Rare's Nick Burton
The director of development discusses the prototypes and evolution of Kinect
With just over a month to release, all eyes are on Microsoft's motion control technology Kinect, with the platform holder pushing the system as a brand new hardware experience and insisting it can change the videogame business and extend the Xbox 360 lifecycle for another five years. But until the system has been on the market and in the hands of the public, there are many questions about just what Kinect is truly capable of and how far it can enable different interactive input and experiences.
GamesIndustry.biz sat down recently with Rare's Nick Burton, development director at the studio responsible for launch title Kinect Sports, to discuss the evolution and prototyping behind the hardware and software for Kinect. And to also answer some of the niggling doubts about lag, the rumours of sign-language recognition and whether we'll ever get used to a button-free controller.
Yes. Going back two years to the beginning of October 2008 we'd been hearing rumblings from time-to-time, as you do as part of first-party development team, about new technology. And this one sounded pretty interesting, there's this guy called Kudo (Tsunoda) who was ringing and saying, "you really ought to come and have a look at this stuff." We'd been prototyping a game and we thought the two might work together - we do lots of different prototypes. So we flew out to their office and it was like some Heath Robinson mad scientist thing. We fired up one of the early prototypes, this was on an early sensor with wires hanging out the back, and there's this skeleton on screen. At the time you had to possess the skeleton by walking into it, it changed colour and then we were off. It was pretty amazing but it was very early. It was exciting but we were just going to play with it - that's what we do as first-party, we take a kit and just play with it. We sat in Kudo's office and he let us take back - you can't really call it a dev kit, but it was 'kit' - that we took back to the UK. Were just spent some time proving out what we could do and some of the designers were starting to think that a skeleton is all very well, but what about if it's an in-game character, or an avatar. And quite stupidly at the time I thought it would be easy to do.
I wouldn't say it was tough, more than lots of really interesting bite-sized challenges. It's never been one insurmountable mountain, but lots of head scratching because we couldn't look it up because no-one had done it before. Eventually you get a head of steam and march through the problems, 'we've figured that one out, gimmie the next one...' And a couple of years later we're showing it at E3 and we're actually still really excited by it.
The best way I could describe it is that it's been productised. It's gone from an incubation project, a research project, to being a proper product and all the things that entails. So a lot of the features have got more robust, more focused. You've got to do that, you can't have it doing 50,000 things okay, it needs to do ten things fantastically. When the 360 was on the cards, we were involved in that, it was the same thing that happened - not the same kind of challenges because it's more of a pure hardware challenge with a console, but Kinect was more of a design challenge. I remember those first few weeks of us having the tech in November 2008, we were going "but you've got to have a button!" To the extent that we built little buttons that you could hold in your hands, just to try it. But we realised we didn't want the buttons.
You fall back to what you know. It's the same in the gaming enthusiast press, they go through the exact same emotional journey that we went through. For us it's quite funny and a little bit frustrating because we're thinking, 'no, you don't'.
The button argument is a brilliant one because that's what we did. They worked 'okay' but there were two many prototypes - and we were doing tonnes of prototypes - insane things, like giving it to a programmer for 48 hours and telling him to do whatever he liked. One guy did a seagull simulator, being able to poo on passers by. But they were the kind of things we wanted to play with. The acid test turned out to be a couple of prototypes we worked out quickly. We weren't thinking about sports, these were just ideas we were chucking around for three or four days at a time. We had a little goal kick game that was to test accuracy of avatars and it worked well because we could place a virtual ball in front of the player. We took it a little bit further and within a couple of days his office became somewhere he couldn't work. We realised we had something there.
And at the same time as that, we realised that kicking doesn't involve a button, there doesn't have to be a button press. So from that we went for a table tennis prototype, with two or three days work. It was wireframe people, table tennis table and a ball, but you could walk back from it and side to side, but it also proved the hand-eye co-ordination worked. And it almost felt like you could feel the ping pong ball hitting the paddle. Perceptually that was really weird.
Totally. That was the shared moment where we realised we don't need buttons. At that point it really evolved for us. It turns out there are lot of ways to interact, it's just a case of finding the right way to do it for your product. I like the hover select we've got in Kinect Sports. It was interesting at that time to go from the assumption that we've got to have buttons to the realisation we don't need them.