Do you think that's really contingent on having not just a small team, but a team with a really good, close relationship?

AE: It's certainly hard to do on a large team. I think that's why you get so many breakaway studios. I think there's another point to make here, which is that it really is a lot of work. People might have seen our presentations and Mark might come across as quite chaotic, but he's actually the most methodical artist I've ever worked with. He gave himself a task week of renaming all of the art files. It's almost obsessive-compulsive.

MH: That just comes from the experience of working with lots of assets though -- knowing how to save time later on.

KE: It's almost a cliché to project that kind of crazy, chaotic image, and it always looks really cool -- but that's not what gets games made. The real genius is a very fine balance between keeping rules and breaking rules, otherwise you end up reinventing the wheel every day. It's important to make your history work for you. We try to use a lot of known techniques and build on them. Ironically, having those structures allows us to be more free in the areas where we want to be more experimental.

MH: When we finished Rag Doll, Lionhead was coming to the end of a natural stage after Fable, and we talked about the opportunities to build on what we'd done. Alex was really excited about branching out, and it was me who was very tempted to stay at Lionhead. I had a job, I got paid well, I knew what I was doing. So the choice was, do I stay in this great job and get paid well doing something I love, or do I start something up with this lot and get loads of stress?

KE: Clearly, you should do the second option.

MH: We talked through lots of ideas together, and finally Alex managed to convince me.

AE: It was very rapid. It was bizarre. We left Lionhead just before the Christmas party in December 2005. Then we were able to get a meeting with a publisher, Sony, which we hadn't really courted.

MH: We had the opportunity to meet Phil Harrison [head of development for Sony Worldwide Studios] -- fantastic.

AE: We thought this was a one-off opportunity we couldn't pass up. We had a few ideas, so we put together a pretty vague pitch, if we're honest. We pitched it in December 2005, and we had our funding in place by the end of January 2006. We incorporated in February. It was just insanely fast. We spent January painting the offices. Everything happened literally in the space of a month. Suddenly you're buying desks and PCs and starting a studio.

People ask us whether we went to lots of different publishers, and the honest answer is "no". We went to Sony, showed them this vague thing, and they just said "Awesome. Go." And even better than that, all the things we really wanted to do that we thought we'd tone down a little for the pitch because we thought they might be a little too weird, these were things he really picked up on, asking why we didn't do more of it. Traditionally, you don't pitch the weird stuff; you pitch the core idea as simply as you can -- so it was great that he picked up on that.

KE: I think one of the most crucial points to stress here is the chemistry of this team, the dynamics of the business. This studio couldn't exist without this huge investment in the relationship between the people in it. This is one of the big problems with the industry today -- you can only get that through time.

MH: It's analogous to a band really.

Media Molecule's current much-awaited project, LittleBigPlanet for PlayStation 3. 

AE: It can come quickly though -- our producer came over from Criterion, where she was working on Burnout, so she was used to working on a game a year, great games. We asked her to come over and help us on the structure side and she came along and instantly worked really well with all of us. We were used to working on these huge five-year projects where you basically kick back for the first three years and let it all sink in and experiment -- which is great if you have the time -- and she was used to this structured sequel process, so we are really getting the best of both worlds.

MH: There's definitely still some chaos. We've managed to preserve that.

KE: This is a great way of working, though. At Lionhead I was part of a central department that contributed to a lot of titles. You find that everything is separate for four years, and then in the last few months you start sticking things together. What we're trying to do isn't the opposite of that -- you can't think of everything up front; games are too complicated for that. There are things that you can learn only through the process of doing them. For example, in areas of designing a character, there are known schools of thought on how to do that. But when it comes to designing an experience, you have to try things out and experiment. Where we can use knowledge that exists, we do.