Crave

Intel CTO Justin Rattner on Thursday showed off this 3D input device that includes 'haptic' force-feedback technology.

(Credit: Stephen Shankland/CNET Networks)

SAN FRANCISCO--Most folks who try the Second Life virtual world grimace as the primitive 3D imagery drags its way onto their screens. Intel Chief Technology Officer Justin Rattner, though, smacks his lips with anticipation.

The chipmaker, always on the lookout for something that will give people a reason to buy a new PC, has reason to be excited about Second Life and its ilk. The technology, while still mostly for a fairly nerdy audience, has the potential to appeal to a broader audience than video games where overmuscled marines blow away aliens.

And just as significantly, Rattner said in a speech here Thursday at Intel Developer Forum, virtual worlds will stress out servers as well as PCs.

Intel has also eagerly anticipated some processor-taxing technologies that have come to fruition, including streaming audio and video, and some that haven't, such as speech recognition.

Rattner showed statistics that indicated a PC's processor bumps up to 20 percent utilization while browsing the Web, while its graphics processor doesn't even break above 1 percent.

Intel CTO Justin Rattner

(Credit: Stephen Shankland/CNET Networks)

But running Second Life--even with today's coarse graphics--pushes those to 70 percent for the main processor and 35 to 70 percent for the graphics processor, he said. The Google Maps Web site and Google Earth software pose intermediate demands.

Running a virtual worlds server is vastly more computationally challenging, though, when compared with 2D Web sites and even massively multiplayer online games such as Eve Online. An Eve Online server can handle 34,420 users at a time, but Second Life maxes a server out with just 160 users. Network capacity also is much more heavily used.

In addition, virtual worlds exercise parts of a processor such as math calculation engines that are idle when handling Web sites.

The ongoing tussle between Intel and AMD has dominated the news in recent weeks, but there's another potential battleground shaping up for Intel that could have a huge impact on personal computing.

A major topic I want to cover over the next several months is the looming showdown as the smart phone industry tries to develop more powerful computers, and the PC industry tries to build smaller and smaller computers. This week has provided a decent glimpse of Intel's vision of where it thinks the industry needs to go with its Silverthorne processor, designed for a new concept of computer called the Mobile Internet Device.

This is a concept Mobile Internet Device that Intel thinks people will be able to build with its Moorestown technology. It folds in the middle, like a book.

(Credit: Tom Krazit/CNET News.com)

We're looking at a major architectural battle over the next three years or so: the ARM instruction set, which dominates mobile phones, versus the x86 instruction set (Intel, please stop calling it Our Architecture). ARM isn't widely known outside the industry, but it designs processor cores for chips that power more than 90 percent of the mobile phones on the planet. Intel, you've probably heard of at one time or another.

Both companies and their partners will be aggressively courting computer users and software developers over the next several years. Intel is trying to find its next big source of growth by scaling down into power-sensitive areas such as MIDs, which are basically minitablet computers. ARM wants chip makers to use its cores in more powerful smart phones, such as Apple's iPhone, which uses an ARM-based chip made by Samsung. Both companies need the support of software developers who will be developing applications for their devices, and whoever has the best combination of compelling design and need-to-have applications will have the early lead as the first quarter of the computer industry winds to a close.

I'll get ARM's side of the story in more detail over the coming days. But Intel is contending that it has a major advantage in that all the software developed on and for PCs will run on its Silverthorne chips for MIDs, said Anand Chandrasekher, senior vice president and general manager of Intel's ultra mobility group.

Chandrasekher asserted that software developed for chips based on ARM cores aren't necessarily compatible. He said that's because ARM's licensees implement slightly different combinations of ARM technology, and there are no APIs (application programming interfaces) that lets application developers write an application that will run across those many different implementations. "If a smart phone is going to become more of a data oriented device, then it's going to have to run applications, then compatibility matters," he said.

Is that an upside-down iPhone? Nope, it's an upside-down shot of another Moorestown concept device.

(Credit: Tom Krazit/CNET News.com)

Intel is promising that any and all software written for PCs will run on MIDs. This does appear to be a more compelling argument for a software developer trying to decide where to place its bets. However, that's hardly a stamp of approval for the MID project at large.

If we've learned one thing this summer, it's that people are finally ready to start figuring out what they want on a mobile handheld device. The physical design matters, the user interface matters, and the applications matter. The key question is in what proportions. Intel might have an advantage when it comes to application development (although ARM probably has a retort), but will that matter in a world where more and more applications are probably going to delivered over the Internet?

I'd like to try and figure this out. Watch our site for a longer piece examining the two chip instruction sets and what they bring to the fight. This will take years to evolve, with things probably starting to heat up around the time Intel releases Moorestown, the successor to Silverthorne. Chandrasekher isn't saying much about Moorestown, but Intel is showing off these concept devices as a preview of what it thinks will be possible with that chip. One of them looks awfully familiar.

SAN FRANCISCO--Intel will begin building flash-memory drives into servers in 2008, starting with 32GB models that the company promises will boost system performance.

Flash drives can perform 10 to 50 times as many input-output transactions per second as conventional magnetic hard drives, said Pat Gelsinger, general manager of Intel's Digital Enterprise Group, in a speech at Intel Developer Forum here. In addition, they consume 4.5 times less power and write data at twice the speed.

Of course, the flash-drive capacities are much smaller. "The cost per bit is clearly going to be higher," Gelsinger said in a meeting with reporters. But some customers are bound much more by performance, he said, often running lots of hard drives in parallel but filling them only to 10 or 20 percent capacity.

Intel will offer samples of the technology this year and sell production models in 2008, Gelsinger added in an interview.

Intel will start with 32GB models, using NAND flash memory. "We'll have nice progression as we go to 64GB and 128GB over time," he said.

SAN FRANCISCO--This Nehalem plan better work out for Intel, because the chipmaker set very high expectations for the next-generation processor design Tuesday.

Pat Gelsinger, general manager of Intel's Digital Enterprise Group, speaks at Intel Developer Forum in San Francisco.

(Credit: Stephen Shankland/CNET Networks)

Pat Gelsinger, general manager of Intel's Digital Enterprise Group, demonstrated a Nehalem-based system at the Intel Developer Forum here that he said will bring major performance improvements for the company's x86 processor line. The processor family itself is due to arrive in 2008.

The Nehalem demonstration featured a system with two quad-core processors; each processing core can handle two independent instruction sequences called threads, and the demo showed all 16 threads at work on various tasks. The processor was the very first incarnation of Nehalem--the "A0" version--built for the first time three weeks ago, Gelsinger said.

"What you saw today was incredible health," he boasted during a meeting with reporters after the speech. "It really is pretty spectacular, and we're excited by the progress."

Nehalem brings major changes not just to the processor but also to the way in which it communicates with memory and other processors, a technology formerly called CSI, which variously stood for Common System Interconnect or Interface, and now branded as QuickPath Interconnect, or QPI. QuickPath reproduces a technique that rival Advanced Micro Devices used for years to market share against Intel and secure a solid position in all four major server makers' product lines.

The Nehalem processors demonstrated Tuesday each had four cores on a single slice of silicon, the approach AMD uses with its new Barcelona member of the Opteron processor family. In 2009, Intel will sell Nehalem processors with eight cores on a single slice of silicon.

Intel also is expected to sell less expensive Nehalem processors with dual cores per die, a source familiar with the company's plans said.

Update: I added some details about USB 3.0 device availability and performance.

Intel showed off prototype USB 3.0 connectors and an add-in card at Intel Developer Forum Tuesday.

(Credit: Stephen Shankland/CNET Networks)

SAN FRANCISCO--Intel and others plan to release a new version of the ubiquitous Universal Serial Bus technology in the first half of 2008, a revamp the chipmaker said will make data transfer rates more than 10 times as fast by adding fiber-optic links alongside the traditional copper wires.

Intel is working fellow USB 3.0 Promoters Group members Microsoft, Hewlett-Packard, Texas Instruments, NEC and NXP Semiconductors to release the USB 3.0 specification in the first half of 2008, said Pat Gelsinger, general manager of Intel's Digital Enterprise Group, in a speech here at the Intel Developer Forum.

In an interview after the speech, Gelsinger said there's typically a one- to two-year lag between the release of the specification and the availability of the technology, so USB 3.0 products should likely arrive in 2009 or 2010. A prototype shown at the speech is working now, and USB 3.0 will have optical and copper connections "from day one," he added.

The current USB 2.0 version has a top data-transfer rate of 480 megabits per second, so a tenfold increase would be 4.8 gigabits per second. Many devices don't need that much capacity, but some can use more, including hard drives, flash card readers and optical drives such as DVD, Blu-ray and HD DVD. The fastest flash card readers today use IEEE 1394 "FireWire" connections that top out at 800 megabits per second.

In addition, USB 3.0 will offer greater energy efficiency, Gelsinger said. It will be backward compatible, so current USB 2.0 devices will be able to plug into USB 3.0 ports.