Forum

Nvidia is developing new CPU cores based on the Arm architecture for PCs and servers that will be able to run Microsoft's upcoming Windows OS, the company said on Wednesday.

The series of CPU cores, which is code-named Denver, will be based on a future Arm architecture and give Nvidia a presence in new markets, Nvidia CEO Jen-Hsun Huang said during a press conference at the International Consumer Electronics Show in Las Vegas.

The announcement of new CPU cores came on the same day that Microsoft announced it is developing a future version of its Windows OS for the Arm architecture. The new Denver chips will support the upcoming Arm-based Windows OS, said Ken Brown, an Nvidia spokesman.

Microsoft's Windows OS, which is used on most of the world's PCs, currently works only on x86 chipsets from companies such as Intel and Advanced Micro Devices. An Arm version of Windows could provide users an option to buy Arm-based systems as an alternative to x86 products.


Intel says Light Peak interconnect technology is readyMicrosoft's slow, steady tablet strategy a big gambleCES 2011 daily video update: Panasonic TV tablet, Verizon 4G phones, Ford carsFord debuts all-electric carAn iPhone for Sprint? Unlikely, but maybe a WebOS device from HPQuickPoll: What's the highest-impact news from CES 2011?Intel to integrate DirectX 11 in Ivy Bridge chipsCES: Lady Gaga unveils Polaroid sunglasses camera, printerOLPC cuts XO 1.75 laptop to $165, power use by halfSprint plans WiMax RIM PlayBook Complete coverage: CES 2011 For many years, users had no choice on operating systems and chip architectures, said Bill Dally, Nvidia's chief scientist, in an e-mail message.

"Microsoft's announcement that it is bringing Windows to ultra-low power processors like Arm-based CPUs provides the final ingredient needed to enable Arm-based PCs based on Denver," Dally wrote.

"Denver frees PCs, workstations and servers from the hegemony and inefficiency of the x86 architecture," Dally said.

The downside is that software written for x86 chips will need to be tweaked before it can run on Arm systems.

Nvidia, primarily known as a graphics card company, has hinted in the past that it was developing a CPU for PCs, but it currently offers only the Tegra 2 chip for mobile devices such as smartphones and tablets.

The Denver CPUs will be based on future Arm processor designs, and Nvidia has licensed the Cortex-A15 processor and its successor design from Arm, Brown said. He didn't specific dates for when the CPUs will be in PCs, but he said they most likely would initially go into servers.

The chips will probably combine Nvidia's graphics processor cores with Arm's CPU cores, much like the Tegra chips. The integration of CPU and GPU technology in a single, power-efficient chip will help the company better serve the PC gamer market, Brown said. Another target market for these chips is cloud computing, Nvidia officials said.

During his speech, Huang was critical of the x86 architecture, which has dominated the PC market for decades. The growth of tablets and smartphones has created a "new PC industry," Huang said, which has led to an increased focus on Arm.

Arm is poised to become the new standard microprocessor architecture, and in a few years there will be more Arm processors than the number of x86 processors that "ever shipped," Huang said.

He provided the example of Apple, which introduced the iPad and iPhone and has revolutionized the way PC makers build and distribute devices, Huang said. Like Apple, a lot more companies are focusing on tablets and other mobile devices.

"The world has changed," Huang said. "The companies in the past that were leaders are readjusting their strategies."

A lot of software developers are also moving to the Arm architecture as they program for mobile devices, Huang said. With Denver, Huang hopes to attract mobile developers to write for PCs and servers.

Intel has dusted off the name Pentium as its mainstream performance brand for mobile computers.

PC Magazine says that Pentiums will appear inside machines in the higher-performance "ultrathin notebook" market segment, which is just above netbooks. Those systems require something with a little more bit-crunching oomph than an Atom processor but less power than a normal CPU chip.

Pentium has been an Intel brand since 1993 and has had good innings in terms of marketing. Apparently the label will be stuck on the 1.3-GHz ULV chip that Intel will call the Pentium SU2700. This beastie runs at a thermal design power of 10 watts

Intel has redesigned its logo stickers in an effort to "simplify" the buying experience. At the linked page they break them down and decode their meanings.

SOME EARLY benchmarks of the soon-to-be-released Core i7 975XE processor have shown up at Xbit Labs. This CPU is the performance king, replacing the i7 965, and breaks in a new – D0 – stepping. With it, Intel lowered power consumption and improved the OC’ing potential, it seems. Check it out at the link ...

Advanced Micro Devices' first 45-nanometer chip, the Shanghai quad-core Opteron, has made its debut at resellers.

The officially unannounced Opteron 837X and 838X series processors are not cheap. Online reseller PC Connection lists the Opteron QC (quad-core) 8384 at $2,509. Another reseller, Buy.com lists the same processor at $2,240.

The 8384 is expected to run at 2.7GHz and draw 75 watts, relatively low power consumption for a quad-core server processor.

The 8385--same clock speed with a faster system bus--is offered for $2,509 at PC Connection.

Other processors listed include the 8382 (2.6GHz), 8380 (2.5GHz), and 8378 (2.4GHz), priced at $2,177, $1,768, and $1,360 respectively at PC Connection. Note that these prices will differ from official pricing from AMD.

The Shanghai Opteron 230X series includes the 2382 (2.6GHz) and 2380 (2.5GHz). These are priced at $1,019 and $814 respectively at PC Connnection.

Rollout of the chip is expected officially on November 13, according to industry sources.

AMD is hoping to make a much better impression with Shanghai. Its first quad-core chip, Barcelona, was rolled out in September 2007 to great fanfare only to be delayed a whopping eight months (or more, depending how the delay is calculated) due to production glitches and bugs. This gave Intel an opportunity to regain ground it had lost to AMD in the server chip market.

Shanghai is in full production right now, Pat Patla, general manager of AMD's server and workstation chip business said last month. The was confirmed during AMD's earnings conference call earlier this month.

Server vendors are expected to be shipping systems as early as this quarter. A Sun Microsystems spokesperson said Tuesday that it plans to offer Shanghai processors on its current x64 platforms running Barcelona. Systems using the new processors are targeted for the first quarter of 2009, the spokesperson said.

At the same clock frequency (speed), Shanghai will outperform Barcelona by about 20 percent, Patla said last month.

AMD is also boosting the size of the cache memory, which typically speeds performance, from 2 megabytes to 6 megabytes. Another speed improvement will come from increasing "instructions per clock."

Patla also said last month that AMD is "turning on HT3 (HyperTransport 3)"--a communication path between chips--and that partners will start to validate systems in the first quarter of next year with this technology.

Intel Corp. expects its forthcoming Intel Core i7 processors to be much more powerful compared to existing central processing units, according to documents reportedly seen by the media. If the information turns to be precise enough, then Intel has all chances to keep processor performance crown even after smaller rival Advanced Micro Devices launches its new chips...

With the fourth quarter now upon us, the market's gaze has begun to turn full force on AMD and the imminent launch of the company's Shanghai processor. Shanghai, for those of you who haven't been following AMD, is a 45nm die-shrink of the company's Barcelona architecture with a 6MB L3 cache (up from 2MB on Barcelona), improved performance and power characteristics, and an additional set of as-yet-unspecified performance tweaks. On Monday, a reporter had a chance to sit down with Patrick Patla, head of AMD's workstation and server division, who offered insight into Shanghai's development process, as well as a few tantalizing details on what we can expect when the new processor ships...

Intel has developed an updated technology that could virtually solve questions of memory speed, the semiconductor firm has revealed at its recent Research@Intel event. Developers at the company have created dynamic RAM that behaves like typical system memory but needs just two transistors for each memory cell and which needs no capacitors. The invention makes dynamic RAM small enough that it can be embedded in a processor rather than put into a separate module, potentially eliminating several bottlenecks inherent to the static RAM used for cache on current processors.
A processor that complements or replaces existing cache would have much more on-chip bandwidth than any current processor and would be tied to clock speed, according to Intel. Where a 45-nanometer, quad-core Xeon currently has some of the fastest bandwidth available at between 18 to 20 gigabytes per second, a basic 2GHz processor made on an older 65-nanometer process could shuttle data at 128 gigabytes per second. Dynamic RAM also promises more storage in the same space and a lower price that could translate to the processors themselves.

The advancement could effectively overhaul Intel's approach to its chip designs as well that of programmers. Very fast memory access is considered critical to future many-core processors and may be necessary for the production equivalent of Intel's 80-core Teraflops Research Chip; developers writing for this and related processors may also depend on always having guaranteed access to cache, which would be virtually guaranteed with the cache becoming faster along with processors.

Intel hasn't said when or if it first expects to implement its two-transistor dynamic RAM into future chips, though mainstream processors with eight to 16 cores are due within the next few years.

Advanced Micro Devices (NYSE: AMD) on Wednesday introduced its two-year product road map for servers and workstations, saying it would release its first six-core chip next year and a 12-core processor on a new platform in the first half of 2010...

AMD is on target to ship its first 45-nm server chip, code-named Shanghai, in the latter part of this year, which would be about a year after Intel shipped its first products using the next-generation manufacturing process that makes it possible to shrink transistor size. Shanghai will be a four-core processor that delivers 25% better performance than the company's current 65-nm quad-core Opteron, formerly known as Barcelona. Shanghai also will ship under the Opteron brand.
Shanghai will use up to 20% less power during idle time than Barcelona and have 6 MB of L3 cache, which is twice the overall cache of its predecessor. In addition, Shanghai's support of DDR2-800 memory translates into a 10% boost in bandwidth.

SANTA CLARA, Calif., March 2, 2008 – The Intel® Atom™ processor will be the name for a new family of low-power processors designed specifically for mobile Internet devices (MIDs) and a new class of simple and affordable Internet-centric computers arriving later this year. Together, these new market segments represent a significant new opportunity to grow the overall market for Intel silicon, using the Intel Atom processor as the foundation. The company also announced the Intel® Centrino® Atom™ processor technology brand for MID platforms, consisting of multiple chips that enable the best Internet experience in a pocketable device.

The Intel Atom processor is based on an entirely new microarchitecture designed specifically for small devices and low power, while maintaining the Intel® Core™ 2 Duo instruction set compatibility consumers are accustomed to when using a standard PC and the Internet. The design also includes support for multiple threads for better performance and increased system responsiveness. All of this on a chip that measures less than 25 mm˛, making it Intel's smallest and lowest power processor yet.1 Up to 11 Intel Atom processor die -- the tiny slivers of silicon packed with 47 million transistors each -- would fit in an area the size of an American penny.

These new chips, previously codenamed Silverthorne and Diamondville, will be manufactured on Intel's industry-leading 45nm process with hi-k metal gate technology. The chips have a thermal design power (TDP) specification in 0.6-2.5 watt range and scale to 1.8GHz speeds depending on customer need. By comparison, today's mainstream mobile Core 2 Duo processors have a TDP in the 35-watt range.

 Articles

 Opinion Polls

  Join CPU-Central