?!?Biased Benchmarks?!?

[san<>tran]

Ok, I got an issue of MaximumPC and they reviewed the ATI HD3870 and they're benchmarks looked something like this. For World in Conflict, running on high and DX10 specs, they had the HD3870 at 23 Frames Per Second, and the 8800GT at 20 FPS. Strange enough that the HD3870 would outperform the 8800GT huh? Well I had to investigate. I searched and found a benchmark on Legion Hardware.com and I found that they did the same test, specs and everything and found that, the 8800GT with 29 FPS, and the HD3870 had something that MaximumPC's results couldn't be further from the truth, Legion Hardware.com found that the ATI HD3850 had 12 FPS. Now I ask why this is, why that some benchmarks can be misleading that how could some be so off it shows that the HD3870 outperforms the 8800GT at its own game? Is there Bias among the results of MaximumPC?

[wormy]

well it depends on the hardware that is used....cpu memory mobo...so on and so forth

[suchuwato]

Every machine and card will produce different results... using the same machine as a comparison is valuable, not comparing different sources.

[GenG]

Other comparisons

 

http://www.extremetech.com/article2/0,1697,2219908,00.asp

 

http://www.techspot.com/review/76-asus-rad...3870/page7.html

 

The review on here didn't have WIC, but from these it is obvious that the 3870 isn't TOO far off. I would say maybe a flawed test rather then a biased opinion, but i even doubt that.

Edited December 7, 2007 by GenG

[Bosco]

There is lots of factors that can make scores different. Hardware, Drivers, Games, Game settings Patches and so on the list goes on and on.

 

I am not one to bash other people for their benchmarks because I am not there, but this stuff is known to happen we even have it happen to use here at OCC once in a while with benchmarks. Gotta love the ones that make you go what the heck? Stranger things have happened lol

[san<>tran]

I checked, the computer specs for both benchmarks differ from little to none.

[Waco]

I checked, the computer specs for both benchmarks differ from little to none.

This says nothing about the driver versions use and/or patch version of the game.

[san<>tran]

Still with the review, a simple patch couldn't sway performance by over 50%.

[spathalogic]

Bad patch + Bad drivers = Bad performance

 

Good patch + Good drivers = Good performance

 

I wouldn't be all too surprised if WIC has some issue with a certain driver version, or if a certain driver version has an issue with WIC. The 3870 HD and WIC are both fairly new.

Edited December 7, 2007 by Instag0

[Bosco]

Still with the review, a simple patch couldn't sway performance by over 50%.

Don't bet on that

[Waco]

Still with the review, a simple patch couldn't sway performance by over 50%.

You apparently haven't looked into this much.

 

Drivers + patching can improve performance well over 100% sometimes when there are bugs in the drivers and/or programs.

 

I wouldn't be all too surprised if WIC has some issue with a certain driver version, or if a certain driver version has an issue with WIC. The 3870 HD and WIC are both fairly new.

The newest patch and drivers for ATI cards and WIC are reportedly MUCH MUCH faster.