Low performance with GeforceFX

Discussion in 'Graphics Cards' started by Gizzy, Oct 19, 2003.

  1. Gizzy

    Gizzy Guest

    Hi guys,

    I just did a major upgrade on my PC. Previously I had a P4 1.7GHz with 512MB Ram and a Geforce 4 with 128MB.

    I now have a P4 2.4GHz, en 1 Gig Dual channel DDR Ram, en a Geforce FX5620 with 256MB Ram. Not to mention, 2 120GB Serial ATA150 harddrives.

    Now, the problem is this:

    Before doing this upgrade I did some benchamarking with Futuremark's 3DMark 2003 and Aquamark3. The results I got then were satisfying.


    Well, Now I have done a major upgrade. I ran both benchmarks again. And I get really disappointing results, which I cant even post here. :mad:


    I am guessing, there is some setting that I have to change? I looked in the BIOS for the AGP apperture size, so maybe increasing that might have helped. But I couldny find that setting anywhere in the BIOS of my mainboard, it is a Gigabyte 8KNXP Deluxe.
    :confused:
     
  2. Petros

    Petros Thief IV

    Messages:
    3,038
    Location:
    Pacific Northwest
    I would guess, from your configuration, that you should be scoring somewhere in the 11,000-point range in 3dMark 2001. I don't know about 03.

    Are your memory timings and your FSB speed in the bios set to the fastest stable settings?
     
  3. Taurus

    Taurus hardware monkey

    Messages:
    3,206
    Location:
    Sacramento, CA
    what gf4 did you have before and what score did you get?

    even if your current score is embarrassing, you should post the compare url so we can see the individual numbers which would help us find out what the problem is.

    is AA, AF, and V-Sync turned off?
     
  4. Sazar

    Sazar F@H - Is it in you? Staff Member Political User Folding Team

    Messages:
    14,905
    Location:
    Between Austin and Tampa
    agp aperture size should be at or larger than current onboard ram on your pcb... think of changing it to some other figures generally if there is a problem with performance or something else but from the sounds of it you should be fine in that respect...

    your gf4ti to gf fX performance figures in aquamark 3 should not change too drastically though there should be some improvement...

    the gf fx nv31 is not a bad card... in dx8 situations though it is only about as fast as a previous gen gf4 ti...

    actually.. what card DO you have?

    is it a gf FX 5600 or an ultra... or is it something else ? if it is a non-ultra... it may actually be slower in situations than your gf4 ti was... assuming you had a gf4 ti (hopefully you were not suckered into buying a gf4 mx...)
     
  5. LeeJend

    LeeJend Moderator

    Messages:
    5,291
    Location:
    Fort Worth, TX
    I've seen that 5620 number somewhere before. It is not a valid FX series number. I searched pricewatch, pricegrabber, and google with no hits.

    Whenever a manufacturer starts "playing" with version numbers be very careful. You could end up with anything.

    We need the manufacturers name, whole part number and any other details you have. Slower benchmarks than a GE4 (non MX) could be real.

    PS DOn't throw away the receipt and packing materials just yet.
     
  6. coathanger007

    coathanger007 Tomorrow Tweaking Today

    Messages:
    1,520
    Location:
    Australia
    GF4 Ti Faster than non-ultra FX5600. I changed from one to the other - worst decision in my life.