Posts Tagged ‘ Nvidia ’

Nvidia comfirms commitment to android – The Inquirer

Nvidia comfirms commitment to android – The Inquirer.

IN A CONFERENCE CALL about Nvidia’s first quarter financial results, the Green Goblin’s CEO, Jen-Hsun Huang said that his plans for a next-gen Tegra include the use of Android.

With every man and his dog seemingly wanting a smartphone, Nvidia is looking to get a slice of the market, and it sees Tegra as the way to get its foot in the door.

Tegra, its mobile web processor, will be able to run the Android 3 OS, Jen-Hsun added, and that will help it take on other, more established firms. “Although it made sense for the first-generation Androids to use available phone processors, the follow-on generations of Android are really going to go after performance,” he said in the conference call, which is transcribed here. “And Iphones are out there, the Iphone 4G is coming, the Ipad is obviously a revolutionary product. The bar is pretty high for all of the mobile players.”

Jen-Hsun knows where the competition will come from, but added that Tegra plus Android 3 will prove to be a winning formula. “Prior to Tegra, there are only two application processor companies out in the mobile space, right? Basically, it’s Qualcomm and TI, and they both make wonderful application processors,” he said.

“Our differentiation and our contribution to the space is where multimedia, high resolution, snappy graphics [are] really necessary. And the first-generation smartphones had pretty low resolution displays. And so snappy graphics and high-performance multimedia and high resolution just wasn’t as much of an issue. But [now] resolution’s a huge issue. And so that’s our contribution and that’s our differentiation and that’s what people are seeking out in the market.”

Responding to questions about its PC-bound heat generator, the Fermi GPU chip, Jen-Hsun, said, “We are ramping Fermi as we had talked about before, and the success of Fermi is certainly as we expected.”

In the meantime, Fermi is still being scrutinised. “The amount of testing that we have to do for Fermi GPUs [is] longer than mainstream products because they’re just much, much larger GPUs. The Fermi GPU, as you know, is some 3 billion transistors, and so there’s a lot of testing to do in it,” he added.

“Now of course we’re ramping into a fresh new market and a fresh new product and there’s a lot of pent-up demand, and so we just needed to keep the pressure on it and just keep cracking through it”.

Yep, we reckon it won’t be long ’til next winter, surely a good time to start seeing Nvidia’s Fermi GPUs. µ


Chief Scientist of Nvidia Condemns Moore’s Law, Microprocessors – X-bit labs

Chief Scientist of Nvidia Condemns Moore’s Law, Microprocessors – X-bit labs.

William Dally, chief scientist and senior vice president of research at Nvidia, said in a column that Moore’s Law was no longer enabling scaling of computing performance on microprocessors. In addition, Mr. Dally indicated that central processing units (CPUs) in general could no longer fulfill the demand towards high performance.

“[Moore’s Law] predicted the number of transistors on an integrated circuit would double each year (later revised to doubling every 18 months). This prediction laid the groundwork for another prediction: that doubling the number of transistors would also double the performance of CPUs every 18 months. [Moore] also projected that the amount of energy consumed by each unit of computing would decrease as the number of transistors increased. This enabled computing performance to scale up while the electrical power consumed remained constant. This power scaling, in addition to transistor scaling, is needed to scale CPU performance. But in a development that’s been largely overlooked, this power scaling has ended. And as a result, the CPU scaling predicted by Moore’s Law is now dead. CPU performance no longer doubles every 18 months,” said Bill Dally in a column published at Forbes.

Perhaps, performance of CPUs no longer doubles every year and a half, but, firstly, those chips are universal and very flexible, secondly, they can be manufactured in large volumes. Graphics chips, which, from time to time, outpace the Moore’s Law, quite often cannot be manufactured in large volumes because of poor yields. Moreover, although GPUs can provide higher horsepower than CPUs, they are not that universal and flexible.

Even though historically developers of central processing units were concentrating on increasing clock-speeds of chips, five years ago Advanced Micro Devices and Intel Corp. concentrated on creating more parallel multi-core microprocessors that work on moderate clock-speeds. However, the vice-president of Nvidia also claims that multi-core x86 CPUs will ultimately not solve problem with the lack of necessary computing performance.

“Building a parallel computer by connecting two to 12 conventional CPUs optimized for serial performance, an approach often called multi-core, will not work. This approach is analogous to trying to build an airplane by putting wings on a train. Conventional serial CPUs are simply too heavy (consume too much energy per instruction) to fly on parallel programs and to continue historic scaling of performance,” said Mr. Dally.

It is rather logical that Nvidia calls central processing units obsolete since it does not produce them or develop them. The big question is whether AMD and Intel give up and let Nvidia to actually capture part of the market of high-performance computing, where multi-core CPUs rule today.

“Parallel computing, is the only way to maintain the growth in computing performance that has transformed industries, economies, and human welfare throughout the world. The computing industry must seize this opportunity and avoid stagnation, by focusing software development and training on throughput computers – not on multi-core CPUs. Let’s enable the future of computing to fly – not rumble along on trains with wings,” concluded the chief scientist of Nvidia. – Nvidia blames sales shortfall on TSMC – Nvidia blames sales shortfall on TSMC.

SAN JOSE, Calif. — Nvidia Corp. posted strong results for the quarter, but the graphics chip maker said that it is still suffering from capacity constraints from its foundry vendor.As reported, Nvidia has been suffering due to a shortfall of 40-nm capacity from its foundry partner, Taiwan Semiconductor Manufacturing Co. Ltd. (TSMC). TSMC’s 40-nm yield problems surfaced last year, but the company claimed it largely resolved the problem.

Nvidia is still having issues with TSMC. In a conference call, Nvidia claimed it could have shipped more products in the quarter. Nvidia said it had a $100-to-$150 million sales shortfall in the period.

But it continues to see a shortfall in 40-nm capacity. ”40-nm is constrained around the world,” according to representatives from Nvidia, during the call. ”Supply will be constrained for the first half of this year.”

Nvidia stopped short of slamming TSMC. TSMC has improved their yields and is doing a ”fabulous job,” the company said.

Nvidia also reported revenue of $982.5 million for the fourth quarter of fiscal 2010 ended Jan. 31, up 9 percent from the previous quarter and more than double the $481.1 million reported in the same period a year earlier.

On a GAAP basis, the company recorded net income of $131.1 million, or $0.23 per diluted share, for the fourth quarter of fiscal 2010, compared with a GAAP net loss of $147.7 million, or $0.27 per share, in the same period a year earlier.

For the full fiscal year, revenue was $3.3 billion compared with $3.4 billion for the fiscal year ended Jan. 25, 2009, a decrease of 3 percent. GAAP net loss for the fiscal year ended Jan. 31, 2010 was $68.0 million, or $0.12 per share, compared with a net loss of $30.0 million, or $0.05 per share, for the fiscal year ended Jan. 25, 2009.

”Nvidia’s business continued to accelerate in the fourth quarter, with strong demand in our PC and workstation markets,” said Jen-Hsun Huang, president and chief executive, in a statement.

For the current quarter, revenue is expected to be flat from the fourth quarter.

Tegra 250 sets mobile speed record


  • second-generation Tegra processor has been sampling since 3Q09 and is expected to reach production in 2Q10
  • contains two Cortex-A9 CPUs, each running at up to 1.0GHz.
  • should score 5760 on the Coremark test (using two threads)
    • more than twice the speed of the iPhone 3GS, Nexus One, or any other smartphone shipping today
    • 40% faster than the Coremark score of the 1.6GHz Atom CPU (also using two threads with hyperthreading).
  • exceeds 1.0W at full speed from both cores
    • 150mW for HD (720p) video playback and just 15mW for audio playback –> similar to original Tegra
  • can decode video at 30fps and full HD (1080p) resolution for H.264 base and main profile
    • restricted to 720p for the ultrachallenging H.264 high profile (Blu-ray)
    • can also encode video at 1080p for H.264 base profile.
    • Few other mobile processors can handle 1080p video at even the base profile.
  • 3D performance : peak speed of 90 million triangles per second.
  • will appear initially in tablet computers and similar-sized devices that will debut by mid-2010.
    • smartphones will not ship until late 2010
  • third-generation processor expects to sample before the end of this year