That's interesting. It's an interesting way to split the characters. The thing about games -- which you're well aware of -- is that they're both like a software development exercise and a creative exercise, so it's like you're talking about something that's almost mechanistic, in a sense. It's a mechanic of the story, the way the story functions. How do you do that division?

RF: Yeah, that was the big thing for us: Marcus is the player, and Dom is the voice of the player; that was the way we always looked at it. But yeah, it's tough, especially because it takes so many years -- that's one of the things that I envy about TV shows: Being able to crank something out every week, and being able to see the fruits of your labors much quicker, and adapt much quicker to whether things are working or not.

It's one of the things that I contend why there's not much humor in games -- it's that no joke is funny for two years. It's one of the things that we still struggle with; whether it be an emotional moment, or a scary moment, after two years of shitting scary, it's not emotional, and you have to worry about getting desensitized, and that's what I think is really hard about games. That stuff doesn't hold up for two years; you start to question yourself.

Even game mechanics where you're like, "This is kickass! This is the most fun I've ever had!" and then two years later you're starting to go, "Maybe I need to add more things to it, and more things to it..."

And I think that's what we've done a really good job of recognizing the saturation point, and saying "Maybe we're going too complex because we're so used to it." And when you expose new players to it, what I love about press visits, is that you get this sort of naive person coming into it, and getting exposed to it, and they're like, "Holy crap!" and you're like, "OK, good, we don't need to do any more with that." So it's been good, from that perspective.

You mentioned humor, then you mentioned getting so comfortable with the game that you can't tell anymore, and both made me think about Portal, and its development process. Obviously, Portal's very funny, and the team also went through a very extensive process where they used lots of tissue testers: they would only play the game once, and they could never use that person again, because they needed to see how they reacted; whether they could solve the puzzles, and whether they both could solve the puzzles and could find the game funny. Have you been able to work with stuff like that?

RF: Yeah, we work with Microsoft a lot, and they do a lot at the User Experience Group, with their playtesting and usability testing; so we do a lot of stuff like that, in terms of getting the first hour -- they do the out-of-the-box experience. They watch the first hour and forty-five, and how do people feel as they go along, and we've done playtesting specifically of the tutorials.

And working with Microsoft is great on that, but we even do less formal stuff, and bring in just relatives and friends and things like that, and getting what their experience is like. I think my son was one of the first non-Epic people to chainsaw somebody, just because it was like, "OK, before we go into E3, we want to see, like, can somebody pick this up? Can somebody learn how to do this? Because it's really complicated."

So, you get your testing where you can, but it's been really great having a partner like Microsoft, that has that process figured out, and formalized, and we've been able to leverage it whenever we can.

But they do it out of Seattle, whereas you're in North Carolina.

RF: Yeah, we get really thorough reports, and we get video captures, and it's always surprising; developers, I think, as a whole, over-generalize. I think developers, as a whole, overestimate their audience.

And not to say that the audience isn't capable, but it's just like [developers] expect that because we can do it, they expect they can do it, and we're always shocked when we get usability tests back, and get a videotape back of what people actually struggle with, and how hard.

Some things that hardcore gamers find easy, that the average person has trouble with. Even when we did the Hollywood Forever event for the Gears 1 launch, sitting down with celebrities and we're like, "Do you want to play the game?" and they're like, "Sure!" and then they start off and are face-down in a corner, and you're like, "OK... This stick is your feet, this stick is your head..." and you realize that we're leaving people behind.

That's one of the [reasons] we're doing the different difficulty levels in co-op, and why we're adding more casual difficulties; we're just trying to reach out to these people, that we think we have a compelling story, and a compelling universe, and we don't want to leave people behind if we don't have to.

I was talking to Scott Brown, from NetDevil, and they're making this game called Jumpgate -- which is an MMO, so it's a totally different kind of game -- but what he just told me is that they did testing on the first mission in the game, and he said they anticipated five to 10 iterations before they would get to the point where they were satisfied with the results that they were getting on the testing, and it was 150.

RF: I can believe that. Yeah, absolutely. That's the hardest part, right? It's finding that sweet spot. And I give lots of credit to people who take that stuff seriously, because it's really important.