5970 Quadfire

Discussion in 'Videocards - AMD Radeon' started by Andypandy1991, Dec 29, 2010.

  1. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    Hello everyone,

    I have 2x5970 asus and i made the performance benchmark in 3dmark
    and i got 32000 gpu score!
    Is that normal?
    Cause i have seen 2x6950 making the exact same score
    Am i lacking something?

    Thnx in advance and happy new year to everyone
     
  2. Sever

    Sever Ancient Guru

    Messages:
    4,825
    Likes Received:
    0
    GPU:
    Galaxy 3GB 660TI
    5000 series doesnt scale as well as 6000 series. 5000 series gives you around 50-60% performance boost per extra card, whereas 6000 series gives you 80-90% boost per extra card (ignoring cpu bottlenecks).

    add in that quadfire generally doesnt scale perfectly either, very few people get any benefits from a fourth gpu.

    and is that 32000 gpu score in the p test? it does seem a little low.
     
  3. Hans Dampf

    Hans Dampf Master Guru

    Messages:
    235
    Likes Received:
    0
    GPU:
    2x 5970 H2O + GTX470 H2O
    32K GPU is far too low! Should be above 40K on stock speed.
    On high OC i can reach GPU 48K.

    You should drop all CCC settings to the bottom level like AA and AF.

    Edit: Just saw that you are running a 980x. So your scores are far far to low.
    Pls check also your BIOS settings. HT should be switched on for higher Vantage results and a proper CPU overclock should be recommended to push the 4 GPU's.
    4.4GHz should be nothing magic for your 32nm CPU.
     
    Last edited: Dec 29, 2010
  4. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I know its low:(
    All the settings are the default settings from when i installed the cards!
    I dont know guys!
    If i drop to the bottom all settings wont i lose image quality?
     

  5. Krogtheclown

    Krogtheclown Master Guru

    Messages:
    960
    Likes Received:
    0
    GPU:
    FuryXfire 3440x14freesync
    ya but if your just benchmarking who cares? People bench to get the fastest score just about everybody turn setting down to nill.
    I got over 40k with an I5 and 5970 xfire so ya you are very low. Didn't have allot of luck with quads though to much problems so I went back to tri 5870 which ran great!
     
  6. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I dont care about benchmarking!
    Really!
    But i said what the hell i will bench them just to see but crap:(
     
  7. Krogtheclown

    Krogtheclown Master Guru

    Messages:
    960
    Likes Received:
    0
    GPU:
    FuryXfire 3440x14freesync
    If you dont care about benchmarking why do it?
    If all you do is gaming then just watch your min fps and try to keep them as high as possible nothing else really matters anyway.
     
  8. Matt26LFC

    Matt26LFC Ancient Guru

    Messages:
    3,123
    Likes Received:
    67
    GPU:
    RTX 2080Ti
    OP doesn't care about benching and thats fair enough, however his score was considerably lower than most with that setup. Im sure he's just worried that something is not setup correctly or functioning improperly.

    I suggest run some games like Krog has suggested and see how they perform, you can use FRAPS to watch your FPS as you game, see whats what.
     
  9. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I will do that my friend!
    Thank you very much all and wish you a happy and prosperous year!
     
  10. antipacker

    antipacker Banned

    Messages:
    207
    Likes Received:
    0
    GPU:
    GTX 480 SLI
    i get more than that m8 with 2 6870's
    something wrong there pal
     

  11. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I know:p
    I just cant figure what!
    But its ok i will figure it out eventually
     
  12. Krogtheclown

    Krogtheclown Master Guru

    Messages:
    960
    Likes Received:
    0
    GPU:
    FuryXfire 3440x14freesync
    tell me what drivers are you using and can you play Stalker COP?
     
  13. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I got 10.12 drivers and i can play stalker great
     
  14. keromyaou

    keromyaou Member Guru

    Messages:
    103
    Likes Received:
    2
    GPU:
    EVGA RTX3080 XC3 Ul
    I have hd5970+hd5870+hd5870 (I do both trifire and quadfire). With quadfire, I can get Vantage GPU 47k with 10.12 driver under the default setting for CCC. If you have Metro2033, can you do the benchmark test using the setting (1680x1050, DX11, Very High, AAA, 16XAF, noDOF, noPhysX)?
     
  15. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    i play at 1920x1080 res and when i get back from work i will do the bench
     

  16. Krogtheclown

    Krogtheclown Master Guru

    Messages:
    960
    Likes Received:
    0
    GPU:
    FuryXfire 3440x14freesync
    Wow 10.12 must be great cause I could never play Stalker with any thing after 10.5 and a few other games. So how does metro do? I always wanted to play that game with my quads but by the time I figured out my game was screwed i had already bought my 580's.
     
  17. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    Options: Resolution: 1920 x 1080; DirectX: DirectX 11; Quality: Very High; Antialiasing: AAA; Texture filtering: AF 16X; Advanced PhysX: Disabled; Tesselation: Enabled; DOF: Disabled
    with these settings on Metro 2033 i get average 58fps

    Also stalker plays great with enhanced dynamic lighting
    If i apply enhanced lighting dx11 the xray engine crashes
     
  18. keromyaou

    keromyaou Member Guru

    Messages:
    103
    Likes Received:
    2
    GPU:
    EVGA RTX3080 XC3 Ul
    I see. I use that condition as a default for Metro 2033 benchmark because there is a thread like this (http://www.overclock.net/benchmarki...tro-2033-benchmark-thread-using-official.html). Under this condition I get 75-76 fps. Under your condition (with 1920 x 1080) I get 69-70 fps. Do you overclock your cpu and gpu? I overclock my cpu to 4.5GHz and my gpu to 974MHz (along with 1244MHz memory clock). If you don't overclock, your score (58fps) should be OK.

    For Krogthecrown,
    10.12 is much faster than 10.5a for Metro 2033 and Crysis Warhead. The Metro2033 benchmark under the standard condition (1680x1050 and so on) is 76fps for 10.12 and 66fps for 10.5a. I can comfortably play Metro 2033 with the maximum settings except for antialiasing (AAA instead of MSAA 4x). I manage to paly Metro 2033 with all maximum settings (with MSAA 4x). But in this case I encounter with some sluggish moments every 4-5 minutes.
     
  19. Krogtheclown

    Krogtheclown Master Guru

    Messages:
    960
    Likes Received:
    0
    GPU:
    FuryXfire 3440x14freesync
    I do to with my 580's hell if I knew they where ever gonna fix the 5970 quad drivers I would have keep them, but I waited a long time and I thought they where just to low tech to fix high tech problems so I gave up.
     
  20. Andypandy1991

    Andypandy1991 Master Guru

    Messages:
    269
    Likes Received:
    0
    GPU:
    3 x gtx 580's PNY
    I have overclocked my i7 980x to 4ghz
    But i dont think the perf loss is from there
     

Share This Page