NVIDIA has responded angrily to claims that it employed underhand tactics to inflate benchmark scores in 3DMark03. Their response is posted in full and unedited:
Accusations of cheating, tweaking and unfair optimisation have been flying around since 3DMark03 was released. NVIDIA feels that at least some of the furore stems from the fact that the company is no longer a member of FutureMark's benchmarking programme.
Andrew Humber from NVIDIA had this to say on the subject: "Since NVIDIA is not part of the FutureMark beta program (a program which costs of hundreds of thousands of dollars to participate in) we do not get a chance to work with FutureMark on writing the shaders like we would with a real applications developer."
NVIDIA reckons that the best way to judge the performance of the latest GeForce FX cards is through playing the games themselves, rather than "performance factors that are not representative of actual game play", as online hardware independent [H]ard¦OCP pointed out recently.
"Our relative performance on games like Unreal Tournament 2003 and Doom 3 shows that the GeForce FX 5900 is by far the fastest graphics on the market today and we actively encourage benchmarkers to spend more time testing actual games and actual game performance as opposed to synthetic benchmarks. As more DirectX 9.0 games become available, this will become more commonplace and synthetic benchmarks will become less important. At the end of the day, what is most important to the enthusiasts and what do they pay their hard earned money for - cards that look great and run fast on their favourite games or an impressive 3DMark score? I can't help but think, the more the industry and our competitors focus on a non-real world tests such as this, the more the gamers will lose out in my opinion."