Crysis Benchmark Tool - Bug

Update on the Crysis Benchmark Tool, Bug:

The Crysis Benchmark Tool has a bug with the settings, found by johnksss, the resolution and graphics quality will be that of the one set in-game no matter what you set in the benchmark tool's window, to obtain realistic framerates, please set both the game's settings and the benchmark tool's as desired (might not work because of that the benchmark tool changes the resolutions dynamically and this might conflict with the correct result), use of the included .bat file benchmark that comes with Crysis is discouraged as well, unless pictures are taken to certify a result, because of the .bat file being flawed (see Update #2 below).

[Link To johnksss's Original Post On The Bug]

If you still want to use the Crysis Benchmark Tool, it is recommended that you run every resolution's test individually after setting the settings you want to use with the benchmark tool in the game's configuration too.

I wondered if this thread could be stickied so that people that use the Crysis Benchmark Tool to measure their graphics cards' performance and / or compare graphics cards are informed about the bug that this benchmark contains, since this bug can be exploited to get unreal scores, also, some of the results of this benchmark are already misleading people.

Updates:

( Nullified) October 01, 2008, 11:47 A.M Mountain Time:

Due to a misunderstanding on my half while talking to johnksss, a few minutes ago, I posted wrong information about that the .bat Crysis Benchmark was flawed too, the truth is that it is in fact not flawed and can be used as a reliable benchmarking tool.

October 01, 2008, 11:59 A.M Mountain Time:

Another report came from johnksss, the .bat is in fact flawed, but not in the way we thought it was, the MS-DOS window that the benchmark opens does not display what resolution the game is really running at, and this might conflict with the real scores of the benchmark, on the other hand, the benchmark's logfile can be modified to display any results anyone wants, another lethal flaw here, people, so still, do not really trust any unreal scores that people post about the Crysis Benchmarks, because there is no way to know that they are real for sure.

 

Update (Nullified):

October 01, 2008, 11:59 P.M Mountain Time:

Due to a misunderstanding on my half while talking to johnksss, a few minutes ago, I posted wrong information about that the .bat Crysis Benchmark was flawed too, the truth is that it is in fact not flawed and can be used as a reliable benchmarking tool.

 

Update #2:

October 01, 2008, 11:59 A.M Mountain Time:

Another report came from johnksss, the .bat is in fact flawed, but not in the way we thought it was, the MS-DOS window that the benchmark opens does not display what resolution the game is really running at, and this might conflict with the real scores of the benchmark, on the other hand, the benchmark's logfile can be modified to display any results anyone wants, another lethal flaw here, people, so still, do not really trust any unreal scores that people post about the Crysis Benchmarks, because there is no way to know that they are real for sure.