It's a picturesque racing game, where you can play with up to eight human players at once, or against computer opponents. You'll be able to master individual cars, and get to learn each of the tracks. The main idea was to make it as fast as possible.

We tried to analyse exactly what it is that makes a great game, and then to apply this to our game. We've studied just about every game out there. We've looked at everything from production values through to the strengths of particular games, such as the behaviour in Mario Kart. We've studied different ways to control the car, the effect of gravity and the importance of track design - Ridge Racer was influential in this respect.

We learnt as much from the mistakes of others as from their successes. One example is the importance of ensuring Pod didn't suffer from any clipping. People hate clipping. Another example of this is the way that we concentrated on making Pod just one game. Too many games make the mistake of trying to include too much in one game, never satisfying either type of customer. It's simply too difficult and distracting to try to make a fighting and driving game.

Overall, I feel that this research has resulted in a game unlike any of its competitors. To win, you must know both the car and the track. If you beat someone, you'll have to be a smarter driver as well as a better driver. Because each track offers a different challenge, you will have to tinker with your car accordingly.

Yes. When a machine comes out on the market, you don't really know what it's capable of. I'm sure you'll be able to do much better things with MMX in the future. Each month we tweak and learn more about the hardware, what needs to be done with the software and how it all works. I think that in two years you'll see something really amazing with MMX. The ability to use 3D specialised sound is a real plus point.

At first people were amazed by the technology, but you cannot rely on the technology. There must be something more. The game must be more than just a cool demonstration of what the MMX can do. It's easy enough to have a hit if you have the technology. However, if you want a mega-hit like Doom, Warcraft 2 or Duke Nukem 3D, you'll need to be very good at both the technology and the gameplay.

We were lucky that we had the help of the people at Intel. We coded the 3D engine, and let them explain how to exploit the chip. The MMX chip understands what you want to do.

No, the two programmes are separate. Inevitably, the Pentium version will not be as rich as the MMX one. The background won't be as stunning, and the frame rate will be lower. However, by adjusting the screen size you can counteract this last problem. Above all else, we didn't want to sacrifice the gameplay. The main thing for a racing game is for it to be very fast. Of course, it needs to be beautiful too.

Duke Nukem and Warcraft are my favourites. What makes them so special for me is that they're multiplayer games. They put high quality in every part of the game. I think that they have combined technical excellence with a desire to meet the needs of the consumer. This is vital. They have placed the consumer first.

I first got into Duke as a single player game. Since then I've played it for hours with friends over a network. I think it's important for new games to give you both. That's what made Doom the game of the century. It was fantastic to play against others, especially your friends.

Of course, I must mention Mario Kart. Few games can touch it in terms of playability. It had everything: originality, humour, gameplay, multiplayer capability... A real classic!

We've spent a great deal of time and energy getting this aspect up and running. I'm really excited by what we have. I want this game to grow, and I see this as one of its strengths. We're creating a multiplayer community. Let's say you think you're the best on any one track. Well, you'll be able to download other people's ghost times from the Internet, and compete against them. We're also going to make extra tracks and cars so that there will always be something new to download.

I'm so impressed by the way iD has given players so much licence to change Quake to their liking. I want this to be a feature of Pod. I want players to design their own cars. I want them to create their own servers so that they can reduce the ping time in multiplayer games. I want players to give me feedback on the game. I want to harness the ideas of the playing community so that we can make Pod a really great job.

It has been a great challenge working in this area. So much of it is unexplored. We as games designers still have to overcome a great many problems. One of these is the question of latency. No matter what the computer's technology is, no matter what the game is, the phone lines are still a major obstacle. We're trying to play through the Internet directly, but we know the latency is still so bad that it's difficult. How much latency is a player prepared to accept? As a player, I am very happy to play against a machine, but no machine can challenge me the way another human being can.

That's why we have really pushed hard to include the Internet in our plans for Pod. Our Web site is going to allow us to download new tracks and new cars. It will also exhibit course records, so this way you'll be able to tell everybody in the work that on a particular track, you are the best. That's exciting.

Check out our features archive