Is FRAPS considered a benchmarking utility?

1ManArmY

Golden Member
Mar 7, 2003
1,333
0
0
I was just curious because everyone is so hung up on their 3D Mark 03 scores while you can just pop in your most current up to date game using directX 9 ( Ravenshield ) and execute FRAPS and see what your FPS are.
 

waylman

Diamond Member
Apr 4, 2003
3,473
0
0
FRAPS is used only for your FPS....it does not benchmark your PC. To benchmark, you have to run a specific test such as 3dmark or a pre-designed game bench.
 

nemesismk2

Diamond Member
Sep 29, 2001
4,810
5
76
www.ultimatehardware.net
Originally posted by: 1ManArmY
I was just curious because everyone is so hung up on their 3D Mark 03 scores while you can just pop in your most current up to date game using directX 9 ( Ravenshield ) and execute FRAPS and see what your FPS are.

I agree fully with your comment and I also use fraps to find out the fps of the games I use.
 

Pete

Diamond Member
Oct 10, 1999
4,953
0
0
Yes, it is. Check out NVNews' reviews to see FRAPS being used instead of the all-too-common timedemos.

waylman, FRAPS lets you log your framerate to a file, which you can use to calculate average, min, and max fps, and make nice chronological graphs like Serious Sam allows (see Tech-Report's latest FX5200U review).
 

waylman

Diamond Member
Apr 4, 2003
3,473
0
0
I understand what you are saying. However, you would have to log your frames at the same time in a game as someone else. Otherwise, it's a useless bench.
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
I had fraps running once when playing Unreal Tournament, and I had the Unreal Tournament timedemo thing up as well, which shows on the fly average and current frame rates, and the 2 reported "current" frame rates were different.
Since the UT timedemo counter was built in to the engine, I was a bit wary of the fraps numbers, and have been ever since.
 
Jan 31, 2002
40,819
2
0
As long as you're both in the same general area, FRAPS is actually quite good as a general performance indicator. Eg "In the level XXX I got about YY fps on average, Z min, W max."

- M4H
 

1ManArmY

Golden Member
Mar 7, 2003
1,333
0
0
Thanks, I know the numbers will vary based on your location and the game your logging your FPS. I just thought it would be logical to use a real world benchmark as opposed to the synthetic Mark 03. I do believe it serves a purpose though but I'm concerned with my FPS during actual competition rather than viewing the pretty pictures in the 3D Mark 03
 

sandorski

No Lifer
Oct 10, 1999
70,798
6,355
126
I can get 300+ fps in Unreal 2 with fraps, if I stare at a wall. That's the biggest problem with using FRAPS as a benchmarker, you need the exact same movement/view on each system being compared.
 

titanmiller

Platinum Member
Jan 5, 2003
2,123
2
81
I suppose you could "wander" around a certain map for 10:00 (minuets) and find the average for that set time on a set map.

 

waylman

Diamond Member
Apr 4, 2003
3,473
0
0
Originally posted by: sandorski
I can get 300+ fps in Unreal 2 with fraps, if I stare at a wall. That's the biggest problem with using FRAPS as a benchmarker, you need the exact same movement/view on each system being compared.

Precisely.
 

1ManArmY

Golden Member
Mar 7, 2003
1,333
0
0
Lonyo
I had fraps running once when playing Unreal Tournament, and I had the Unreal Tournament timedemo thing up as well, which shows on the fly average and current frame rates, and the 2 reported "current" frame rates were different.

What was the difference of margin? Was it significant?
 

LesPaul

Senior member
Dec 4, 2002
248
0
76
You could record a demo in UT (not sure if this game lets you.. never tried), then run that as your own time-demo with fraps going to record the FPS... Then nividia or ati can't do any 'optimizations' for it in the drivers (other than entire game optimizations - which are good), because they don't know how your demo is setup.

This might be a good way for reviewers to start benching hardware, to avoid all the possible "cheats" and accusations going on so much now. You wouldn't really be able to compare these benchmarks from site to site, because each demo would be different, but you could compare benchmarks from the same site.. ie. Compare AT's custom demo benchmarks of 9800pro vs. 5900u or whatever.. You could get a good idea of how each card performs against other cards, you just wouldn't be able to cross compare results with other sites.

What do you guys think?