nVidia GeForce4 MX 440 in the test: Inno3D, Gainward, Leadtek and Prolink in comparison
- 1 Foreword
- Overview of maps
- 3 The chip in detail
- 4 Test system
- 17 FSAA performance
- 19 Overclocking
- 21 Summary
Since 3DMark 2001, like 3DMark 2000, is a purely synthetic benchmark, it also says relatively little about the expected game performance However, it is a good starting point for future games, as it is based entirely on DirectX 8, in particular the Geforce3 as a DirectX 8 card clearly stands out from the rest of the test field, among other things because of the Nature test, whose points match the other cards completely withheld. That's why we ever give herethe values with and without consideration of the nature scene.
Since the GeForce4MX, contrary to its misleading name, does not have any pixel and vertex shaders that the 3DMark2001 SE makes excessive use of, is for you below 16Bit compared to the fill rate superiority of its predecessor GeForce2 Ti, no herb.
As before, however, its bandwidth-saving measures pay off under 32Bit color depth and it leaves the GeForce2 Ti with a 7-21% advantage even at the lowest resolution standing in the rain.
The Kyro II unfortunately only ranks under 'also ran ...', although in 1600 it comes close to the GeForce2 Ti without both w Deliver really usable results.
On the next page: Vulpine GL v1.1p