EliteBastards editorial: Can you trust a benchmark if you have no proof of what's going on? I think that it's true and we don't have enough proof about anything. We should all be wanting proof instead of speculation. I say wait for proof. nVidia insists that it is a rendering bug of intelligent culling and clipping. There are thousands of line of code and we all know that the newer drivers are still buggy. Let's wait and see what happens. If nVidia can produce scaling results when changing a variable in their intelligent culling and clipping, then we know it's legit. If the score is the same and the "hall of mirrors" effect is gone, we know it's legit. If it stays the same in the next big driver release, we know it's most likely a sham.