PART FOUR

Q: The Gamecube appears to be designed to make it easy to develop for as possible, but I was wondering what steps were taken in that area. Also, as far as launch titles go, would five titles be about right? Or can third parties be included? Would the launch be delayed if the games aren't there yet?
S: That's a lot of questions for one time (laughs), but to begin...
M: First, about software, we'll go official on that next May. I can't make any promises right now. As for why next May, I can't tell you why yet. Right now, though, we are pushing for five titles on our own. Beyond that, if I say too much then Xbox might start changing their strategies and I'd be in trouble (laughs).
Q: (laughs)
M: I really don't want to say anything. Very generally speaking, there's a lot to do. To make a game, you have to move polygons around. For 3D games, you know. And to display polygons, you have to do a lot to get them going. The mathematical calculations for polygon displays, light sources, shape changes, textures, and so on. Recent hardware has been all about how many polygons, but in general if you do textures that number's cut in half. And if you do lightsourcing, then it's cut in half again, and whatever you do, it's half, half, half. And then when you notice what you're doing you're going at about a tenth of the speed you were. And then in the hardware there're bottlenecks that cause everything to stop if you hit them, which will cut everything to a tenth. So, we can say it does a hundred million polygons, but in reality it might only show five million. There are lots of bottlenecks in drawing graphics, and we're using a graphic chip for that now, but the CPU is still calculating everything else. Even if it's a super high speed CPU, it's still doing a lot of graphics work. Also, we had this problem with the N64 too, but the sound is done via the CPU too. It's a bad habit becuase we say "The CPU's incredibly fast, putting sound on it's no problem", and then we can create some nice sound. It's the same thought process as we had with the Famicom. The Super Famicom was too much trouble, so that has a special sound chip. The N64 is a million times faster, so we had the CPU take care of it. Once we actually put it in, though, it got very, very tough. (laughs) It becomes a competition in the CPU between sound, graphics, and game-related things. Just like battling for memory, like battling for processing for speed, like the artist and musician and game designer battling each other. Everyone has to share. That sort of thing. You can't just make a game by moving polygons around. Oh, and there's also terrain mapping, for collision detection. When things move, you have to make sure they're still touching the floor. The CPU does all of that too, not the polygons. And if the CPU has too much work to do, then everything else has to wait. So, basically, there's a ton to do besides merely moving polygons, so we have to divide the work evenly. If you look at the Gekko, or how many blocks or bits there are, you might think there's not much going on in there. The cutting edge of technology now is nowhere near the cutting edge next year. It's last year's cutting edge. 0.18 is this year's cutting edge, but if you talk about the future, it won't be the cutting edge for long. However, if you see how everything is balanced, then maybe the Gamecube will be on the edge for a few years. And so... what was I talking about? (laughs) This is tough, quit talking to me in numbers. (laughs) Some of it's a lie, though. With the Mario demo yesterday, maybe it wasn't really 128 of them moving around. Marios and Koopas were going around on that pizza-like table; they were fighting in Mario 64, but. And there was fire there too, but that was it. However, yesterday there were maybe three Marios, and the rest was just pictures. Like, maybe we can get ten going at once, so don't worry. That's the best modern hardware can do, and all. Tens of millions of workstations are built into that system, you know, and I doubt that's going too far.
S: But you didn't really do that, did you?
M: No. (laughs) So don't write that. Really, don't write that. (laughs) I wouldn't do anything like that, so calm down, calm down (laughs). The point is, we don't worry too much about that sort of thing while making games, and the game comes first. I don't want to say things like it's not an entertainment system, but as a concept, you have to realize that it was planned to be a machine for games. As far as selling points and prices go, even if all I say is about the game aspect, if you look at what's written down here, you'll probably think it's a multimedia machine. It's powerful enough to do anything, after all. It can do it, to be sure, but the idea is, "Let's make some really cool games!" - that's the hardware's concept.

Q: And as for the development kits...
M: Oh, right, right. I've dealt with them a fair amount, but I'll let PR handle this one.
S: They've been publically released.
Q: To the third parties?
S: Well, not just second parties. Of course, people close to our company have begun their development already, the third parties. I'm not entirely sure who's going to be a third party and who's not, but in Miyamoto's announcement yesterday, a lot of what he was saying was toward licensees, in the content. There was a lot geared toward game creators. Many only come in once they see the stance and ideas of the company. As for any concrete dates, no one has released anything on that yet.
M: I don't know. (laughs). Straight from PR.
S: Our current stance is first to make games that stand upon themselves, to show what you can make with the system, to show that Nintendo means business.
M: Are you doing anything?
S: Hmm?
M: We're going to be holding announcements and meetings for the third parties, right? Is any of that planned yet?
S: Yeah. We'll be dealing with that around September.
M: Yesterday was the starting point, so. Once we get into next year I think we'll be getting to a fair amount of people. Obviously, in the end, we can't do it all by ourselves. The designers, all of them, want the development kits. If a team has 50 people then they want 50 kits, but with 50 kits 10 different companies can do development work. And once they come out in volume, development becomes faster. On the other hand, if you release kits too fast then they're usually imperfect and hard to use. That just makes third parties mad at us when they buy them. So, I think that development kits will be perfect once next year comes along, but it'll get gradually better and better as this year goes on. People might think we're doing horrible if you write it like that, so make sure you confirm that with PR.

Back to part 3
To part 5