what is THE gl bench test?

[Alex]

k 3dmark2001 for d3d... what about gl?

vulpine...aquamark...or whatever theyre called


which ONE is the best/most accurate and, hopefully, most FREE!!!

thx!

 

[Rand]

Quake 3.

And IMHO, 3DMark is far from the most representative benchmark of real world gaming conditions.

 

[nemesismk2]

<< Quake 3.

And IMHO, 3DMark is far from the most representative benchmark of real world gaming conditions. >>



My vote goes with Quake 3 as well!

I find it really funny that there's still people out there that think 3dmark scores mean something, some people must be actually buying it just to benchmark their pc, how lame is that?

 

[vss1980]

Quake3 is a bad choice in some benchmarking respects as it doesn't take advantage of all the abilities of a the graphics cards available today. For example, it will offer no information about how much faster GF3 vertex/pixel shading is than an ATI 8500 shading. As a general real-life performance test Quake3 is still probably one of the better tools.

In that sense of testing new features, Vulpine's glmark is better as it does test more advanced stuff.
Another possibility is GLexcess. Its not amazing but does give information about the seperate tests it does to compare to others.

 

[nemesismk2]

<< Quake3 is a bad choice in some benchmarking respects as it doesn't take advantage of all the abilities of a the graphics cards available today. For example, it will offer no information about how much faster GF3 vertex/pixel shading is than an ATI 8500 shading. As a general real-life performance test Quake3 is still probably one of the better tools.

In that sense of testing new features, Vulpine's glmark is better as it does test more advanced stuff.
Another possibility is GLexcess. Its not amazing but does give information about the seperate tests it does to compare to others. >>



True, just using one benchmark is not the way to go but most people have Quake 3 even if it's just the demo and it's a standard benchmark used by almost every single review. Other benchmark's should obviously be used but not artificial benchmark's which are a waste of time IMO!