1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Is there an error in the 5970 reviews?

Discussion in 'Comments & Feedback' started by Black Panther, Jan 10, 2010.

  1. Black Panther

    Black Panther Senior Moderatorâ„¢ Staff Member

    Joined:
    May 30, 2007
    Messages:
    8,561 (3.27/day)
    Thanks Received:
    1,916
    I was going through the reviews for the 5970 and the 5970 in CF.

    How come the comparisons in the CF Review show the single 5970 performing much better than in the single 5970 review itself :confused:

    This is what I mean:

    [​IMG]

    [​IMG]
  2. InTeL-iNsIdE Guest

    For a start 4 gpus dont scale as well as 2 and that lessens with 3 and more so with 4 etc, also drivers are still very immature for the HD 5 series, I can see in 2 months or so time we will have a nice stable set of mature drivers for the 5 series cards. Not to mention a beast like the 5970 crossfired IS going to be bottlenecked by a 3.8ghz i7, or at least I would think so, I am betting with 4.5ghz or so you might start to see different numbers


    Damn just reread all of your post and I was on about something completely diff, nvm lol, though yea, what gives??
  3. Black Panther

    Black Panther Senior Moderatorâ„¢ Staff Member

    Joined:
    May 30, 2007
    Messages:
    8,561 (3.27/day)
    Thanks Received:
    1,916
    Lol yes what I meant is (forget the crossfire) both graphs are Crysis benchmark 1920x1200 4xAA.
    In the first graph the single card 5970 gets 44.9FPS and in the second graph the same single card 5970 gets 74.5FPS :wtf:
  4. Kreij

    Kreij Senior Monkey Moderator Staff Member

    Joined:
    Feb 6, 2007
    Messages:
    13,881 (5.08/day)
    Thanks Received:
    5,615
    Location:
    Cheeseland (Wisconsin, USA)
    I think that when the lone 5970 actually had some completition from its siblings that it became more serious about performing. Peer pressure?
  5. DirectorC

    DirectorC New Member

    Joined:
    Nov 4, 2009
    Messages:
    1,624 (0.94/day)
    Thanks Received:
    268
    Location:
    Florizy
    I'm guessing there's no way the 5970 actually gets 75fps.
  6. Fourstaff

    Fourstaff Moderator Staff Member

    Joined:
    Nov 29, 2009
    Messages:
    9,158 (5.38/day)
    Thanks Received:
    1,959
    Location:
    Home
    What are you suggesting, Kreij? :slap:

    I think its the ATI cooling version versus the HIS version, both in crossfire
  7. Kreij

    Kreij Senior Monkey Moderator Staff Member

    Joined:
    Feb 6, 2007
    Messages:
    13,881 (5.08/day)
    Thanks Received:
    5,615
    Location:
    Cheeseland (Wisconsin, USA)
    I'm not suggesting anything, just making a less than helpful comment. :rolleyes:
    I do believe that W1zz did both of the reviews so only he can tell us what's up with the descepancy in the numbers.
  8. Fourstaff

    Fourstaff Moderator Staff Member

    Joined:
    Nov 29, 2009
    Messages:
    9,158 (5.38/day)
    Thanks Received:
    1,959
    Location:
    Home
    So, when can we get the mighty W1z to come and banish the doubt?
  9. department76

    department76 New Member

    Joined:
    Jan 7, 2008
    Messages:
    519 (0.22/day)
    Thanks Received:
    60
    Location:
    U of WY, USA
    since they are different reveiws, i'm guessing different drivers were used. the most recent of course with possibly a newer driver.
  10. 3volvedcombat

    3volvedcombat New Member

    Joined:
    May 10, 2009
    Messages:
    1,514 (0.79/day)
    Thanks Received:
    261
    Location:
    South California, The desert.
    Ya i dont think a HD 5970 is going to be getting 80 fps averages LOL.
    and yet the first test it gets 45-50 fps averages.

    BUT NONE THE LESS, WE DO KNOW THE CARD IS VERY FAST ANYWAYS.


    :roll::roll::roll::roll:

    So i say good job reviewing the card wizzard, even if thats a error, it takes time and money to review and he does it for almost a living haahaha.

    (I would pay to review this though haahahah).
  11. HTC

    HTC

    Joined:
    Apr 1, 2008
    Messages:
    2,237 (0.97/day)
    Thanks Received:
    302
    In this particular bench, W1zz made a mistake and benched in "high" instead of "very high", which is why there is such a difference.

    He said so in post #96 of this thread.

    I thought it had been corrected already but apparently not.
  12. zithe

    zithe

    Joined:
    Jun 16, 2008
    Messages:
    3,087 (1.38/day)
    Thanks Received:
    345
    Location:
    North Chili, NY
    Still amazing performance. No need to use all very high when you don't see some of the effects.
  13. Black Panther

    Black Panther Senior Moderatorâ„¢ Staff Member

    Joined:
    May 30, 2007
    Messages:
    8,561 (3.27/day)
    Thanks Received:
    1,916
    It was corrected in the 5970 review, but apparently not in the 5970 crossfire review.

    W1zzard?
  14. Bjorn_Of_Iceland

    Bjorn_Of_Iceland

    Joined:
    Jan 2, 2008
    Messages:
    3,172 (1.32/day)
    Thanks Received:
    375
    oh ok.. so it looks like 44.9 is the correct very high value.. pretty much not far off in todays standards hehe.

    Seriously, I was peeing my pants when I saw that 74.5 @ very high lol.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page