Articles: Graphics
 

Bookmark and Share

(8) 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 ]

Far Cry

There are probably even more pixel shaders in Far Cry than in Halo; moreover, these shaders are very sophisticated, and the game itself boasts high-resolution and beautiful textures and not-very-simple geometry. All these things taken together can load any graphics card to the full. We tested the RADEON X800 PRO in the Shader Model 2.0b mode, while the graphics cards with NVIDIA’s GPUs were using Shader Model 3.0. Thus, we made use of the innovations implemented in both competing architectures (R420 and NV40).

The ASUS V9999 Gamer Edition does well in this test, although it is slower than the RADEON X800 PRO which is capable of fast processing of complex shaders. Note that the difference between the ASUS and the GeForce 6800 is again negligibly small, since it is the GPU clock rate that mostly determines the results here, as the speed of pixel shaders depends on the graphics core frequency.

The influence of fast memory on the performance becomes stronger in the eye candy mode. The RADEON X800 feels at ease in such modes, and almost overtakes the 16-pipelined GeForce 6800 GT. The V9999 Gamer Edition takes the third place. Our overclocking of the ASUS card brings the biggest reward in 1024x768 (in higher resolutions, it is the memory subsystem speed that determines the result).

It’s practically the same on the Research level as on the Pier one – thanks to its ability to effectively process complex computations-rich pixel shaders, the RADEON X800 is 8-12fps faster than the ASUS V9999 Gamer Edition. Overclocking helps the ASUS card to make the gap as small as 2-6fps.

The gap between the two rivals (ASUS V9999 Gamer Edition and RADEON X800 PRO) is diminishing to become the smallest in the highest resolution. Since the graphics memory speed is most important for this operational mode, the results of the overclocked ASUS look natural enough.

 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 ]

Discussion

Comments currently: 8
Discussion started: 10/20/04 01:58:34 AM
Latest comment: 11/27/05 04:48:00 AM

View comments

Add your Comment