News and rumor site Fudzilla claims to have had an "interesting chat with key people involved in Nvidia's Ion project," all of which sounded pretty confident about the company's next generation Ion, or Ion 2.
As previously reported, the new Ion chip will act like a discrete GPU, a necessary transition because of how Intel has designed Pinetrail. It will also support Windows 7, boast over 5 hours of battery life, and according to Nvidia, has the potential to run 5 to 10 times faster than Atom with Pine Trail graphics. And while Pinetrail will only be able to play Flash content in SD, Ion 2 will support both SD and HD playback, according to Fudzilla's chit-chat. On top of it all, Ion 2 will support Blu-ray and 1080p, as well as mainstream gaming.
So what's the caveat? Put simply, Ion 2 will cost more than an Atom platform built around Pine Trail. That doesn't seem to bother Nvidia a whole lot, who says that its customers will put a higher value on HD, Blu-ray, transcoding, video editing, and casual gaming capabilities.
The University of Antwerp gave everyone a chuckle last year when they built a quaint little supercomputer made out of four high end Nvidia GPUs. Apparently, that was just a practice run. The same group has now constructed a 13 GPU monster of a supercomputer called Fastra II.
The rig contains six dual-GPU Nvidia GTX295 cards and a single GTX275. As you can imagine, there were a few issues getting the whole system up and running. Motherboard manufacturers don’t usually anticipate someone needing to run 13 GPU cores. With a little persistence and a custom BIOS from ASUS, the tiny supercomputer was up and running. The whole affair cost only 6000 Euros, and is capable of twelve teraflops.
The value per teraflop is high considering most conventional supercomputers cost millions of dollars to build and run. You can check out some possible applications and crazy benchmarks here.
Intel said today that it’s much hyped Larrabee graphics chip won’t be happening after all. The Larrabee GPU will exist only as a software development platform for the time being. According to Intel spokesman Nick Knupffer, "Larrabee silicon and software development are behind where we hoped to be at this point in the project. As a result, our first Larrabee product will not be launched as a standalone discrete graphics product.”
The Larrabee chip has already been delayed repeatedly, so this news isn’t entirely surprising. There have been indications of trouble ever since Larrabee missed its original debut in 2008. A recent demo of the chip at SC09 highlighted lackluster performance one wouldn’t expect after such a long development. It would have been Intel’s first standalone graphics part in over a decade, but now is little more than a sore spot.
Intel gave no clear guidelines on when the SDK can be expected, just sometime in 2010. The cancellation of Larrabee has no bearing on Intel’s upcoming hardware graphics solution for the Atom “Pine Trail” chip. Pine Trail will have graphics processing integrated in the CPU. Oh well Intel, you’ll always have Atom.
MSI is giddy as all get-out that one if its videocards -- MSI R5870-PM2D1G -- set the record for the highest 3DMark Vantage Performance score on the planet achieved by a single-GPU graphics card. The score to beat is now 24,486.
Benchmarking that high on a single-GPU videocard doesn't come easy. Renowned overclocker "Deanzo" from New Zealand on hwbot.org used a healthy dose of LN2 to cool down the card to -180C. He was then able to supercharge the GPU to 1,380MHz, or 62.4 percent higher than reference clocks.
As for the software, MSI says Deanzo used the company's Afterburner overclocking utility, which supports GPU, Shader, and memory clockspeed adjustments on both ATI and Nvidia videocards. The utility also makes it possible to increase the voltage and adjust the fan speed.
At Microsoft’s Professional Developers Conference the Redmond giant showed off an early build of Internet Explorer 9 complete with GPU acceleration. Not to be outclassed, Mozilla has indicated they too are working on GPU acceleration for the popular Firefox browser. After the Microsoft demo, Mozilla Director of Developer Relations, Chris Blizzard tweeted, “Interesting that we're doing Direct2D support in Firefox as well - I'll bet we'll ship it first. :)"
Later, Firefox developer Bas Schouten wrote about the addition of Direct2D to the browser. He said the browser wouldn’t look much different, but rendering should be much improved. Schouten provided benchmark data for Direct2D rendering compared to standard Windows Graphic Device Interface (GDI). While some sites showed little difference, several saw dramatic reduction in rendering times. Hopefully we’ll see this technology sooner rather than later. However, there are currently no ship dates for either product.
Perhaps looking to steal a bit of thunder from AMD's awesome HD 5970 videocard, Nvidia PR guy Brian Burke today posted a picture of the green team's upcoming Fermi-based graphics card. Isn't marketing fun?
On his Twitter account, Burke referred to the Fermi card as a GeForce 100, which he said is the first GeForce GPU based on the new architecture. A screenshot in the background shows the videocard running the Unigine Heaven DirectX 11 benchmark.
That's all that was said (and shown), but the bigger message is the unspoken one that says, "Hey, we're still here, and we're poised to kick AMD's tail." No one from Nvidia actually said that, mind you, but they might as well have if they're going to post a pic of their upcoming graphics card when the talk of the town is centered around AMD's flagship GPU.
Like everyone else, we're chomping at the bit to see what Nvidia's next-gen Fermi graphics chipset (GT300) brings to the table, so we're a little bummed it's been pushed back until Q1 2010. The delayed launch, however, isn't expected to negatively impact Nvidia's sales for the remainder of 2009, DigiTimes reports.
When Fermi does ship, Nvidia will position the GPU into three different product lines, including the GeForce brand, Quadro, and Tesla, according to company CEO Jen-Husn Huang.
While the delay is unlikely to hurt Nvidia's bottom line, that could change if Fermi gets pushed back a second time, some analysts warn. They say Nvidia could start to lose some of its market share should Fermi arrive in April 2010, as AMD will in all likelihood have already started shipping its entry-level Radeon HD 5600 and 5300 GPUs.
The shortage of the 40nm ATI Radeon HD 5000 series is being blamed on Taiwan Semiconductor Manufacturing Company's inability to keep pace with the demand due to low yields. On the other hand, the dearth of 55nm GPUs is due to the fact that they no longer figure prominently in AMD’s plans.
The report further claims that AMD has delayed the shipment of “its ATI Mobility Radeon HD 5000 series (Manhattan) GPUs for notebooks to the first quarter of 2010 from the originally planned fourth quarter of 2009.”
Intel hopes to eventually make a thunderous entrance in the discrete graphics market with its upcoming GPGPU chip codenamed Larrabee, and to do that, the company needs to line up some chip partners willing to jump on board. Intel CEO Paul Otellini set out to do that recently, talking privately to several China-based videocard makers.
According to what un-named sources have been whispering in DigiTimes' ear, Intel plans to offer preferential pricing for just the GPU by itself, as well as when bundled with other Intel products. This is a similar strategy to what Intel has been doing with its Atom platform, and it remains to be seen how many graphics partners will warm to Larrabee in this manner.
As it stands, some first-tier graphics card vendors are a bit leary about Larrabee on fears that the first release may end up buggy. But within the next couple of years, vendors expect Larrabee will be able to hold its own against what AMD and Nvidia have to offer.
Last month we posted a link to some early spy shots of the Radeon HD 5850 X2 & 5870 X2, but aside from the pictures, the post was a bit short on details. The card which still hasn’t been officially unveiled by AMD is still somewhat of a mystery, but the gang over at Alienbabeltech.com got a hold of some new photos and information that answers some, but not all of the questions we have about the new design.
Based on the reference card shown in the pictures, it appears as though this monster will require two power connections, one 8-pin and one 6-pin which are located just above the fan. Even if the power requirements of this beast don't shock you, the overall length of the card just may. Measuring in at a whopping 13.5 inches long, most enthusiasts without a full tower case will have a hard time fitting this into their machines. AMD responded quickly by stating that the photos were of an engineering sample, leading us to wonder if the card won’t shrink an inch or two prior to release.
The other big change since our last update is the new naming conventions. It now appears as though AMD will be dropping the “X2” designation and the cards will be known as the Radeon HD 5950 (Dual 5850’s), and the Radeon HD 5970 (Dual 5870’s).Questions remain, but hey, its better then nothing right?