Thoughts on Microsoft's new Xbox One

Yesterday, Microsoft unveiled its new Xbox One game console to an assembly of press, analysts and Microsoft employees on its Redmond, WA campus. The Xbox One has faster processors and more memory than the Xbox 360. A new version of its Kinect 3D digitizer with a 1080p camera is included as standard equipment. The Xbox Controller has also been redesigned, although the changes are mainly cosmetic. In addition, the Xbox One has an HDMI input, so that selected cable, satellite and IPTV set-top boxes can be connected to and controlled by the Xbox One.

Microsoft spent the first half of the presentation focusing on the Xbox One's TV-related features. For example, the Xbox One will have a built-in Electronic Program Guide (EPG) that supports many video operators. Users will be able to change channels and look for shows to watch by voice. In addition, the Xbox One will enable navigation via Kinect gestures.

The second half of the presentation was devoted to games. Only a handful of game publishers were represented on stage, and none of them showed actual game play; instead, they showed trailers. To my eyes, the most impressive trailer was for Forza Motorsport 5, which is the only game title that's been confirmed to be released day-and-date with the Xbox One. It looked great, with visual elements such as metallic paint and realistic depth-of-field rendering that would have been possible only in pre-rendered cutscenes not long ago. Unfortunately, that wasn't the case with the demos from the other game publishers. Electronic Arts, for example, appears to be using the Xbox One's additional horsepower to add more intelligence to the game play in its sports titles rather than for improving how its games look.

Microsoft is apparently concerned about the future of the Xbox given the falloff in sales of console games and the rise of casual games on smartphones and tablets. As a result, it's trying to position the Xbox One as both a set-top box (one that can both connect directly to content over the Internet and indirectly through cable, satellite and IPTV set-top boxes) and a high-performance game console. The problem is that those are two very different markets, with different use cases and consumer expectations. For example, Google TV provides most of the same non-game functionality as the Xbox One, albeit without voice recognition or gesture control. On the other hand, you can buy a Google TV-based set-top box from Vizio for $99, while I expect the Xbox One to be priced around $399. You can also buy an Apple TV or Roku set-top box for $99 or less. I simply don't see very many people buying the Xbox One for its set-top box features, since they can get most of its functionality from less expensive competitors. That means that most of the Xbox One's buyers will be hard-core or moderate gamers, which won't expand the potential market for the device at all.

I suspect that Microsoft's corporate leadership has fallen victim to Shimmer Syndrome (named after the combination floor wax and dessert topping in the famous Saturday Night Live commercial parody.) As with Windows 8, which works both on tablets and on conventional PCs but is compromised on both platforms, it's trying to make the Xbox One work both as a set-top box and game console. The compromise on the set-top box side is clearly price; we don't yet know what the compromise is on the game side, but it may be lack of attention that opens the door for Sony to offer a superior developer and gaming experience.

Someone at Cisco is having a very bad day

Engadget is running a story about Oliver Kreylos, a lecturer and researcher at UC Davis. He was one of the earliest experimenters with open-source drivers for Microsoft's Kinect, and he's posted a video demonstration of his primary project: 3D telepresence. By placing two Kinects almost 180 degrees apart and facing each other, he's able to create a live 3D image of a remote participant in a teleconference, matted into a virtual set that looks like an office. Here's a video of the demonstration:

To navigate around and through the remote scene, he's using a Wiimote to move a virtual camera (the actual Kinects stay in fixed locations). The person in the remote location can see the location of the "camera" by donning 3D glasses.

The quality of the video is still somewhat lacking--the resolution of the Kinects is 640 x 480--and there's just so much live streaming data that you can push over a USB 2.0 interface. But, Kreylos did this with two $150 Kinects--everything else either already existed in his lab, or he could get it easily.

Cisco sells telepresence systems that are designed to make you feel that you're in the same room as the other person, at prices of $300,000 a system and up. They have less-expensive systems that are essentially minor variations on teleconferencing systems sold by many companies. If a developer at UC Davis could cobble together a 3D telepresence system using $300 worth of off-the-shelf hardware, how can a business justify $300,000 for a system that does something similar?

We've been waiting for Microsoft's "next great innovation" for years. Surface wasn't it, nor was Zune, and neither will it be Windows Phone 7. It looks like Kinect is the first truly game-changing (no pun intended) innovation from Microsoft in years. So, start thinking about what can be done with HD Kinects using USB 3.0 interfaces. This is the 3D future, folks--not Jeff Katzenberg rolling out endless inane 3D movies, but rather, individuals, businesses, schools and institutions integrating 3D into our everyday lives.

Microsoft (finally) does the right thing with Kinect

Earlier this month, shortly after Microsoft's Kinect shipped, a group of open-source hardware developers called Adafruit Industries offered a $1,000 bounty to the first person who wrote and released open-source drivers for the device. In response, Microsoft told CNET: "Microsoft does not condone the modification of its products. With Kinect, Microsoft built in numerous hardware and software safeguards designed to reduce the chances of product tampering. Microsoft will continue to make advances in these types of safeguards and work closely with law enforcement and product safety groups to keep Kinect tamper-resistant." In response, Adafruit increased its bounty to $2,000, and then $3,000.

To an extent, Microsoft's position was understandable--the Kinect was intended to drive sales of Xbox 360s and Xbox games, and a Kinect sold for some other purpose wouldn't generate additional revenue for Microsoft. Also, if the number of Kinects available for the holiday season is limited, Microsoft might not have enough to meet demand. However, the ham-handed way that Microsoft went about threatening anyone who dared write a driver for the Kinect, which at the end of the day is simply a USB 2.0 device, backfired.

Less than a week later, a Spanish developer Hector Martin wrote and released his open-source driver and won the Adafruit bounty. Since then, other developers have begun adapting the Kinect for a variety of applications, such as 3D video and contactless measurements, and as the vision system for a robot. It has great potential in education, machine vision, communications and a variety of other applications.

Yesterday, after the EFF weighed in, Microsoft apparently "saw the light." CNET reports that a Microsoft representative on NPR's Science Friday said that the Kinect was left open by "design", and a tweet from the Science Friday account stated that "(Xbox director of incubation) Alex Kipman says Kinect interface was left unprotected 'by design.' [And Microsoft's] Shannon Loftis says she's 'inspired' by community finding new uses." Adafruit Industries replied on its blog with "Congrats to everyone in the open source community, in about one week we turned 'work closely with law enforcement' to 'inspired' by community finding new uses for Kinect."

Microsoft's Applied Sciences Group demonstrates glasses-free 3D system

At last month's SID conference in Seattle, Microsoft's Applied Sciences Group demonstrated a 3D display technology that doesn't require glasses. It uses a "wedge lens" which modifies the critical angle of light coming from the lens/backlight to determine the direction and location of light coming through the LCD. The system also requires a camera to determine the location of viewers in order to provide a convincing 3D effect, but Microsoft's new Kinect (formerly Project Natal) should do the job.

There are some limitations to the system: A minimum 240Hz refresh rate is needed for two people to be able to simultaneously view 3D video, and the viewing angle is currently only 20 degrees, although Microsoft hopes to increase it to 40 degrees. Microsoft is also encouraging LCD manufacturers to go beyond 240Hz in order to increase the number of simultaneous viewers.

My belief is that 3D in the home won't really take off until glasses-free systems become practical, both technically and economically. We may still be five years away from that happening, but once it does, demand for 3D will explode.