Pixel filling power
Fill rate and memory bandwidth aren't the be all and end all of graphics performance, but they're a good place to start. Here's a look at how the S4 Pro's theoretical fill rate and memory bandwidth peaks compare with the competition:
| 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) | |
| Radeon 9600 SE | 325 | 4 | 1300 | 1 | 1300 | 200 | 64 | 3.2 |
| Radeon 9550 | 250 | 4 | 1000 | 1 | 1000 | 200 | 128 | 6.4 |
| GeForce FX 5200 | 250 | 4 | 1000 | 1 | 1000 | 250 | 128 | 8.0 |
| DeltaChrome S4 Pro | 300 | 4 | 1200 | 1 | 1200 | 300 | 128 | 9.6 |
| GeForce FX 5200 Ultra | 325 | 4 | 1300 | 1 | 1300 | 325 | 128 | 10.4 |
Wedged between the GeForce FX 5200 and 5200 Ultra, the S4 Pro certainly has the fill rate and memory bandwidth to compete with other budget graphics parts. However, these are just theoretical peaks. What matters is how much of their fill rate potential these cards can realize.


Despite having a lower theoretical peak multi-texturing fill rate than the GeForce FX 5200 Ultra, the DeltaChrome S4 Pro leads the pack in 3DMark03's multi-texturing test. Unfortunately, the S4 Pro's single-texturing performance isn't as impressive, but it is faster than the Radeon 9550.
3DMark03 also has synthetic vertex and pixel shader tests that we can use to evaluate the relative strength of each card's shaders.

![]()
The S4 Pro's vertex shader performance is a little off the pace, but that's not the worst of it. The card's half as fast as the Radeon 9550 in 3DMark03's Pixel Shader 2.0 test.
| Friday night topic: The trouble with Best Buy | 129 |