by Anton Shilov
10/30/2004 | 05:02 AM
ATI Technologies, the world’s top supplier of discrete graphics processing units, said it would adopt technologies that its arch-rival NVIDIA Corp. has been promoting for a while in future. While the Markham, Ontario-based graphics chip designer did not provide any concrete plans, typically graphics firms discuss details of their next-generation products, not hypothetic long-term plans.
Pixel Shaders 3.0 – When Needed
“ATI will support Pixel Shaders 3.0 when it becomes readily available in games and applications. Currently, this feature is not readily used by the developer community and today’s top titles still largely rely on Pixel Shaders 2.0. As this trend starts to change ATI will incorporate PS3.0 into its feature set,” said John Carvill, who is public relations manager on Integrated and Mobile Products for ATI, in an interview to an Asian web-site OC WorkBench.
ATI Technologies has been consistently downplaying Shader Model 3.0 supported by NVIDIA Corp.’s GeForce 6-series GPUs since the release of the GeForce 6800 product in April, 2004. An internal document allegedly made by ATI stated: “Steer people away from flow control in ps3.0 because we expect it to hurt badly. [Also it’s the main extra feature on NV40 vs. R420 so let’s discourage people from using it until R5xx shows up with decent performance...”.
Makers of graphics processors usually present new graphics architectures or seriously revamped architectures once a year. Some sources expect ATI’s next-generation R520 graphics chip to emerge in the first half of 2005. Before that ATI will offer its code-named R480 product, a remake of the current RADEON X800-series.
Multi-GPU Technology – Probing
The representative for ATI also made some comments in regards multi-GPU technology, a capability that allows two visual processing units to work in parallel and achieve higher performance in resource-demanding 3D environments than single VPU configuration. Currently multi-GPU ability is available on specially designed platforms by Alienware, a maker of high-end gaming desktops, and also from NVIDIA Corp., who supplies graphics processors that are natively able to operate in pairs.
“Our [graphics] cards can already support dual-GPU configurations in such platforms as Alienware’s ALX. We have a strategy in place for dual-GPU on the chipset side but it would be premature to discuss at this time,” Mr. Carvill indicated.
NVIDIA’s SLI is a technology that enables two NVIDIA-based graphics cards to operate in a single workstation or PC delivering higher graphics horsepower. A special mainboard with two PCI Express x16 is required for such configuration. According to NVIDIA’s estimates, typical performance advantage dual NVIDIA GeForce 6800 Ultra bring is about 75% - 80% when operating on a system running NVIDIA nForce4 SLI chipset that boasts special enhancements for Multi-GPU technology. NVIDIA’s approach requires special circuitry to be incorporated into GPUs and, for extra speed gain, into core-logic. Alienware’s Video Array technology does not require any special logic to be incorporated into graphics or system chips.
ATI’s potential desktop multi-GPU approach is unclear, however, the company has some logic in its VPUs to allow them to work in parallel, as the firm supplies its graphics processing units to Evans & Sutherland for high-end graphics systems with up to 256 chips.
ATI can make chipsets for both AMD and Intel microprocessors. NVIDIA Corp. does not have license to make core-logic products for Intel Corp.’s chips.