PomPom Games' Michael Michael

Microsoft definitely helped us with those first two games - Reloaded and Empire - so that was when we had a little bit of funding from somewhere. There's no way that Miles and I could have done something like Empire under our own steam, with our own finances. That's why Empire was full of these little design touches. I'm really proud of how that game went, actually.

It's true that when too many of a certain kind of thing comes out, people tend to be turned off. I see user reviews of games, and it's: "Oh, just another twin-stick shooter". With the combos, the truth is I just like that kind of stuff. Deep scoring mechanics, things you can do to affect your score.

It's always good to put in something that gives the really good players a reason to play the early, easy levels of a game, and still have something to go for. You just put this new layer in that encourages people to go back.

With something like a twin-stick shooter, you have to find these things: things that have the scope to change the kind of score you get in a level, because otherwise it's just plus or minus 10 per cent of the same score each time they play.

So really, I guess it's just me being jealous of the things other developers do, and wanting to try something like that myself. I'm not sure how successful it was putting that in - maybe we should have chewed it over a bit more - but that was our attempt anyway.

Not very well at all. God no. In terms of numbers it's hard to know exactly, but it certainly wasn't very high. Probably a quarter of what Reloaded sold. But you have to remember, something like Geometry Wars - which, for the record, I totally love - just did gigantic numbers, so even though Reloaded was profitable for us, it was still a long way behind something like that. That game hit a chord with the public and it was just huge.

Empire and Bliss Island, for the PSP, were both bigger games, and they proved to us that we can't really spend a whole year making a game any more. You just have to make smaller games more regularly and profit off the back of that, because it's not going to work otherwise.

Astro Tripper was a case in point - we knew the game was there, we knew it was already fun, and we knew that a few thousand people had already played it, and it was just a great time to go through and touch it up. Poppi didn't sell very well, but it was the same idea. It was an experiment, really, and we're now going to present it as a proof of concept to a company.