X-bit labs

Nvidia GeForce 8800 Specs Posted by Web-Site

by Anton Shilov

[ 09/19/2006 | 07:56 AM ]

A web-site from <%BANNER[article_nw]%>Taiwan claims that it has received specifications of Nvidia’s upcoming code-named G80 graphics processor, which, according to the publication, will be available later during the year. The information contradicts the data previously published by X-bit labs.

According to VR-Zone web-site, the new GeForce 8800 will feature unified shader architecture and will consist of 700 million transistors. The publication notes that the part will have two times higher pixel shader and twelve times higher vertex shader performance compared to the existing “G71”.

“Twelve times higher vertex shader performance” would mean that the part features either 96 unified shader pipelines at about 650MHz (the clock-speed of the GeForce 7900 GTX and the chip has 8 vertex shader processors) or 48 unified shader pipelines at 1300MHz. “Two times higher pixel shader performance” would mean that the part either has 48 unified shader processors at 650MHz (since GeForce 7900 GTX has 24 pixel shader processors) or 24 unified shader processors at 1300MHz. If the part had 96 unified shader processors at 650MHz, it would have four times higher peak speed when calculating pixel shaders compared to the GeForce 7900 GTX.

The web-site further suggests that the new code-named G80 offering from Nvidia Corp. will have 384-bit or 320-bit memory interface, will support new full-scene antialiasing mode called VCAA and so on.

The whole Nvidia G80 specifications list by VR-Zone looks as follows:

• Unified Shader Architecture;
• Support FP16 HDR+MSAA;
• Support GDDR4 memories;
• Close to 700M transistors (G71 - 278M / G70 - 302M);
• New AA mode: VCAA;
• Core clock scalable up to 1.5GHz;
• Shader peformance: 2x Pixel/12x Vertex over G71;
• 8 TCPs & 128 stream processors;
• Much more efficient than traditional architecture;
• 384-bit memory interface (256-bit+128-bit);
• 768MB memory size (512MB+256MB)

The publication further claims that there would be two models at launch, which is claimed to coincide with the release of Intel’s quad-core Intel Core 2 Extreme microprocessor in November:

• GeForce 8800GTX: 7 TCPs chip, 384-bit memory interface, hybrid water/fan cooler, water cooling for overclocking. $649;
• GeForce 8800GT: 6 TCPs chip, 320-bit memory interface, fan cooler. US$449-499;

The web-site did not elaborate why the top model based on the chip that has “8 TCPs & 128 stream processors” only sports “7 TCPs”. It is also unknown what “TCP” abbreviate stands for is. The term “stream processor” is also not explained.

Earlier it was reported that Nvidia’s first DirectX 10 chip will not have unified shader architecture and will incorporate 48 pixel shader processors and an unknown number of vertex shader/geometry shader processors. The Inquirer web-site, however, reported that the G80 has 32 pixel and 16 vertex and geometry shader processors.

• VR-Zone: G80 De-Mystified; GeForce 8800GTX/GT.