We'll get back to that in a bit. In terms of the free-to-play mechanics that are coming into it - because it came as part of a wave of Microsoft games that introduced mechanics more typically found in free-to-play games in full-price games, what's your take on that and how do you justify their inclusion?

Dan Greenawalt: So that's how you felt about Forza 4?

I did feel that way about Forza 4, so I'll admit an inconsistency there, but it wasn't pronounced as much. It's certainly much more of an issue in Forza 5.

Dan Greenawalt: I understand that if it looks like a duck and it quacks like a duck... I know the statement. But honestly if you look at free-to-play games they usually have things called paywalls, where you're slowly wearing something down and the only way to get around it is to pay. That's not what we implemented in Forza 4 and that wasn't our goal in Forza 5 either. We don't have paywalls. We have acceleration, and that was based on feedback from players in Forza 4 - there's a small group of players that can't be bothered to do things and they have disposable income. They're the sim guys in a lot of cases. They don't want to do the career, and they don't value those aspects, and that's alright by me. With Forza 4 we had car tokens that range from one dollar to three dollars - the most expensive car was ten million credits in game, and it only cost three car tokens which would have been three dollars.

That felt like it was not making the car exclusive enough for those who are willing to pay. So we made car tokens equal to credits - it's not about making more money, it was actually about saving people's time when doing the grind.

I remember you saying last time we spoke that this felt like the first Forza that wasn't compromised - but do you think you could have done with another 12 months of development, and didn't have the pressure of having to launch alongside Xbox One?

Dan Greenawalt: Not really. We've developed a team that's made to have process around concept, and prototype and production and close-down in a very set cadence. It's how we hire and how we staff, and it's kind of how we are. In statistics there's the idea of the inverse U - the more time you have something can get better, and the more time it takes eventually it gets worse. In game development it's similar that way. You can't make a triple-A game in a month - I'm being hyperbolic here - and as you take more and more months, you get more time to get the quality required, the innovation required and you have to be able to throw things away to make a triple-A game.

But - and I'll be hyperbolic again - after, say, six years, your technology starts getting old. It starts getting outdated, and you have to rewrite it, so you're in the state of constant rewrites. Looking at most games being developed, the two to three year basis is the sweet-spot for triple-A games. But we've optimised our team to do the two year product cycle. It's hugely disruptive to add another year - you're having to throw a lot of work away. Since we've made a team that's able to make games in two years, I think another year we'd have to change our processes to make the most of that.