I guess motherboards are only so exciting, because Tyan has decided to dip its toe into the graphics board market with a couple of ATI-based products. The Tachyon G9000 Pro is the first of these graphics cards to become available, and it sports ATI's new Radeon 9000 Pro chipset, a killer DirectX 8.1-compliant GPU that's poised to wipe the GeForce4 MX off the map completely.
We've given Tyan's freshman offering a thorough hazing on our test bench and compared it to ATI's own "Built by ATI" Radeon 9000 Pro solution. How does the Tachyon G9000 Pro stack up in features, performance, quality, and even retail bundles? Let's find out.
The Tachyon G9000 Pro features ATI's Radeon 9000 Pro graphics chip. Despite a few minor architectural changes, the Radeon 9000 Pro is very similar to ATI's Radeon 8500, and it retains the 8500's full DirectX 8.1 feature list, which is going to become increasingly important as this year's holiday titles start to require true pixel and vertex shaders to run with all the visual goodies turned on.
ATI's Radeon 9000 Pro is putting big pressure on NVIDIA's GeForce4 MX graphics chips, which lack hardware vertex and pixel shaders. Cards based on the Radeon 9000 Pro chipset have also come to market at very low price points, which should help aid the penetration of DirectX 8.1 features into price-sensitive mainstream markets where even the cost of a GeForce4 Ti 4200 is a little too steep.
If you want a detailed analysis of the Radeon 9000 Pro, you can read TR's comprehensive review of that chip. This article will be an examination of Tyan's Radeon 9000 Pro implementation rather than an in-depth look at the chip itself. However, let's take a quick look at some fill rate and memory bandwidth numbers to set the stage.
|Core clock (MHz)||Pixel pipelines||Peak fill rate (Mpixels/s)||Texture units per pixel pipeline||Peak fill rate (Mtexels/s)||Memory clock (MHz)||Memory bus width (bits)||Peak memory bandwidth (GB/s)|
|GeForce4 MX 440||270||2||540||2||1080||400||128||6.4|
|GeForce4 Ti 4200 128MB||250||4||1000||2||2000||444||128||7.1|
|GeForce4 Ti 4200 64MB||250||4||1000||2||2000||500||128||8.0|
|SiS Xabre 400||250||4||1000||2||2000||500||128||8.0|
|Radeon 9000 Pro||275||4||1100||1||1100||550||128||8.8|
|GeForce4 MX 460||300||2||600||2||1200||550||128||8.8|
|Radeon 8500 128MB||275||4||1100||2||2200||550||128||8.8|
Though the Radeon 9000 Pro only lays down one texture per clock, it's capable of "looping back" within the rendering pipeline to lay down multiple textures in a single rendering pass. So, while the Radeon 9000 Pro's texturing fill rate looks low on paper, we're not taking into account the chip's "loop back" functionality.
In terms of memory bandwidth, the Radeon 9000 Pro's DDR SDRAM runs at 275MHz on at 128-bit memory bus. That gives it as much memory bandwidth as a classic Radeon 8500 and NVIDIA's GeForce4 MX 460.