3DMark has always been a popular platform for testing video card graphic performance even though static benchmarks have gotten a lot of criticism about testing true performance in video cards. I myself have always believed 3DMark is a useful test but not the final way to tell true performance. The graphs below show that the new drivers did help the card to achieve a higher score. Not overclocked the card gained 155 points that's an increase of 1.29 percent.  On the other hand the video card did not fair as well while it was overclocked gaining only 83 points giving it an .63 percent increase in performance. All this tells me is that it looks like my overclock lost its effectiveness in this test.

I was blown away by the difference the new drivers made on 3DMark 2003. I was expecting to get the same results as I got in 3DMark 2001 but instead I gained 93 points giving me an increase of 6.4 percent and my overclocked performance gained 118 points giving it an increase of 7.03 percent. So far these tests are proving that 3DMark does benefit from the driver upgrade.

After getting the results from this benchmark I was thinking that all the talk might be right. The non overclocked test lost 19 points giving the video card a .9 percent loss in performance, and the overclocked test lost 8 points giving it a .3 percent loss in performance. I am not usually surprised with a 1 percent difference in scores when dealing with benchmarks so in this case it doesn't seem as if the drivers did anything at all. So how do you explain the performance increases in 3DMark 2003 when there's no affect here. Lets look at some more benchmarks and see what they say.

GLMark is basically a benchmark that works a lot like 3DMark but is for OpenGL. If you take a look at the graphs below you will notice that the drivers had no affect on this benchmark as well. Lets get away from the static benchmarks and move on to real time benchmarks.