When I played Happy Action Theater, it felt a lot more natural than most Kinect games. How much of that is good design, and how much of that is your tech?
TS: Kinect is really cool tech in itself, and then our team came up with a lot of cool tech using some samples that Microsoft had, and some of our own original stuff, so you can, say, pull a character out of the screen. We can take your video image and yank it out of the background, so it looks like you disappear from your own living room, which is pretty cool. And just other tech to make snow appear on objects in your room, and we made pigeons land on your head using some cool tech that we wrote.
But there's also design in that. From the very beginning, I thought about, "What is the Kinect really good at tracking?" and what special features it has. And with those as the building blocks, what kind of experience can you craft out of that?
Well it has a camera, and it has this great depth sensor, which is new. There's been webcams before, there's been EyeToy and camera stuff before where you could make camera-based games, but there's never been one with an accurate depth sensor, right?
Not that I'm aware of, at least not as a commercial product.
TS: So what can you do with that that you couldn't do before? So we have an activity where it takes a picture of you, and then you can interact in 3D with that picture. So you can stand next to yourself, or intersect yourself, or hold your own hand, or wrap arm-and-arm with yourself. Did you get to see that one?
No, I didn't.
TS: It takes a 3D picture of you, so it takes a picture of the video feed, but also with depth stored to it, so you can run around yourself in a circle, and things like that. Stuff you could never do before with just a basic webcam. So it was designed from the ground up to be a Kinect experience, I think that's important to Kinect. If you just see Kinect as a different controller for games -- like, we're going to take some game that already exists and just make it work with Kinect -- I think you're going to have a really hard time.
I think the best games are yet to come with Kinect. Like this, where people just start from scratch with no preconceived notions about what a game can be and just build up the experience from nothing using what Kinect is good at. I think it's going to be good. I think we're going to come up with whole new genres of experiences that just haven't existed before.
How do you playtest a toy like this?
TS: It's fun! In the office we'll set it up in our main conference room. People go up and play it. The fun thing about it is that it's, by design, a really easy thing to jump in and out of. So in the hallway, people passing by will be like, "Whoa, I'm in a snowstorm!" And they'll stop and throw a few snowballs, and then they'll walk on. We can record that footage, we can record that depth information, and use it to test the game later.
We brought our kids, and families, and their friends and families to play it, and that's where we really understood the game. It's the kind of game where you explain it to an adult, and you can experience it as an adult. You can say "it's really cool, it's cool tech, it's a cool simulation," but not until you get three kids in front of it do you understand that it's this crazy imagination machine.
It's like "Oh, we're in hot lava! And this guy is chasing me, and now I'm splashing the lava, and now the lava is here, and we've got to get on the couch!" And you see the kids using their imaginations and creating their own experiences, and you realize what Happy Action Theater is all about.
Double Fine Happy Action Theater
We used to call that "emergent gameplay".
TS: Yeah [laughs]. Oh, and also, Microsoft. This is the great thing about working with Microsoft: having access to their usability research department. They have a lot of smart people, and a lot of great facilities for it. And basically we just watched live tests of our game every Thursday, just seeing kids coming in, and standing in front of the machine, and seeing what they do.
We learned crazy things -- like kids didn't want to get out of the hot lava. They wanted to get in the lava! They want to splash around in it, and, in fact, lay down in it. So we made gameplay based on that activity. So we could kind of follow the kids as they led us through how their imaginations told them to play the game.
So the playtesting actually did affect design -- beyond the lava stuff.
TS: Yeah. We implemented a very basic level of functionality, and then watched the tests to see what the kids wanted to do. So we would make balloons -- we would do a physics simulation of balloons falling, and you being able to knock them around. I thought kids would just want to bop balloons up in the air, and it would just be fun to keep the balloons floating. I watched the tests, and all the kids wanted to do was pop their balloons. I don't know why it didn't occur to me. It was like, "Oh, of course." So we added a thing where you could pop the balloons.
But then some of the older kids, after they played it a while, they wanted something else to happen. They wanted some reason to pop the balloons. The younger kids don't really care, but the older kids wanted that. So we started putting in a progression of collectables you find in the balloons. It goes "bling!" and it makes a sound -- but there's no real score, no inventory or anything. It's all just kind of pretend, as in you pretend like you're playing a game. But it's really to give you something that you discover over time.
And there are little things like that in all the activities. There are behaviors that you don't understand until you do something strange, like stand still for a long time, or lay down under the lava, or collect a certain number of balloons, or pop this -- the hidden stuff comes out. Most of that came from watching playtests, and seeing what the kids wanted, what more experiences they wanted.