RJARRRPCGP Posted January 18, 2006 Share Posted January 18, 2006 Windows 98 SE for most of the time seems to give me better benchmarking results, especially with Nintendo 64 emulation, but 3D Mark 2001 SE always gave me a lower final score than when running under Windows 2000 and when running under Windows XP. I dunno why. Link to comment Share on other sites More sharing options...
Lunac Posted January 18, 2006 Share Posted January 18, 2006 (edited) It could be device driver version, driver configuration, or 3DMark2001SE settings. Some driver versions give better performance under one OS than they would under another OS. Also, are you even running same version of drivers under both OS?Depending on the brand of your video card, you should research driver versions and their performance for a particular OS.For instance, I got an increase of 1201 3Dmarks (from 9840 to 11041) by upgrading my video card drivers under Win98SE. (nvidia ver. 32.82 to 40.72). A 12.2% increase in performance. I was using 3DMark2001 SE.Also, check that your driver/video card settings are identical, and do the same for 3DMark2001SE settings. Edited January 18, 2006 by Lunac Link to comment Share on other sites More sharing options...
RJARRRPCGP Posted January 18, 2006 Author Share Posted January 18, 2006 (edited) It could be device driver version, driver configuration, or 3DMark2001SE settings. Some driver versions give better performance under one OS than they would under another OS. Also, are you even running same version of drivers under both OS?Depending on the brand of your video card, you should research driver versions and their performance for a particular OS.For instance, I got an increase of 1201 3Dmarks (from 9840 to 11041) by upgrading my video card drivers under Win98SE. (nvidia ver. 32.82 to 40.72). A 12.2% increase in performance. I was using 3DMark2001 SE.Also, check that your driver/video card settings are identical, and do the same for 3DMark2001SE settings.This was when I decided to test 3D Mark 2001 SE based on the driver version number that had the best results.44.03 is the version number of the Detonator drivers that gave me the best 3D Mark 2001 SE results with my eVGA GeForce 4 Ti 4200 64 MB video card. Thus I decided to try the Windows 9x version of 44.03 under Windows 98 SE then that's when I noticed this phenomenon. Edited January 18, 2006 by RJARRRPCGP Link to comment Share on other sites More sharing options...
miko Posted January 19, 2006 Share Posted January 19, 2006 (edited) the 44.03s are somewhat famously hacked to give artifically good result in benchmarks, it could have something to do with that.some linkage - here and here45.23 (WHQL in XP) are nice on GF 3s & 4s (if i remember rightly) you should also possibly try the 45.32 betas(after much testing i ran one of these two on my GeForce 4 Ti 4200, just can remember which ). Edited January 19, 2006 by miko Link to comment Share on other sites More sharing options...
RJARRRPCGP Posted January 19, 2006 Author Share Posted January 19, 2006 (edited) the 44.03s are somewhat famously hacked to give artifically good result in benchmarks, it could have something to do with that.some linkage - here and here45.23 (WHQL in XP) are nice on GF 3s & 4s (if i remember rightly) you should also possibly try the 45.32 betas(after much testing i ran one of these two on my GeForce 4 Ti 4200, just can remember which ).I don't like 56.72, because it causes 3D Mark 2001 SE to start stuttering!! Edited January 19, 2006 by RJARRRPCGP Link to comment Share on other sites More sharing options...
miko Posted January 19, 2006 Share Posted January 19, 2006 i didn't recommend 56.72 Link to comment Share on other sites More sharing options...
Lunac Posted January 19, 2006 Share Posted January 19, 2006 (edited) I've owned many video cards, including three (3) nvidia boards. GeForce4 MX 420, GeForce3 Ti500, and now GeForce4 Ti4600. In that order. In tons of extensive comparison driver tests done in the last four years or so this is what I concluded: 32.82 and 40.72 are pretty much best nvidia drivers for Win9x. In terms of stability and performance 32.82 is overall best for GF3 series while 40.72 is best for GF4. (Compatibility with latest games and nvidia hardware is another thing). Are you sure driver/video settings are identical under both OS's? Also, miko is right, 44.03 are "influenced" drivers.I must admit, I never really tested any nvidia boards under XP. Probably because I haven't used XP in 4+ years now . I have ran all of these boards under Windows 2000, but never really bothered to do any extensive benchmarks. I mean, why? Call me old fashioned but running 3D benchmarks under any NT OS for the sake of gauging system performance seem kind of inane. Edited January 19, 2006 by Lunac Link to comment Share on other sites More sharing options...
RJARRRPCGP Posted January 19, 2006 Author Share Posted January 19, 2006 i didn't recommend 56.72 Sorry about that. I was just saying that I decided to test 56.72 a while ago, because I switched back to a Nvidia from an ATI. Then I noticed that. Wonder if the 5xxx versions require a GeForce FX for good performance? Link to comment Share on other sites More sharing options...
Lunac Posted January 19, 2006 Share Posted January 19, 2006 What results did you get with 40.72? Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now