Graphics Wars Heat Up As Nvidia, AMD Launch Cards

Nvidia and Advanced Micro Devices released new graphics processors Monday as the two chip makers continued to compete on discrete graphics products while emerging as allies of a sort in promoting the possibilities of GPU computing in the face of CPU giant Intel's uncertain steps in that direction.

Nvidia's new GeForce GTX 200 series includes the GeForce GTX 280, available in quantity starting Tuesday for a suggested manufacturers' price of $649, and the GeForce GTX 260, available June 26 with a price tag of $399. The GTX 280, with 240 processors and a full gigabyte of frame buffer memory, is clearly Santa Clara, Calif.-based Nvidia's latest edge-pushing high-end consumer card, while the 192-processor, 896MB GTX 260 has humbler specs, but not by much.

Up the road in Sunnyvale, Calif., AMD's latest pair of consumer cards from its ATI graphics division target a lower portion of the discrete market. The new ATI Radeon HD 4800 series will deliver a teraflop of graphics performance, according to Rick Bergman, GM of AMD's Graphics Products Group.

And the Radeon HD 4850, set for release June 25, will do it for less than $200, Bergman said. The second card announced Monday is the Radeon HD 4870. It's a bit more powerful and a bit more pricey at around $300 than the 4850, and isn't scheduled for availability until July 8.

id
unit-1659132512259
type
Sponsored post

"We've adopted a new model for this product and going forward," Bergman said, outlining what he described as an important shift in AMD's strategy for discrete graphics. "It's targeting the $200 to $300 part of the market because we think that's the meat of the market."

Of course, there will be a third card in the 4800 series aimed at the $500-range "ultra enthusiast" segment. That product is still going by its code name at AMD, R700, and should be ready sometime in August. Until the R700 gets an actual name and slot in enthusiast systems, AMD and Nvidia won't be competing head-to-head on price with their latest product lines.

Still, that seems to be largely immaterial in the hyper-competitive world of discrete graphics and each chip maker was touting the alleged technological superiority of its own products in the run-up to Monday's launch announcements. Even as AMD pointed to its leadership on DirectX 10.1 support, 55nm process technology and GDDR5 memory, Nvidia touted its tripling of threads in flight over its GeForce 8 and 9 series architectures.

Next: Tesla Vs. FireStream

One area where Nvidia and AMD seem to be taking each other on more squarely is the relatively rarified world of general purpose GPU (GP-GPU) computing on high-performance compute clusters. Graphics-only Nvidia, perhaps naturally, has been more vocal about the potential of GP-GPU computing in recent months, devoting an entire Technology Editor's Day last week to its Tesla specialty cards and CUDA language for parallel programming for GPU clusters in areas like financial analysis and medical imaging.

In addition to the consumer products, both graphics chip makers also had second-generation, teraflop-performing treats for HPC partners Monday. Nvidia unveiled its $1,699 Tesla 10P GPU, while AMD countered with its as-yet-unpriced FireStream 9250 card for GP-GPU computing on single-socket servers and workstations.

This space may still be somewhat esoteric, but with Nvidia pushing hard for CUDA -- not to mention its own Tesla products -- while AMD touts FireStream and an open-standards approach to parallel programming, it's certainly competitive. Yet much as it is on the consumer client side, the Nvidia-AMD competition for HPC customers is tempered by a mutual interest in confronting the leviathan looming over all things silicon.

That would be Intel, which has rather famously dismissed GP-GPU computing and discrete graphics leadership as necessary to its immediate business. Intel CTO Justin Rattner reiterated the message at last week's Research@Intel Day, telling a crowd of researchers, analysts and journalists at the chip maker's annual R&D extravaganza that Intel sees ray tracing as the future of visual computing, set to replace traditional raster-based graphics in the years to come.

Rattner also promised that this would all become clear when Intel presents a paper on its next-generation of graphics technology, codenamed Larrabee, at August's Siggraph conference. That would certainly be something, as Intel has been remarkably cryptic about Larrabee and its overall graphics ambitions even as client computing has entered an era where visual leadership is increasingly important from the enthusiast segment across the commercial space and down to the entry-level consumer.

Whatever Larrabee turns out to be, Nvidia and AMD can at least take heart that it won't arrive until the second half of 2009. That's a mighty long window for the two companies to continue playing the part of "frenemies" when it comes to jointly evangelizing discrete graphics writ large while fighting tooth and nail in the product trenches.