Call of Juarez is going to be one of the first PC games which will make use of DirectX 10, and now that ATI's first desktop DX10 part is out (the HD 2900 XT) Legit Reviews have decided to throw it into the ring with NVIDIA's equivilent offering in the 8800 GTX to see which one deals with DX10 content better.
Using a pre-release cutdown build of this game which was provided purely for benchmarking purposes, it's funny that although AMD/ATI were the ones to hand it out in conjunction with their launch of the HD 2900 XT, the results clearly indicate NVIDIA's card to reign supreme by a good margin.
The irony doesn't stop there though, it was also mentioned during the article that the particular benchmark build LegitReviews used wasn't working properly in conjunction with NVIDIA's card, a newer version of the benchmark was made to work equally for both cards yet as much as they tried they were strangely unable to source this build from either parties.
I'll let the results do the rest of the talking.
By now, we've all been beaten to death with all the DirectX 9 benchmarking on DirectX 10 capable cards because no DirectX 10 benchmark game titles were available to be benchmarked. I've been saying for months that the NVIDIA and ATI battle doesn't mean anything till we see some performance numbers on the DirectX 10. Before the launch of the ATI Radeon HD 2000 Series AMD and ATI seeded the press with benchmark copies of the upcoming DirectX 10 game title Call of Juarez. Finally, we had a benchmark that we could use to run DirectX 10 material and it was from a real game that consumers were able to purchase.
Once we fired up the benchmark and started running it on the NVIDIA GeForce 8800 GTX we noticed the application wouldn't run when any level of Anti-Aliasing was enabled. I contacted both ATI and NVIDIA letting them know that a benchmark was given to the media that didn't run correctly on one of the brands. It's here where things started to get a bit interesting!