4in1 3D Perf Test

Discussion in 'Windows Desktop Systems' started by Taurus, Jun 27, 2002.

  1. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    some people with via chipsets claim significant 3d performance gains/losses with different 4in1 drivers... something i've never experienced. i formatted fresh today and decided to run a little experiment.

    the tests were run with my system (below) with detonator 29.42 installed. as you can see, there was never more than a half a percent difference from one driver to the next.

    if you have a different chipset and/or video card, feel free to do the same and post your results. }:>
     
  2. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    yeah, via's site didn't have the 4.37's, but didn't bother looking elsewhere..

    so, i'll do those later

    edit: found 4.37 and 4.39beta... now off to bed
     
  3. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    2nd time...

    did it again today with 4.37 and 4.39beta included...

    again, no real difference for me... and the different results in the two tests don't really add up... meaning the <1% difference isn't related to the 4in1's... maybe it's heat, voltage fluctuations, or aliens.
     
  4. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    oop...
     
  5. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    i did this morning before the 2nd run... but not in between each driver install, which i've been thinking is the way to go. is there a utility that will wipe any installed 4in1's and totally revert back to the 4.33's that came with xp? doing the uninstall option doesn't really fair well with me... it gets rid of the detonators too, somehow.
     
  6. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    so doing a fresh format between each test would be the way to get true results... no leftover driver remnants from the previous version.

    i do install xp in about 25 minutes (hd to hd)... so it's not that out of the question. maybe later.

    i want to do the same with the detonator drivers, also... as the only series of tests i've seen were with a gf3 and the results could differ for my card. is removing the driver from add/remove programs and device manager enough or should i use that utility that wipes detonators?
     
  7. naheed

    naheed Guest

    so what version of these drivers allow for the best performance?
     
  8. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    agreed.

    so, based on the results from my sdram/geforce2 machine and catch23's ddr/geforce3, it's safe to say that any recent 4in1 driver (4.33 or later) will not affect 3d performance in any significant way.

    it would be nice to see someone with a kt333 chipset and, say, a radeon 8500 to do the same series of tests to see if we can get any contrasting numbers. and even though all we did was test direct3d performance, i don't think opengl tests would yield different results since i believe that only makes a difference after the agp slot--not with the chipset or agp slot, which is what these 4in1 drivers concern. maybe some harddrive or memory benchmarks might show us some real differences between the versions? eh.. i'll pass. }:>
     
  9. naheed

    naheed Guest

    so they are all pretty much the same then right?
     
  10. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    when it comes to 3d performance, it looks like it... but we didn't test ide speed, memory, and overall system compatibility within the different drivers... which they could all affect.

    like catch said... when in doubt, go with the latest... and if you notice a difference that you don't like, go back to the previous version.