Clay Shirky has been on a mission to debunk Linden Labs‘ claim of “two million registered users” for Second Life (or, specifically, Linden’s refusal to correct those who’ve made that claim), and has done so with icy effectiveness. He’s emerged as one of the leading Second Life skeptics, asking more questions about Second Life here and here. I should also say that Valleywag commenter tparisi wrote a thoughtful rebuttal to Shirky (one, however, that doesn’t question his statistical conclusions.)

IBM’s a great believer in the potential of virtual worlds; the team at eightbar created perhaps the best Second Life blog around, and Irving Wladawsky-Berger has offered some deep insight on what the growth of virtual worlds might lead to. Over 1,000 IBMers have joined IBM’s Second Life community; our CEO led an in-world event and most recently we introduced 12 new islands. All levels of the company - including some functions you wouldn’t normally expect - are aggressively exploring how to effectively use Second Life during the course of business.

With that level of participation comes the responsibility to ask tough questions. I think you’ll see IBM exert that kind of leadership as we continue exploring the potential of virtual worlds in the new year.

The new Forbes cover story explores the power of the Cell processor. Here’s how it begins:

IBM’s radical Cell processor, to debut in Sony’s PlayStation 3, could reshape entertainment and spark the next high-tech boom.

Later this year millions of homes will get a new supercomputer for the living room. Or maybe the playroom. Sony’s long-awaited PlayStation 3 game console, a slender yet muscular machine the size of a DVD player, performs a mind-boggling 2 trillion calculations per second. This kind of power, once reserved for seismic exploration and nuclear-weapons design, will let programmers create videogames that look as realistic as film.

Some techies say PlayStation 3, which may debut by midyear and could end up in 100 million homes in five years, will usher in the next microchip revolution. The Sony system owes its prowess to a microprocessor called Cell, which was cooked up by chip wizards at IBM (with help from Sony and Toshiba) at a cost of $400 million over five years. The Cell chip, based on a design inspired by supercomputers, runs at least ten times as fast as Intel’s most powerful Pentium. More important, Cell boasts a staggering fiftyfold advantage in handling graphics-intensive applications that will define the next generation of visual entertainment–blindingly fast and seductively immersive games, virtual-reality romps, wireless downloads, real-time video chat, interactive TV shows with multiple endings and a panoply of new services yet to be dreamed up.

The whole article is a must-read.

IBM today released the specifications of the chip that’s at the heart of the new XBox 360 at the Fall Processor Forum in San Jose.

The chip is a customized version of IBM’s industry leading 64-bit PowerPC core. Some highlights include:

– 3 identical multi-threaded PowerPC-based CPU cores operating at 3.2 GHz enhanced with specialized function VMX acceleration for gaming applications and a high speed 128-bit vector unit

– 1 MByte Shared L2 Cache with custom logic for high-speed data streaming for graphics and system applications

– 5.4 Gb/s per-pin Front Side Bus (with an aggregated bandwidth of 21.6 GBs)

– Highly configurable and programmable utilizing eFUSE technology

The chips are being made at IBM’s fab in East Fishkill, NY, and at a plant in Singapore owned by Chartered Semiconductor, who was a partner in the chip’s development.

Dean Takahashi writes a great post that explains the significance of this announcement to IBM and the technology industry as a whole. Pay attention to his discussion of the close collaboration between Microsoft and IBM that occurred throught the development.

This was a remarkable development cycle. Microsoft was absolutely determined to get the XBox 360 on the shelves globally before Sony could do so with PlayStation 3. As a result, IBM had to accelerate the development process, shaving as much as 30% - 35% off the “normal” development time while holding to Microsoft’s exacting quality and performance standards.

For IBM, this chip - along with the forthcoming Cell processor for PS3 and the chip we’re building for Nintendo’s forthcoming Revolution console - is physical validation of its microelectronics strategy. The kind of computing power used by the NextGen consoles strongly foreshadows where IBM believes computing is headed. A whole generation of developers, raised on the intense visual interface offered by computer games, will look to harness these capabilities to solve real-world problems.

As Takahashi notes of the Microsoft-IBM partnership at the end of his blog, “Talk about strange bedfellows. It’s like the PC wars never happened. And there’s new enemies to fight.”

Indeed.

Tags: IBM, xbox, Microsoft

Mercury Computer Systems just announced the first-ever Cell microprocessor-based product for non-gaming uses. It’s a Cell-based blade (for IBM BladeCenter servers), designed for use in graphics-intensive environments, like the medical, aerospace and semiconductor industries.

The blade software environment will run on Linux, and Mercury will provide the Eclipse-based open source software framework necessary to harness the Cell BE processor architecture – integrating the compilers, debuggers, math libraries, utilities and middleware in a seamless fashion.

Mercury expects to ship the product in early 2006. You can check out the specs here.

Hi Everyone,

My name is Peter Hofstee, I am one of the architects of Cell, primarily responsible for the synergistic processor. I have given quite a few technical talks on Cell, but in this blog I’d like to talk a bit about some potentially further reaching implications of the next generation of gaming technology.

For me the most significant aspect of this next generation of technology is the transition from consoles and games designed primarily for stand-alone use, to systems and games defined by broadband connectivity and interaction in real-time within a virtual 3-D environment.

The ability for large groups of people who are only virtually co-located to interact in real time I think is profound and is likely to have an impact far beyond gaming. It is likely to change education with virtual classrooms that can become better than real ones, for example by allowing everyone in the class to get an instant good view of someone asking a question. It is likely to change research and collaboration by significantly lowering the barriers to organizing an international conference or increase the value of a smaller meeting. It is likely to change the way sports are watched by allowing personalized points of view and audience participation. It will likely change business, with new opportunities for advertising and new ways for people to respond to advertisements and new ways for businesses and consumers to interact.

A key enabling technology that needs further development is the creation in real time of 3-D models of the real world (just like cameras construct a 2-D model). This brings world-data into the format that games and CG movies use. Today the process is in its infancy, for example requiring actors to put on black suits with white markers so the segmented motion can be tracked. In the case of medical data, x-ray or MRI data need a considerable amount of off-line processing to create a 3-D image that isn’t even based on objects. Still imagine for a moment that true 3-D recording technology to did exist and classrooms or sportsgames could be captured this way. Once the data is captured in this format, it can be manipulated, for example by adding other participants to the modeled classroom or conference. Also, just as in games, it allows an infinite choice of camera locations for final display as in the sportsgame example. Applications are endless; a surgeon can see an overlay of 3-D data from an MRI scan to what is visible to the eye, and may prefer this type of virtual view even when not operating remotely. The technology will improve virtually enhanced displays, not just for fighter pilots and airplane mechanics, but for car drivers and mechanics and plumbers, and electricians etc. etc. etc. I think this type of technology would have a lot more impact than conventional 3-D displays that offer only very limited freedom in the choice of a point of view, though of course 3-D displays and 3-D modeling and distribution are mostly orthogonal technologies and can be used in combination.

While the goal of full real-time 3-D recording will take some time to achieve, in many cases data that has been collected off-line addressing the static aspects of the scene that is being modeled can make the problem tractable. Examples include merely recording the positions and orientations of pre-modeled racecars on a pre-modeled race track to allow gamers to participate in the race, or recognizing the players and their body configurations in a soccer game to apply it to detailed 3-D models of the players that were previously built to allow viewers to control how they watch the game. Another example is the camera recording your body position and motion and applying it to a virtual character in a game. An example where the data is (almost) static is the collection of 3-D models of the world itself, in particular cities, to allow real-time exploration in 3-D. Perhaps an ultimate example is world modeling with enough detail to make a virtual hike worthwhile. Several of these exist already, what I have sketched is far from an all-or-nothing proposition. Most likely it will follow the paradox of recent technological progress (think MMOGs); viewed up close the rate of adoptions seems to be very slow, and yet, all of a sudden it seems to be there in a big way.

One last comment. For a field to truly drive the development of new technology, there have to be significant problems that cannot yet be solved, and drive the value people find in new generations of technology. This is now case with gaming and microprocessors, and the one class of problems I have described here a little bit is only one of many examples.

There was an interesting article in last month’s IEEE Spectrum on 3D displays. The article discussed three basic types of 3D displays. Stereoscopic, swept volume, and dithered slice (my words for the types). Relatively speaking the computational requirements to support the different types are quite different. The stereoscopic type requires 2X the graphics performance as the scene is displayed from two slightly different viewpoints. The swept volume type could take 1000X of the graphics performance as the third dimension is displayed. Lastly the dithered slice display which might have 16 layers in the display with computation needing to be done for each slice.

The primary issue will probably be the end user cost of the display technology. It seems to me that only the stereoscopic and dithered slice display approaches might be possible for the new generation of game machines although 16x or so graphics performance might put a strain on these machines. It’s more likely that there would be at least the opportunity to support stereoscopic 3D assuming a low cost hardware solution was available.

I suspect that it may be a few more years before the dynamics of cost effective displays, compelling 3D content (although most game format might support stereoscopic and dithered slice display technologies), and game and graphics performance align.