Results 1 to 6 of 6

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1
    Joined
    Apr 2017
    Posts
    2

    1070 vs 1080 not what I expected

    I have put together 18 systems this year for my office. 10 with EVGA GTX 1070 SC and 8 with Gigabyte GTX 1080 Windforce. I am by no means an expert and only ran two different free bench mark programs but found my systems running the 1070 ran better than the 1080...

    All machines on MSI z270 M3 boards, i7 7700K, 32gb G.Skills Ripjaws, and Samsung Evo 850 500GB SSD.

    GTX 1070 Nova bench average graphics score: 1891
    GTX 1080 Nova bench average graphics score: 1467

    We are a engineering shop and live on Autocad so I also used Cadalyst Systems Benchmark

    GTX 1070 Cadalyst Systems Benchmark 3D index: 1962, 2D Index: 549
    GTX 1080 Cadalyst Systems Benchmark 3D index: 2037, 2D Index: 526

    I used the same system for the test, did a clean install of the most current driver after GPU swap.... Not sure why the Nova Bench scores for the 1080 are so bad and I would have not expected the Cadalyst Systems Benchmark to be a dead heat. Anyone have thoughts on this? What am I missing here!?

    Thanks in advance!
    Last edited by hammershed; 04-19-2017 at 10:36 AM.

  2. #2
    sttubs is offline Always learning something
    Joined
    Feb 2004
    Location
    Rock Falls, IL
    Age
    51
    Posts
    984

    Re: 1070 vs 1080 not what I expected

    Are all of the 1080's performing like that?
    AMD 8350 w/Xigmatek Dark Knight Night Hawk Ed - Asus M5A99FX PRO R2.0 w/4x4gb GSkill Ripjaws 1600 CL7 - Samsung 850 Evo 500gb - WD Black 5Tb - Gigabyte 980 Ti Xtreme Gaming OC Ed - Win 10 Pro 64bit - Lite-On iHES108 - White Corsair 500r w/Corsair AX860i 860W - ASUS VG248QE 144Hz

    Heat: http://www.heatware.com/eval.php?id=55337
    Ebay: http://myworld.ebay.com/sttubs/

  3. #3
    Joined
    Apr 2017
    Posts
    2

    Re: 1070 vs 1080 not what I expected

    Yes

  4. #4
    Joined
    May 2017
    Location
    Texas
    Posts
    15

    Re: 1070 vs 1080 not what I expected

    This is what i am afraid of. Have been seen it for a while, but does not really know if its worth for the price. Check some information and see Digital foundry test and found that there is no much significant improvement. So, for me not and it would be good for anyone who is not having yet 1070.

  5. #5
    Joined
    Jun 2012
    Posts
    177

    Re: 1070 vs 1080 not what I expected

    I'm only guessing, and maybe this will spark something, but here's a few thoughts:

    First, it looks like Novabench may be dependent on both CPU and GPU, so part of the results could be influenced by the other components, including the CPU but also maybe RAM. According to Novabench, the GTX 1080 should beat the 1070 by about 5%.

    Second, the 1080 is designed to beat the 1070 in gaming performance - other productivity purposes may not see similar results. You would think that the 1080 should perform better, but if the benchmark is testing a part of the GPUs where the differences are small (or worse, they're the same), then you see results like you describe.

    Can you think of any other factor outside of the GPUs that could be causing this mystery?

  6. #6
    Joined
    Aug 2000
    Location
    Vvardenfell
    Age
    56
    Posts
    10,836

    Re: 1070 vs 1080 not what I expected

    Some other obvious thoughts: were the settings identical in the graphics? At all levels? They ought to be, but that is not necessarily true. Silly things like FSAA level can kill performance.

    Have you tried swapping GFX over between machines and seeing if the scores stay with the CPU or the GPU? Again checking that the GFX hardware isn't causing settings to change.

    But yes, without some investigation it makes little sense.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •