Interview With Victor Zordan, Expert In Video Game Technology
GF: How accessible is motion capture technology to indie game developers?
The Kinect system is giving motion capture to the masses. It’s not yet there, in terms of the set of applications and drivers that are being used, but I believe at $150, it’s really giving it to the masses. It’s available, but it’s still very young. We’ll now see that there’s this whole aftermarket of people who are developing ways in which you could use this technology to do other things, like pure motion capture.
I have a student who is working at a local community college with limited resources and he is setting up a system where he’ll have a Kinect in a room dedicated for recording motion capture. The community college’s animators will be using a Kinect to get their motion capture.
How easy it is for the Kinect to be set up as a motion capture system — and the quality of the system — are the two things in question; and those will both move in the right direction as this technology matures. In a year or two, it will be surprising what we can already do.
GF: In movies, you can sometimes tell CGI from an actor because the animation feels too smooth, too fluid. Is that a reflection of the motion capture technology?
Motion capture introduces potential high-frequency outliers and problems that you can eliminate by filtering. Filtering will smooth the data. But it has the impact of making the character’s motion look too fluid.
When we move, we have high-frequency content that is unfortunately removed when you smooth and filter the data. That high-frequency content is information that we perceive in other humans, but it also includes the information that is the outliers, the noise, the things you don’t want.
It really has to do with our perception. If we look back at special effects from movies in the ’90s, we cringe at them now, because we have begun to fine tune our perception of CGI. But when we saw these movies the first time, we accepted them. As our awareness of CGI became heightened, we became more sensitive to it, and we’ll see motion capture go through such a transition too. It’ll become grittier as they remove the need to smooth, and our perception will become less accepting of the smoothed data.
GF: Between the game industry and the film industry, who is pushing us to the limits of CGI technology?
The movie industry gets the luxury of doing anything they want at their leisure, meaning they can take 24 hours to render something. They can take advantage of cutting-edge techniques that are too slow for a game, which in the end, makes them the visual powerhouse.
The idea that the game engine has to be fast and has to be interactive means that developers do not get the luxury of doing anything they want. Instead, they have to put up with the limitations of the hardware and be able to render in real-time.
That said, there is a trade-off, because there are actually things that are really nice about having interactive graphics and there’s discussion about using interactive graphics to do videography in the movie industry. Being able to watch what’s happening in an interactive way and being able to control your camera — this is really the movie industry drawing form the game industry. Camera control is something the game industry has had to fight out, because it has to be right. You need to be able to clearly see if you’re moving through the world. This is an example of where the game industry is the cutting edge.
But if I have to paint the plainest picture possible, the fact that the movie industry gets to do everything offline means that they get to do everything first, because they can do it slowly and play it back in real-time. The game industry has to wait until innovation has come to find the shortcuts that will make the technique fast enough for a game. The difference, though, between real-time and offline rendering is shrinking.
GF: What else has the film industry borrowed from the game industry?
In the movie Sucker Punch, they practically used in-game scenes in the movie, which is almost the inverse of what is done in the game industry with cinematics. They’re exploiting a certain level of visual appeal that is related to the games.
It’s almost like the first time you went back and used an 8-mm camera on purpose because you wanted your film to look grainy. People were moving away from this grainy look until someone had the idea that looking grainy was cool, and Sucker Punch has this sort of effect with games. They decided, “You know what? Being in the game is cool.” So they made these things that purposely looked bad in comparison to the rest of it, and bad in a way that was good.
I feel that the movie industry moving to 3D is also an example of this because you end up traveling with the character. You’ll have this big chase where you’re practically in the first person perspective, and I think that smacks of the excitement of the games, where you get to “ride down the tunnel.”
In many ways, we’re starting to see a marriage of these two media forms. Some movie scenes seem built specifically so that they can be brought into the game. When you see the characters riding that cart down the tunnel, you know it’s going to be in the game.