Xfire cables - one or two??? and where.

Discussion in 'Video Cards & Monitors' started by S1L, Nov 12, 2010.

  1. S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    I found something interesting the other night. When using two xfire bridge cables to connect my two 5870's I found the performance improved when compared to using one cable on either slot.

    The test was done unigine heaven latest realease and running on all stock settings on the benchmark 1920*1080 etc
    Now I am not just talking a few frames etc I am talking almost 100fps.

    I will post the results later just uploading pics however I am interested to hear peoples opinions of this as I have scoured the forums and found that the general consensus is that it doesn't matter which one you use or how many the results will remain the same.
     
  2. DiGiTaL MoNkEY

    DiGiTaL MoNkEY Inverted Monkey

    Joined:
    Jun 28, 2005
    Messages:
    26,894
    Location:
    Melbourne, Victoria
    I swear this was debunked some time ago, but maybe this series has some advantages with using two...either-way I'm interesting to see some results :thumbup:
     
  3. terrastrife

    terrastrife Member

    Joined:
    Jun 2, 2006
    Messages:
    18,772
    Location:
    ADL/SA The Monopoly State
    one of your cables is broken.

    the sli/cf bridge has about pcie 1x abdwidth, not much.
     
  4. DiGiTaL MoNkEY

    DiGiTaL MoNkEY Inverted Monkey

    Joined:
    Jun 28, 2005
    Messages:
    26,894
    Location:
    Melbourne, Victoria
    Could be quite possible as well.
     
  5. gaz@jb

    gaz@jb (Banned or Deleted)

    Joined:
    Mar 2, 2007
    Messages:
    1,308
    Location:
    West coast
    I agree it is possible.

    OP:
    Retry with both bridges only using one at a time, you should be able to see the difference as soon as the benchmark starts, that is if it is broken, if not the results should be identical.

    Then retry the dual bridge to see if your frame rates go back up 100fps

    It shouldn't take too long :thumbup:
     
  6. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    These are the results.
    Now I couldn't get the actual ss of the finished becnhmarks as they are firefox docs and I haven't been able to upload them anywhere or save them as jpg.
    The settings in unigine are stock and res is 1920 x 1080.

    The two 5870 I have are the Asus 5870 1GB and the Gigabyte 1GB 5870 ultra durable.

    Both are in 16x pcie slots.
    With one xfire cable and monitor running from the asus card I got:
    32.9 avg fps
    16.4 min fps
    85.3 max fps
    Score = 829

    The same config but with two xfire cables:
    65.8 avg fps
    22.6 min fps
    178.6 max fps
    Score = 1659

    With two xfire cables but with gigabyte card running monitor:
    53.5 avg fps
    18.6 min fps
    133.4 max fps
    Score = 1347
    With one fire cable and monitor running from gigabyte card:
     
  7. terrastrife

    terrastrife Member

    Joined:
    Jun 2, 2006
    Messages:
    18,772
    Location:
    ADL/SA The Monopoly State
    you seem to be missing the 'test one cable at a time' theory.
     
  8. t8y

    t8y Member

    Joined:
    May 5, 2002
    Messages:
    4,311
    Location:
    SA, 5011
    seems you left some testing out? your post didnt sound completed (how it ended with "With one fire cable and monitor running from gigabyte card: " and then no further result..)

    also on the .jpg issue, try using printscreen button and then paste into mspaint?


    but yeah, subscribed as im curious to see how this pans out with the testing suggested by the other fellas here
     
  9. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    I thought the single cable with either card would yield similar results especially since that when tests were run on a single card both cards scored as follows.

    Asus 5870
    33.8 avg fps
    4.7 min fps
    91.7 max fps
    score = 851

    Gigabyte 5870
    35.3 avg fps
    16.5 min fps
    92.4 max fps
    score = 890

    The scores are so similar the results I believe would only reinforce the fact that using two xfire cables allows more bandwidth between the cards increasing fps etc.

    However for all you skeptics out there I will run the test now with the Gigabyte card running the monitors and post the results.

    I will also try the ss into mspaint idea and see how we go.
     
  10. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    If you read the post above yours you will find that I have covered the single xfire cable with the asus card running the monitors.
     
  11. terrastrife

    terrastrife Member

    Joined:
    Jun 2, 2006
    Messages:
    18,772
    Location:
    ADL/SA The Monopoly State
    but you havent covered both cables.
     
  12. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    What do you mean both cables?

    Also just got the results for the Gigabyte running the monitors.
    The results are as follows.
    32.7 avg fps
    14.5 min fps
    85.5 max fps
    score = 825

    Also I ran the test again with two xfire cables and the Gigabyte card running the monitors and these are the results.

    65.1 avg fps
    18.5 min fps
    177.2 max fps
    score = 1641

    As you can see it is pretty much double the results of the previous test using the exact same setup but with two xfire cables instead of one.
     
  13. thrillhouse

    thrillhouse Member

    Joined:
    Mar 3, 2002
    Messages:
    1,825
    Location:
    sydney sydney sydney
    1. Run test with cable 1
    2. Redo test with cable 2

    Post results.
     
  14. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    WTF do you mean the test I just ran was exactly that. With one xfire cable and then two.

    The cables were just opened a few days ago from brand new. The cables aren't going to make any difference?
     
  15. Autti

    Autti Member

    Joined:
    Aug 27, 2009
    Messages:
    6,570
    Do a test with ONE of your cables, then repeat the test with the OTHER cable.

    To compare the two cables against each other.
     
  16. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    Is this because you are concerned about the integrity of the cable?
     
  17. HeXa

    HeXa Member

    Joined:
    Jul 7, 2001
    Messages:
    10,216
    Location:
    Canberra, ACT
    wasn't that obvious with the 2nd reply to your thread and the two people who agreed? :confused:
     
  18. mojododo

    mojododo Member

    Joined:
    Feb 14, 2004
    Messages:
    3,440
    Location:
    Melbourne
    This will be interesting to see if it is indeed a broken xfire cable.
    (the whole reason people wanted you to do the tests originally BTW)

    Either way, youll find out if 1. a cable is rooted, or 2. two xfire cables actually help.
     
  19. Nosrulz

    Nosrulz Member

    Joined:
    Nov 9, 2009
    Messages:
    929
    Location:
    Brisbane
    Another option would be to turn off crossfire and test.

    Considering your benchmark scores are nearly doubling when you use 2 xfire cables it sounds like a broken cable.

    Edit: Just found a screenshot of the unigine results that I ran with my 5850 OCed to 950/1200. My score basically matches your score when you used 1 xfire cable. I'm 100% certain you've got a broken cable.

    [​IMG]
     
    Last edited: Nov 14, 2010
  20. OP
    OP
    S1L

    S1L Member

    Joined:
    May 18, 2010
    Messages:
    70
    Location:
    Melbourne, Australia
    Well as I said the cables were only opened a few days ago and I find it hard to believe that either cable had been damaged.

    However to satisfy you all I ran the test with the other cable and the results were the same.

    Gigabyte powering monitors using the second xfire cable.
    34.3 avg fps
    15.3 min fps
    89.4 max fps
    score = 865

    As you can see the results are very similar to the results after using the gigabyte powering the monitors with a single xfire cable.

    Gigabyte xfire using first cable.
    32.7 avg fps
    14.5 min fps
    85.5 max fps
    score = 825
     

Share This Page

Advertisement: