably good [UNINTELLIGIBLE] film actually takes about 18 months from beginning to end. It's not something, even with the DV and all the tools the amount of time that it actually takes to do it right is comparable to the amount of time that it takes to make a video game title.

MAN: The problem is if you're planning to compete on Playstation Two, for example, the development systems are extremely expensive. And so that's always the thing. It's the most common e-mail I get actually is OK, I'd like to make 3D graphics, I know video games are where I'm going to end up, but the tool that you're all using is $3,000, where can I get a free one.

PESCE: The other side of that, and I think Sony is taking care of this, is they're coming out with a Workstation line later on this year that's based on the Playstation chip set. And the idea is that will actually become a low end development platform for the Playstation. So Sony, I think, is going to try to jump start that as well.

PEARS: I'm Celia Pears from USC. I actually teach in Mark's program and I teach writing, guess what, for interactive media. It's not so much a question, it's just a comment that I want to say that's been touched on throughout. But this whole discussion that it's impossible to write a good story in interactive media is something that I just patently reject. I think that if you look at the history of computer gaming that it's very clear what the priorities have been between the CD-ROM market and the console and arcade market. That story has always been a low priority. Speed has been a very high priority in a lot of these genres that we're talking about. And you guys touched on it a little bit but to me as a story teller the biggest thing you need to look at is character and empathy and this idea of caring about characters. Which I think, Mark, you've started to do with your virtual creature. And certainly people who have played Creatures, if they like it, find that they care about their characters. I want to bring this up because empathy is really the heart blood of almost every other fiction form or literary form that we have from the novel to the film to television. And it's almost the antithesis of the convention in the game industry. And a lot of, particularly these shooter games we've talked about, empathy is an impediment in those games. So I'm just putting this forth as both a critical and creative almost evangelization to say that there is room for story and character. It just means that we need to prioritize that as something that we think is important. And I don't know if you guys have seen any of Alex Mayhew's work. He works in England. But he's doing a lot of amazing stuff with character driven games. There are a lot of other people out there trying to do this and I think we shouldn't just throw it as being impossibly difficult. I think to me it's the number one thing we should be working on right now. Because I think we've gotten a lot of these other things down and now it's just a question of higher resolution.

MOLYNEUX: I absolutely agree with you. I think that story and characterization is something that we should focus on. And if people turn around and say we had in the industry this fight about interactive movies, the idea that you could play through a movie and the plot would develop according to what [UNINTELLIGIBLE]. We had this fired a few years ago [UNINTELLIGIBLE] and it never worked out. It didn't work out because really it was a realization that you had to have an infinitely big script to actually have an infinitely branching movie. But in reality the most creative story based games that I've ever seen have always been text adventures. Where you play through a text adventure and in your head there's the images there. And what ruined that story telling process is when we actually started putting visuals in them. So it's a very tough challenge and I hope I'm involved.

PERRY: I think a lot of it as well, I keep touching on the quality of the graphics. Nowadays quality of graphics actually can help achieve that goal. So if you're told your goal is to do some specific task and you will save the peasants, if you do this task and the peasants survive but really don't then that's not so great. But if the peasants come over and very convincingly thank you, really, really thank you, and you feel thanked, it actually happened, then you're going to go oh, that was cool. And you'll like that. But right now they don't thank you and you don't feel at all like, you'll think to yourself I wonder what will happen if I don't help them. [LAUGHTER]

COLER: Chris Coler [SP?]. I'm teaching a course right now at Tufts University on the history of video games. And my question is this. Given the obvious pros of two dimensional graphics, of hand drawn graphics as an art form, do you see in the future as 3D graphics get more and more and more involved, do you see any editing time, possibly a backlash of people wanting to go back to, rather than computer generated art the hand drawn art, to the 2D game mechanics? Because I mean 3D is not necessarily being an improvement over two dimensions in all cases. I mean one example right now being Sparesoft [SP?] with Final Fantasy Nine going back to more iconic, still with their higher graphic standards but going back to more deformed type characters, more 16 bit looking characters. Do you see this as possibly happening down the line?

PESCE: There's an entire movement afoot in computer graphics on what they call non photorealistic rendering. And there's a lot of technique and academic research going on in this. I have a student who's developing his thesis project right now. And the entire world, although it's in 3D, is developed with a very specific visual feel which is not based in reality. It's based in a particular artistic perception of the world which is courtesy of the point of view of the character. So in other words, it's a particular drawing style which is used to advance story. So I think that's one of the areas it's going in.

PERRY: I think it's a harsh horrible business part of this industry is that gamers tend to be quite fickle. They want the coolest newest looking thing. And there's definitely a good drive towards retro stuff right now but the fact is, I think probably the best example of that was when Donkey Kong Country came out on the ten to 64. Because Rare [SP?] had prerendered the characters you were now seeing smoother, more rounded characters than you had really seen before. And people flocked to it. And of course, it still has to be a good game. But if they can clearly see the difference. I did a game one time called Aladdin. Aladdin actually, to be honest, I was really surprised to see how many children spotted the difference between the animation quality. Because we didn't draw it, we had Disney draw it. And Disney did a fantastic job. And kids could see, I mean they would say wow, it's so smooth and it feels so real and they noticed. And I just find that just constantly nice is people are always looking for the next cool looking thing. So I think financially for our business it would be a pretty scary place to be, to take a big step back.

MAN: What if you had your own way? Would you go back and do stuff like Aladdin [OVERLAPPING VOICES].

PERRY: Actually you know something? I personally think 2D is 3D without moving the camera. So you could play Black and White and just never move your camera. It wouldn't be much fun but you effectively can create a 2D image in a 3D world. I tried that with one game and it bit us really bad. Because it wasn't state of the art. It was an old style game using two and a half D, which I thought would have been fine, but the market just said no. So now we go super hard core for like the best technology there is. It's just the business. That's part of staying in the business.

MEAD: Gary Mead from Sloan School of Management. I used to be a former employee of [UNINTELLIGIBLE] Industries. So I'd like to think what the panel thinks about online gaming and where that space is going compared to single player games.

PERRY: For me the big thing is, I mean we as a company, it's so funny, every time we ship a game people say so is it going to be multiplayer. And I go maybe but we never do. And so the fact is we have now got to the point where yes, we're absolutely going multiplayer wherever we possibly can. It is definitely the most fun when you can, it adds so much life to a game after you've had the initial experience when you're able to go play your friends. And especially I heard of a great example of this. If you're sitting in some scummy hotel room that you can connect into this world and still be there with your friends even though you're out in Wyoming in this horrible hotel. You can still get into this world and have a great time with everybody else. And ideally I think still those games, the guys in my office play some of the massively multiplayer games and while I appreciate them I think they're so simple. And still there's no --

[END OF SIDE A]

[BEGINNING OF SIDE B]

PERRY: -- compelling world to get into. But I'll tell you, we will be there very soon. And it's something that will just continue to get wider as time goes on. I would love to see like the CETI program where you see all the shared computers basically working together on the same problem. I'd love to see the actual game getting shared across all the computers. That would be a really nice concept. The more people you add the more processor power you add to the emotion and having a scalable world in that way. That would be great.

MOLYNEUX: They're embarrassing me. I've been saying that multiplayer games, online games, will be the next big thing since 1989 and it's taken like ten years for the world to catch up. For me playing against a human player, playing online in any way is just the coolest thing. Especially if you can bring something that's unique to the online world I think there's going to be innovations that we can't even dream about yet on the online side. And it really does bring people together. The Quake clans. There are people from different cultures and different ages all residing together. And that's an amazing thing. So I think the future is very bright and certainly every game I'm going to be involved in is going to have online.

MAN: I'm afraid this game is going to have to continue online. We've used up the time. Please look for the transcripts on the web site. And while you're waiting send e-mail to a friend, talk about something you've learned here. Just talk about what you've seen at MIT's comparative media studies program. And come back for some of our future events. [APPLAUSE]