Performance in Synthetic Tests
Some professional reviewers have noted that the fill rate tests of previous 3DMark versions have been somewhat bandwidth limited. Bandwidth is tightly tied to fill rate, since any game that does lots of fill, also
has to use a large amount of large texture maps, which in turn stresses the bandwidth.
The fill rate tests of 3DMark05 are a bit different than before, more theoretical, minimizing the influence of bandwidth and attempting to concentrate on measuring fill rate. The tests do not look as nice as the previous versions, but any nicer effect would require larger texture maps and thereby again move the bottleneck towards graphics memory bandwidth, Futuremark says.
- Single-Texturing. 64 screen size quads are single textured by a 2x2 map. This is basically the same test as in 3DMark03, except for the smaller texture to reduce bandwidth influence. This test uses the DirectX fixed function rendering path.
- Multi-Texturing. 8 screen size quads are mapped with 8 additively blended textures. The maximum amount of fixed function texture layers rendered in a single path is 8, otherwise there could have been more textures per object. This is also basically the same test as in 3DMark03, except with only a 2x2 texture. DirectX 9 hardware should support in addition to PS 2.0 also the capability to render 8 texture layers per pass. Therefore, no one really needs the functionality that 3DMark03 had, where quad texture cards for example rendered 16 objects with four texture layers each.