Our review of the ATI FirePro V8750 workstation GPU showed that the card is blazing fast at DCC. We tested the V8750 against the venerable NVIDIA GeForce 8800 GTX, and the difference in performance was staggering. But the 8800 GTX is almost four years old, and the question kept pouring in: How would the V8750 fare against the fastest single GPU on earth?

It just so happens that Icrontic recently fitted its test bench with such a GPU, the ATI Radeon HD 5870. And with that, it’s time to finally put the question to rest: How does the best workstation card AMD offers compare to the Radeon HD 5870, a desktop card that’s 100% faster on paper? The results may shock you.

The ATI Radeon HD 5870 is the flagship DirectX 11 GPU. Released on September 23, it is the fastest of the four available GPUs compatible with DirectX 11. As with any new flagship GPU release, gamers and enthusiasts have flocked to NewEgg and other retailers to buy it. It’s fast, it’s efficient, and it answers “yes” to the legendary “Can it play Crysis?” question. But will it blend?

(more…)

Over the weekend EVGA announced the release of the EVGA GeForce GTX 275 CO-OP graphics card which places a GeForce GTX 275 and GTS 250 on the same board, with the latter serving as a dedicated PPU.

The EVGA GeForce GTX 275 CO-OP.

Technologically, the card is an interesting achievement in that “first and only of its type” sort of way.

But when it comes to value, the card takes a big tumble; there aren’t very many AAA titles with a notable NVIDIA PhysX implementation. We also have to consider the GTX 200’s march towards EOL and the introduction of the Radeon HD 5000 series, both which further erode this product’s value.

The GPU’s price is also of concern. With an MSRP of $349, the card is similarly priced to the GTX 285, an adapter which definitely has a leg up in the performance department (to say nothing of the 5850).

Maybe somebody really loves Arkham Asylum this much.

After an OEM-only soft launch in July, NVIDIA has finally entered the 40nm game with a pair of 40nm GPUs in the retail channel.

Called the GeForce GT 220 and 210, respectively, the new cards are a die-shrunk version of the G200b core that powers adapters like the GTX 275.

As the table illustrates, the new goods are nothing to write home about. On paper, both of them are slower than the rustic GeForce 9600 GT, but they are the company’s first DirectX 10.1 and SM4.1 products.

The 220 also features an onboard audio controller for 7.1 sound via HDMI, so some may find the card useful for HTPC applications.

The GeForce GT 210

The GeForce GT 210 ($44.99) and the 220 ($69.99) are available from your favorite etailer immediately.

NVIDIA has just revealed the first details of their next-generation video card during the keynote address delivered by NVIDIA CEO Jen-Hsun Huang.

They call it “Fermi,” and it’s NVIDIA’s newest and most radical GPU architecture. Huang implied that the card is the symbol of the company’s efforts to embrace the GPGPU, or a video card which can handle CPU-like tasks.

An x-ray of the Fermi GPU die.

(more…)

Advanced Micro Devices has begun an initiative to create an open source and cross-platform solution for GPU physics.

Dubbed “Bullet Physics,” AMD has partnered with Pixelux Entertainment to end the clash between proprietary physics solutions like AMD’s Havok and NVIDIA’s PhysX. The new engine is written in OpenCL, a vendor-neutral platform which simplifies the process of coding for today’s processor-like GPUs. The duo is also working to develop Bullet Physics for DirectCompute, or the DirectX 11 standard for GPU offloading.

AMD’s CTO of Graphics Eric Demers said that an open physics engine will end the petty war between incompatible physics engines.

“Proprietary physics solutions divide consumers and ISVs, while stifling true innovation; our competitors even develop code that they themselves admit will not work on hardware other than theirs,” he said. “By working with Pixelux and others to enable open support of physics on OpenCL and DirectX 11 capable devices we are taking the exact opposite approach.”

The move is an obvious slap at rival firm NVIDIA which moved in June to disable PhysX offloading if a GeForce is not the primary renderer. While that change and its motivations are only just now coming to light, AMD seems to have short-circuited the brewing conflict with an engine that will run on GeForces and Radeons alike.

“Pixelux wants ensure that our technology can take advantage of the computing resources that any particular hardware platform offers without locking in our users to any single platform,” said Mitchell Bunnell, CEO of Pixelux. “By working with AMD to run our software in OpenCL we stay true to that goal.”

The maturity of the engine and a list of supporting developers were not immediately available. We have also contacted NVIDIA for a statement regarding their perspective on the new initiative.

It’s Monday! That means you’re once again chained to your desk, blearily contemplating the most creative ways to dodge the work that should be done at said desk. It also means you’re ready to reconnect to the internet for a dose of what you missed while you were being a responsible-get-projects-and-chores-done-on-the-weekend person (RIGHT?)! And, hey, if you need a dose, we have just what the doctor ordered.

NVIDIA to finally offer retail 40nm?

Coming in at numero uno is NVIDIA, with word that the firm might eventually push select 40nm parts into the retail channel. The firm’s shift to 40nm has been a painful affair plagued with poor yields and epic delays, but a few GPUs on the 40nm node have been available to OEMs for a while. According to the Chinese Digitimes publication, that OEM exclusivity may soon end.

The Digitimes report says that the DirectX 10.1-capable GeForce 210 and 220 products could hit retail as early as October. Even though GeForce 210 and 220 parts are far from benchmark-busters, the move indicates that NVIDIA’s 40nm process is (finally) maturing to support retail sales. If this is indeed the case, we are at last moving closer to the GT300.

Dell boards the Windows 7 train

If you’re saddled with a box from the boys at Dell and have been itching to try Windows 7, fortunes just may have turned in your favor. The big ol’ OEM took the weekend to announce a full roster of Windows 7-ready drivers for a wide variety of systems past and present. If you’ve been cautious about taking the leap to 7 to avoid driver issues, maybe it’s time to jump.

(more…)

Despite ongoing legal proceedings designed to block NVIDIA from producing Nehalem-compatible chipsets, big green is allegedly working to do just that.

A report from the Chinese tech website HKEPC suggests that NVIDIA is preparing two IGP chipsets for Nehalem and Westmere CPUs. The first, codenamed MCP99, is said to be a single-chip solution employing DMI, which means it is being prepared for the upcoming crop of 45nm Lynnfield chips which we have previously discussed. MCP89, meanwhile, will offer a similar set of features in a lesser performance profile for the more budget-conscious amongst us.

The MCP85 is also mentioned as a SoC solution which condenses the GPU, northbridge and southbridge functionality down into a single chip to simplify boards and cut costs.

NVIDIA has predictably refused to comment on unconfirmed chipsets, but the wily NVIDIA chief exec Jen-Hsun Huang was cryptic and suggestive as usual.

“We’re not necessarily building chipsets for future Intel buses. We’ve not commented anything on that and so you are just going to have to wait to see what we come up with,” he said during a July 26 analyst call. “Our company is…pretty darn clever. There is a lot of ways to skin the cat.”

What the MSI N275GTX Twin Frozr OC loses in spelling accuracy, it makes up for in cool, quiet operation and performance.

The United States Patent and Trademark Office has rejected another batch of infringement claims leveled against NVIDIA in an ongoing dispute regarding memory and memory controllers.

The eight bounced claims brings the total to 49 after the USPTO opened infringement proceedings with 41 rejections in June. The growing volume of rejections is expected to significantly influence the conclusion of a parallel case addressing similar claims in the ITC.

Big Green’s spinners are lauding what is thus far a one hundred percent success rate in battling the litigation which stems from memory tech in the GeForce 9000 series.

“We are pleased that the USPTO decided to review the patentability of these two additional Rambus’ patents and continued to agree with NVIDIA’s challenge to these eight claims,” said David Shannon, NVIDIA executive vice president and general counsel. “The USPTO has now initially rejected all of the patent claims asserted by Rambus against NVIDIA in the ITC.”

It is expected that RAMBUS will queue for another slap on the nose when the case progresses to the appeals stage around mid-August.

Enthusiast memory manufacturer Mushkin has recently announced that it plans to launch NVIDIA and ATI GPUs after a relatively unsuccessful bid in January 2008.

The new stable of products will debut as NVIDIA parts billed under the ultimateFX branding. The following GPUs are on deck, while a roster of ATI cards will follow later in the year:

Mushkin ultimateFX GeForce GTX260
Mushkin ultimateFX GeForce GTX260 (1792MB)
Mushkin ultimateFX GeForce GTX260 OC
Mushkin ultimateFX GeForce GTX260 OC2
Mushkin ultimateFX GeForce GTX275
Mushkin ultimateFX GeForce GTX275 OC
Mushkin ultimateFX GeForce GTX275 OC2
Mushkin ultimateFX GeForce GTX285
Mushkin ultimateFX GeForce GTX285 (2048MB)

Products in the ultimateFX line will arrive in a wooden box designed to foster an upscale image.

1. Cut a hole in the box.
2. Put a card in that box.
3. Make ‘em open the box, and that’s the way you do it!

NVIDIA revises the GTX 295 to use a single PCB now. Guru3D investigates.

The ASUS EN9400GT video card is aimed at the HTPC market because that’s about all it’s good for.