Nvidia Fermi GTX 480M Coming in June

 
 
By Nathan Eddy  |  Posted 2010-05-04 Email Print this article Print
 
 
 
 
 
 
 

An online notebook vendor accidentally offers a mobile version of Nvidia's Fermi architecture based GTX 480 GPU, the 480M. Although the offer was later removed, the listing suggests a mobile version of the GPU will hit markets sometime in June.

An online notebook maker posted the configuration for its Eurocom Leopard notebook being released in June, listing as part of the components Nvidia's GTX 480M chip, based on the company's new Fermi GPU architecture.

While the site later removed the listing, the leak spread across Web-based news outlets, leading to further speculation that a mobile version of the Fermi GPUs are almost to market. Nvidia had previously released the Fermi architecture-based graphics cards GTX 480 and GTX 470, gaming-centric GPUs designed with tessellation performance in mind. Tessellation techniques are often used to manage data sets of polygons and divide them into suitable structures for rendering.

The company's PolyMorph Engine, a scalable geometry processing engine built from the ground up for DirectX 11 tessellation, also includes high-speed 32x anti-aliasing smooth edges for improved visual quality and is capable of rendering more than 746M pixels per second at full HD 1080p for a 5,760-by-1,080 gaming experience. Nvidia also offers software that automatically converts more than 400 games to stereoscopic 3D, part of the company's 3D Vision Surround technology that expands the gaming real estate across three monitors in full stereoscopic 3D.


Nvidia, currently in second place behind rival (and business partner) Intel, jumped its market share from 31.5 percent in the fourth quarter, according to Jon Peddie Research. Nvidia has aggressively been pushing its GPU technology into more mainstream computing environments, particularly in such areas as HPC (high-performance computing). The Fermi GPU architecture, which incorporates more than 3 billion transistors and 512 CUDA cores, is part of that campaign.

While Nvidia and competitors Intel and AMD continue to unveil ever more powerful chipsets, Nvidia's chief scientist and vice president, Bill Daly, warned Moore's Law, which posits the number of transistors on a chip would double every 18 months, thus doubling the performance of the CPU in that time, may run out. In a column on Forbes.com posted last week, Daly argued that the energy needs for the CPUs Intel and AMD are pushing out there are creating an environment where Moore's Law can no longer continue. Daly argued Nvidia's GPU technology is the answer to the problem.

The news comes as the SIA (Semiconductor Industry Association) announced chip sales rose to $23.1 billion in March, increasing 4.6 percent from February's sales. The SIA report found global chip sales increased from $43.7 billion in the first quarter of 2009 to $69.2 billion during the first three months of 2010. The good news echoes a report issued by research firm IDC in April, which found chip unit shipments showed a seasonal decline of 5.6 percent over the fourth quarter of 2009, but a 39 percent jump over the same period last year. Rising demand, particularly in developing markets, is helping the rebound in the semiconductor industry, aided as well by a return to technology spending.

 
 
 
 
Nathan Eddy is Associate Editor, Midmarket, at eWEEK.com. Before joining eWEEK.com, Nate was a writer with ChannelWeb and he served as an editor at FierceMarkets. He is a graduate of the Medill School of Journalism at Northwestern University.
 
 
 
 
 
 
 

Submit a Comment

Loading Comments...

 
Manage your Newsletters: Login   Register My Newsletters























 
 
 
 
 
 
 
 
 
 
 
Rocket Fuel