And how did GoldenEye come about?

MH: As I understand it GoldenEye was proposed to Rare at the highest level - Chris and Tim Stamper and Joel Hochberg. They weren't that excited but were talking about it. I heard about this, and I said, 'I'd like to do that GoldenEye game', and that was that.

Were you a big Bond fan?

MH: I still am a big fan of the Bond films. Back then though, there hadn't been a Bond film for years and years, and the last ones hadn't been terribly inspiring. They'd lost their Bondness. That was one of the reasons no one at Rare was bothered, but I loved Bond, so it was very exciting and it was an open brief. I could choose what kind of game to make. I wasn't experienced in the commercial world of video games, and I wasn't aware of the stigma of a film license, so I came at it with an open mind. The first decision was 'Let's make a shooting game', then I was asked to do the design doc and build the team.

The development process was prolonged wasn't it?

MH: I started out with Mark Edmonds and Karl Hilton, and it was the three of us for about year. We started out small and grew very slowly. By the end of the third year, we were about a dozen, and we were pretty much left alone. Rare was never a talkative company. Tim and Chris weren't over all the time seeing what we were doing. The essential process was throwing some talented people together and letting them get on with it. We didn't have project managers, there was a smidgen of a schedule, but it wasn't as if we were told to deliver milestones. Of course, GoldenEye was awfully, tragically, disastrously late.

Did you ever think the project would be canned?

MH: I was aware of the pressure, which was intense. It was clear to everyone Nintendo wasn't interested in something so-so. We responded with fabulous work. We were lucky to get hold of exceptionally talented people and we had resources most other developers couldn't dream of. But looking back, I imagine there were lots of very serious, ongoing discussions about cancelling the project.

For example, most of the team were new hires without any experience of game development. The original idea was there would be a glittering launch simultaneously with Project Reality [what became called the N64], which would be synchronised with the release of the GoldenEye movie. We were actually closer to the release of the Bond movie after. [GoldenEye, the game, was released in August 1997, nine months after the US N64 launch and 22 months after the GoldenEye film.]

Equally, did you think it would be the success that it was?

MH: Anyone in the industry with a bit of wisdom will tell you, it's always very unclear whether a game's going to be a success or not. No one ever came to me and said, 'You know that game you made, that's going to sell eight million units'. No one could have predicted it. Nintendo couldn't have predicted it, and it has people whose job it is to predict sales, because the sales were very flat. It's uncommon for a game to sell about the same every month for years and years.

Why didn't you do another Bond game?

MH: We were offered the sequel. The rest of the team were keen, and in one respect, out of all of them, I was the one most likely to say, 'Yes' because I loved Bond. But I was able to say, 'No' in a second. A lot of the high level decisions on Perfect Dark were made to try and be different to GoldenEye but still reuse some expertise and engine. Really though, I needed to work on a game more different than Perfect Dark for it to be interesting.