It's interesting that camera is still a wild frontier where you can still try new things. I think what you guys have done is definitely a good way to go, but it's also crazy that after all these years of 3D games, we still don't have something to point to and say, "This is how we should do it for sure."

MM: Yeah, and I think it depends so much on the environments of each individual game. I wish you could just have a "this is the best-in-class camera, let's now implement it." But, for example, when we were doing the original Alan Wake, and you have a lot of pine trees around you, one of the big issues is to make sure that the camera doesn't collide with tons of branches. And that's maybe a very different challenge if you have an urban environment, and very clear geometry, and then your challenges are different from that.

And it's kind of neat that now we're to the point where the games choose what stylistically and artistically fits this game. So you basically have a director of photography for your game, in a way.

MM: Yeah, very true. And I think it's cool that we also have -- because the productions are larger -- dedicated people who a large part of their job is to focus on one particular area. And whether it's lighting, or camera, or directing cinematics, or things like that, I think those are very cool things to have, which we obviously didn't have, even a few years back.

Back to animation, how difficult is it to contextualize the dodges? Because enemies are attacking horizontally or vertically, you've got to make sure that Wake is dodging in an intelligent way.

MM: Those are things that you first build a general idea, and a plan of what you want to implement, and you have it planned out on paper. But, ultimately, it just comes down to tweaking, and testing, and just iterating. I don't know anybody who can really just do that on paper and go, "This is the way we're going to do it."

Games, first and foremost, they're interactive entertainment. Going back to the blindingly obvious -- games are interactive. How they feel, and how they flow, and how they react is at least as important as what you're showing, what you're telling people. And it's about the interaction from the gamer to the game, and then responses back.

And if it doesn't feel good, and if the camera and the character don't behave in the way that you intend them to behave, then it's instantly unfun, no matter how good the story, and how good the looks are.

And I think that's the biggest difference between passive media, where you're just watching and taking it all in. Sometimes you need to make compromises, and decide what you are going to accentuate. But I think it's something that should be seriously considered, because it's not a CG movie.

There's always going to be edge cases that are going to break the immersion a little bit -- "Oh, that guy just dodged in the wrong direction" or "I expect to be able to do this, but I can't do it."

MM: It comes down to a level of simulation as well, and kind of how far you want to take that. But as long as it's consistent, I think that's the big thing. If your level of abstraction is very high, then you can get away with much less. But the closer you are getting to real, and fidelity, then people expect to be able to open drawers, and closets, and everything. Which isn't necessarily fun -- you don't open every drawer in every office. [laughs]

There is the other problem. The first time I played Shenmue on Dreamcast it's like, "Okay, I can open drawers. Why?"

MM: Yeah. [laughs]

"Now I feel like I have to open every drawer!" And game players have a real tendency to be like, "If there's something I can do, I have to do it." Not everyone feels like that, but there's the core that will ruin games for themselves that way.

MM: I think a good game that pulled it off very nicely was L.A. Noire. You had a huge world, and you were clearly interacting with smaller objects and doing things there, but at least the visual language, for me, was very clear on what is interact-able.

Going along those lines, you were talking about the environment being more destructible, and dynamic, and things like that. Actually, it looked like you were really just using planned areas of destruction to keep the player on a linear path.

MM: Yeah, there's a lot of that. And then also where enemies come from, and having that destructible -- whether they break a door with an axe and come through that and so forth. But I think it's more about how the player perceives it, and how it feels, than genuinely having destructibility as a gameplay element, as you would in Battlefield, for example, where it makes perfect sense.

So the PC version is still happening?

MM: Yeah.

Why did it take so long between the 360 version and the PC version?

MM: That's a very good question. What I can say is that we did take it as soon as we could to the PC, and the stars weren't aligned when we were shipping for the 360, but we're happy to be bringing it over to the PC early this year. We've had a ton of feedback from people asking for it, and clearly we have our roots in PC development -- with Death Rally, Max Payne, Max Payne 2 on PC. And then all the fan reactions, all the mail, all the Facebook sites, all the community sites. And we do actually listen to the folks. So we're stoked to be able to bring it out now.

On a similar tangent, a lot of the feedback we got for American Nightmare, feedback that we've implemented, was that folks really liked the original Alan Wake for the story and the setting. But there was a lot of feedback on the combat -- they wanted more combat variety, and that's something that we started to address with these test levels and sandboxes with different enemies. And it kind of all grew from there, for this title.

Why are you now able to release Alan Wake on PC with the same timing as American Nightmare?

MM: Microsoft have been kind enough to give us the freedom to take it to PC, and we're really thankful for their support.