CNET también está disponible en español.

Ir a español

Don't show this again


You can believe in Microsoft's Project Natal

It may sound too good to be true, but Microsoft seems to have put together something that does exactly what it says it will. Now, can game developers find a way to incorporate the technology into their projects?

LOS ANGELES--About halfway through a closed-door demo I was in this afternoon of Microsoft's just-announced full-body motion-sensitive control system, Project Natal, another reporter told our host that he was skeptical of what he was seeing.

A minute later, after taking the virtual controls himself of the game "Burnout Paradise" and giving Natal a test, the reporter walked back over to where I was standing and when I asked him if he was still skeptical, he gave me a chastened look and said, "It's interesting."

In other words, he was won over.

Project Natal, as you probably know by now, was Microsoft's big announcement at its E3 press briefing here. It has gotten a lot of attention because of its promise to make it possible to incorporate all kinds of hands-free control into a wide variety of games: racing games, painting games, shooting games, sports games and so on.

Daniel Terdiman/CNET Networks

But until now, I hadn't seen the technology close up. I was part of a small group that got the demonstration deep inside Microsoft's cavernous E3 booth. We were allowed to ask whatever we wanted, but we weren't able to take video or photos of the technology.

One thing that came out of the presentation: Tsunoda said unequivocally that the software behind Natal was developed entirely in-house at Microsoft. But he wouldn't address the question of where the hardware came from, or specifically, if it came from 3DV Systems, a company that has been working on this kind of technology, and which Microsoft recently acquired.

Having seen the 3DV technology a couple of years ago, I can say that what Microsoft is showing today is very much the same but with a much more user-friendly front end. In other words, I would bet that 3DV is the source of that hardware.

Here's my quick impression: Natal is for real, and it may well change the way people experience video games, as well as anything else that is run through an Xbox 360.

Tsunoda began our demonstration by explaining the problem Microsoft had set out to solve: To make the gaming experience fun for everyone, while not alienating the core Xbox 360 fans.

"The control system is simple," Tsunoda said of Natal. "People can just jump in and have fun. (But there's) an extra layer of fidelity for core gamers."

Project Natal (see video below) is designed to be a one-to-one avatar control system, Tsunoda said. Wherever you move your hands, your body or your legs, the system captures it and mirrors it on the screen and in whatever game you're playing. "No (other) controller in the world allows you to control your whole body," he said. "Every part of the body is in play."

One interesting thing that came up in the demo is that when a woman stepped up to use it, the system recognized she was female and represented her on-screen as a female avatar with long hair. Tsunoda said that ideally, Natal will recognize users and be able to grab their existing Xbox avatars, but that in such a demo environment, it simply represented her the best way it could, given what it could see of her skeletal structure.

Another interesting point was the way Natal recognizes people's skeletal structure and analyzes how we move. Tsunoda made the point that Natal will continue to work even if someone walks in front of a player because it knows how the human body works. So, if a player had his or her arms blocked, but Natal's cameras could still see part of their arm, it can fill in the rest based on algorithms that tell it how that arm should look.

And it's the software, Tsunoda said, that's the "magic" behind Natal, and that allows the technology to "extract the human skeleton."

Natal is designed to work whether someone is standing up or sitting down, and can recognize users very quickly. We saw that in action when, one-by-one, we were invited to step up and play either a kickball game or a driving game. With a couple of exceptions where the player didn't stand in the right place, Natal did seem to almost instantly recognize that a new person was playing and, then, respond to their movements.

This may have been most impressive during game play of the racing game, "Burnout Paradise," when it was clear that Natal was doing a fine job of translating the player's hand movements--mimicking holding and turning a steering wheel--into moving the car on-screen.

Tsunoda said the technology behind Natal includes an RGB camera, an infrared camea, a multi-array microphone and a depth map. These features allow the system to track a player in 3D space, as well as to capture spoken commands from multiple people, none of whom have to wear a headset.

Asked how Natal differs from the many other motion-control cameras that have come along over the years, Tsunoda simply said that nothing that has ever come along before has been able to instantly work when a new player steps up in front of it, or when the lighting conditions change, or when someone else steps in front of a player.

"Ours, you can play any way you want," he said. "This just works the way you want it to."

Clearly, Microsoft is banking heavily on software developers--who are just now getting development kits--being able to utilize Natal in their games in such a way that players don't have to do any kind of configuration or tinkering in order to get it to work. Absent that instant-workability, the system loses a lot of its allure. But Tsunoda was insistent that that is Natal's value proposition and that developers will have no trouble making it work that way in their games.

If that's true, then it would seem that Microsoft has a real winner on its hands. As I wrote Monday, any success depends, though, entirely on price point, and how users get their hands on it. But right now, having seen Natal close up, I have to say I'm a believer.