CN: But it's interesting to talk about process, right? Like last time we spoke, we talked about your script writing and your process with the actors and scriptwriters and that. I think that kind of stuff, for our audience, is exactly what's useful. On the other hand, inspiration is something that you can't talk about concretely. It comes to you.

PM: Well, there are two ways inspiration comes... One is, I look outside this window now, and I can almost see two or three game ideas that you could latch onto, and I think that's part of the process of being a designer.

That's the sort of the more mechanical side of inspiration. And a lot of times, you do think deeply about how you're going to get people to feel like this, or how you're going to get them to experience that.

If you think of Fable, it's how you get people to feel like they've got choice. And that's one side of inspiration.

The other side of inspiration is that moment where you're never quite sure when it's going to come. It can come at the most surprising, shocking times. I think the worst time for me that I've ever had an inspirational idea, which was an idea for a feature in Black & White, was actually when I was being intimate with my wife.

CN: [laughs]

PM: Having to sort of stop and write it down was... Yeah, I wasn't popular. My mind wasn't on the job. It was at work.

Something that makes me think about is the creative tension in games. I think a lot of it arises from the fact that there's a tension between the demands of software development, which is feature-driven, and creative expression, which is driven by inspiration. It seems like what you said might speak to that, to an extent.

PM: Yeah, here's the key thing. If you really want to get in some features, and sometimes they're very, very difficult inventions, to get a key feature in a game...

You know, if we take the dog, it was an invention, just like inventing anything. It's all very well for me as a designer to come in and say, "Hey, let's have a dog in the game." That's one sentence, and there you go. But actually, what's required to invent that is a lot more process-driven.

The dog started in a very, very different direction than where it ended up. It started with your ability to be able to praise and reward the dog, stroke him, hit him, and his behavior being driven by that.

We got to that prototype, and we kind of see this prototype and kind of look at it and think, "You know what? This doesn't make the dog feel real at all. It makes him feel completely artificial."

CN: Like a Tamagotchi.

PM: Like a Tamagotchi. It kind of trivialized him. The player would have to worry about... They weren't worrying about being a hero. They're worrying about, "Has my dog seen this? should I be punishing the dog about that? And should I be rewarding him about this?"

That wasn't the direction of the game. The direction of the game was that this dog was a dog, and he kind of learned the things that you did, and you didn't have a button to press to change his mind or behavior.

And so, how that changed was a real process. It was this experiment that started where you had a dog, you could stroke him and you could hit him, and what was fascinating about that is the one we had...

A lot of times with prototypes, the gameplay testers play the prototypes, and 90 percent of the gameplay testers, when they actually touched this dog demo, immediately went to the dog's bottom and started fondling the dog's bottom, which is obviously completely... That's not what we wanted.

[all laugh]

PM: We didn't want people to feel like they could do that. But that really showed us, "Hey, this doesn't feel like stroking the dog. This doesn't feel like owning a real dog." The next process was to say, "Okay, this doesn't work. This isn't the core mechanic. The core mechanic of Fable is actually running around the landscape and exploring. How can we help the dog with that?"

We had this one moment, there was an experiment, in which we used the Fable 1 engine, and again I'm going to show you this [in my GDC presentation]. In almost all games, when you're traveling with something, the thing that you're traveling with tends to stick behind you. Companions always travel behind your or alongside you.

Well, what we did was we just very simply said, "Well, let's put the dog in front of you." And suddenly, you could see it -- in this demo. And suddenly, the emotion of feeling like you were with something that had its own mind suddenly was there. It wasn't because of any brilliant AI, it was just simply moving the coordinates of the dog so it was in front of you.

That is all part of that process, but it's done in a way so that you're not limiting the amount of time that you're going to executing that idea, because this is where it gets really scary. If you've got an invention, an idea, or a prototype, and you're eating into your development time, you're actually limiting the amount of time that you can polish that feature, so we did a lot of prototyping here, which enabled us to polish the dog... Does that make sense?