The V49 benchmark apparently fails to correctly load Tomb Raider's GeForce FX-optimized code path, so it's more a reflection of how the GeForce FX cards perform with default DirectX 9 code than anything else. Running the normal Tomb Raider game executable, without the benchmark mode enabled, loads the correct GeForce FX code path and promises better performance.
The way the words sounded to me is that the Tech Report reviewer repeated what (perhaps) NVIDIA told that reviewer. Are those statements of fact backed up by the reviewer's own investigations, or statements provided by NVIDIA?
Subsequent to Beyond3D making known the fact that TRAOD can be benchmarked, NVIDIA have released drivers that improves their cards' performance in the benchmarking of this game. Can anyone guess why?
Perhaps, the scenario is such that NVIDIA didn't know the game can be benchmarked with v49. Perhaps NVIDIA had no such "GeForce FX-optimized" (whatever that may mean... anyone knows?) drivers for the game at the time that Beyond3D debuted this game as a benchmark.
After the fact that Beyond3D made known the fact that the game can be benchmarked with the v49 patch, we have had NVIDIA drivers that demonstrably improves NVIDIA cards' peformance in the benchmarking of this game. This can be easily investigated by comparing the NVIDIA drivers used when
Beyond3D first presented the TRAOD benchmarks and every NVIDIA drivers after this.
To further investigate the Tech Report reviewer's words, FRAPS can be used to compare the performance of actual gameplay versus benchmarking of the game... it appears, from the Tech Report reviewer's words, that there is a difference between actual gameplay performance and benchmark performance
regardless of version of drivers used, all because benchmarking the game involves adding a "-benchmark=" command-line parameter that appears to change performance depending on whether we're talking about benchmarking or actual gameplay .... but take in mind the above paragraph -- to compare actual gameplay performance versus benchmarking performance according to what the Tech Report reviewer stated, using FRAPS while using the drivers before and after the fact that Beyond3D first presented the TRAOD benchmarks.
FYI, this is what the TRAOD PC programmer (from Core Design) told me regarding the v49 patch on a general basis :
But no matter what anybody says about the benchmark, the better the scores it reports then the better the game plays.
This directly contradicts what the Tech Report reviewer states, updated video card drivers excepted.
I am not satisfied with what the Tech Report reviewer stated about this game in his review. He should say if what he "reported" in his review about the benchmarking of this game is the result of undeniable fact based on his own investigation... or if it is directly what NVIDIA or Core/EIDOS told him. There's a huge difference between the two, and it is my impression that it is the latter, based on the way the words the Tech Report reviewer used. Which, of course, he shouldn't have used if this is the case.
I really wouldn't have any reason to post any of this if this Tech Report "review" was clearly labelled "
Considering a NVIDIA video card? Read this!". If this was the case, then the reviewer have a right to say what he did, which was to provide comment (the way he did) about whether TRAOD was "GeForce FX-optimized" or not... anyone intending to buy a NVIDIA card
and nothing else would love to read about any NVIDIA optimizations in any game. But this "review" is a shootout, with ATI cards present. Why would/should this Tech Report reviewer state what he did regarding the benchmarking of this game, without any apparent effort to prove the legitimacy of what he stated, nor provide any IQ investigations/comparisons among the various cards in a
shootout when he mentions "GeForce FX-optimized" ?
This is exactly the sort of thing that can be labelled "spin". I'm not questioning the Tech Report reviewer's ethics... I'm just wondering why he felt any need to state what he did given that he provided no backing up for his statements, especially in a "review" that is a shootout.
) why would they bring 5900SE? It just doesnt make sense especially as margins on slower NV35 are certanly worser than NV36, or at least should be by some common logic.