Finally we have Microsoft with their Xbox 360. They had a video showing off their Project Natal, which apparently no one can agree on how to pronounce. The idea here, similar to Sony’s, is to use cameras to track what players are doing. Unlike Sony’s approach, though, Natal will use multiple cameras to accurately track players’ positions in 3D and, more importantly, doesn’t require holding anything special. As they say, “you are the controller.” In addition, they claim that you’ll be able to log in using facial recognition and interact with the Xbox using voice recognition. They even showed a demo in which a player had a conversation with a game character on screen, even sharing a note hand-written on a sheet of paper. All of this is much more ambitious and difficult to pull off than even what Sony is doing and if any of these 3 companies have the money to pull something like this off, it’s definitely Microsoft. That said, though, they’re promising the moon here and it would be pretty shocking to see them pull it all off.

There were several games on the PS2 and even a couple now for the 360 that track players’ motion with a camera to control the action on screen. Most of them haven’t done it very well and have ended up being more frustrating than anything else. The technology has improved, though, and I imagine that Microsoft has made it function a lot better, but it would still be a very impressive feat to get the games they showed in their video to actually function in a real-world situation. After all, just the problem of separating the person out of the cluttered background of your average house is difficult (what if there’s a picture of the family hanging on the wall behind the players?), but figuring out what objects are hands or faces is something that you usually see in academia. Even more difficult, Natal will have to be able to track which hands belong to which players (what if they walk past each other?) which is something I’m not sure any known research project has done yet. Again though, all of these problems are simply hard, not impossible. Someone is going to solve them eventually and there’s no reason it can’t be Microsoft. At least that part of their presentation is theoretically possible in the next couple of years, even if it’s not terribly likely. The same goes for using facial recognition to log in. It’s certainly not a super easy problem, but it doesn’t need to be highly secure (at least I hope we don’t start protecting our military installations with Xboxes), so that should be even easier than the motion sensing. Simple voice controls are easy, too. In the video they show a couple watching a movie and using voice instead of a remote control to pause or rewind. That’s all pretty straight-forward. The rest of it, though, having a conversation with a character on screen, taking a picture of a hand-written note to share with the character, etc. is just a pipe dream. Voice recognition just isn’t good enough yet to handle such complex conversation, especially if the house is noisy at all. Furthermore, even the most advanced AI researchers haven’t been able to make programs that are capable of holding conversations like that. The folks at Microsoft are incredibly smart, but I have a hard time believing they’ve succeeded there in a few years where many, many more have failed for decades. We’ll get there someday, but not anytime soon.

Also, nothing in their video mentioned how difficult it’s likely to be for game developers to actually make use of all these advanced features. Which is to say that it’s not likely to be easy at all. At best developers will get the same information about player movement that they get from the Wii and it will still be just as hard to interpret. We’ll see a few gimmicky games come out using Natal and maybe a few bigger games that support it as an option. It won’t be until at least the next generation of consoles, though, and maybe even the one after that before we see this kind of control scheme done well enough to be useful and fun.

With all three new control schemes, cost is likely to be a big barrier to entry. Guitar Hero has proven that people are willing to shell out the money for special controllers if you have a compelling enough experience, but how many games can provide that kind of experience with motion controls? As always there’s a bit of a chicken and an egg problem. Game developers don’t want to make a game that relies on these if most consumers don’t have them. At the same time, consumers won’t want to buy the new controllers if there aren’t any games for them. So that’s all just to say that we’re likely to see some of this come out for the current generation of consoles, but as I said, it won’t become wide-spread or that useful until at least the next generation of consoles.

Pages: 1 2

This entry was posted on Thursday, June 11th, 2009 at 1:22 am and is filed under Science & Technology, Technology. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.