EVGA

benchnmarkking

Author
patchesanook
CLASSIFIED Member
  • Total Posts : 3903
  • Reward points : 0
  • Joined: 2006/02/04 05:54:08
  • Location: in a cave in west virginia
  • Status: offline
  • Ribbons : 9
2018/11/06 16:20:45 (permalink)
hi,
a friend of mine asked me what the best software to benchmark 2   1080ti,in sli
i have slacked-off in personal  pc building so i know is out of date. is 3Dmark still the king,or could you guys recommend something.
thank you.



#1

6 Replies Related Threads

    Sajin
    EVGA Forum Moderator
    • Total Posts : 49167
    • Reward points : 0
    • Joined: 2010/06/07 21:11:51
    • Location: Texas, USA.
    • Status: online
    • Ribbons : 199
    Re: benchnmarkking 2018/11/06 16:24:12 (permalink)
    3dmark time spy extreme will do the job. 
    #2
    TECH_DaveB
    EVGA Alumni
    • Total Posts : 4893
    • Reward points : 0
    • Joined: 2008/09/26 17:03:47
    • Status: offline
    • Ribbons : 46
    Re: benchnmarkking 2018/11/06 17:52:30 (permalink)
    Timespy did the trick for mine.  Superposition can really put the hurt on things, but as of last July when I was benching mine, it was still kinda unstable.
    #3
    Sajin
    EVGA Forum Moderator
    • Total Posts : 49167
    • Reward points : 0
    • Joined: 2010/06/07 21:11:51
    • Location: Texas, USA.
    • Status: online
    • Ribbons : 199
    Re: benchnmarkking 2018/11/06 17:59:52 (permalink)
    TECH_DaveB
    Timespy did the trick for mine.  Superposition can really put the hurt on things, but as of last July when I was benching mine, it was still kinda unstable.


    I would have recommended running unigine superposition on its 8k preset due to it being more stressful than 3dmark, but the op specifically stated testing sli. To get sli working in unigine superposition you have to do some sli profile tweaks to make it work. For 3dmark sli just works without having to tweak any sli profiles.
    #4
    TECH_DaveB
    EVGA Alumni
    • Total Posts : 4893
    • Reward points : 0
    • Joined: 2008/09/26 17:03:47
    • Status: offline
    • Ribbons : 46
    Re: benchnmarkking 2018/11/06 18:15:55 (permalink)
    I got some testing done in SLI on Superposition, but I honestly don't remember what I changed to make it work, no profiles though. I spent more time benching clocked in 3D mark, I did have some superposition scores but they were mainly stock, and also since heaven, I have been able to stabilize WAY higher GPU clocks in Unigine software than anywhere else.  Get a great score in Unigine, 3d mark immediately crashes to desktop, and a game only stays up and running 2 minutes.  Not sure why, but I have had that from my 8800 Ultra and testing Heaven before I got my 480, up through my 1080TI's.  So I do some testing in Unigine, but spend most of my serious benching time in 3D Mark.
    #5
    Sajin
    EVGA Forum Moderator
    • Total Posts : 49167
    • Reward points : 0
    • Joined: 2010/06/07 21:11:51
    • Location: Texas, USA.
    • Status: online
    • Ribbons : 199
    Re: benchnmarkking 2018/11/06 18:35:15 (permalink)
    TECH_DaveB
    I got some testing done in SLI on Superposition, but I honestly don't remember what I changed to make it work, no profiles though. I spent more time benching clocked in 3D mark, I did have some superposition scores but they were mainly stock, and also since heaven, I have been able to stabilize WAY higher GPU clocks in Unigine software than anywhere else.  Get a great score in Unigine, 3d mark immediately crashes to desktop, and a game only stays up and running 2 minutes.  Not sure why, but I have had that from my 8800 Ultra and testing Heaven before I got my 480, up through my 1080TI's.  So I do some testing in Unigine, but spend most of my serious benching time in 3D Mark.


    Sounds pretty normal to me as different applications will stress the gpu differently. You could run/pass the most stressful application in the world with your highest overclocks applied, but then later have those same overclocks fail in a less stressful application due to the way that specific application stresses the gpu. It's always best to test stability per application than to rely on one stressful one. 
    post edited by Sajin - 2018/11/06 18:42:06
    #6
    TECH_DaveB
    EVGA Alumni
    • Total Posts : 4893
    • Reward points : 0
    • Joined: 2008/09/26 17:03:47
    • Status: offline
    • Ribbons : 46
    Re: benchnmarkking 2018/11/07 16:14:34 (permalink)
    I agree completely, im just not serious enough with the clocking to optimize for each test.  But from my experience, a clock that is 3d mark stable is most likely to be stable in most/all other applications, whereas optimizations for other benches may not be as reliable outside of that specific test.  Once I have my peak 3d mark I see how it behaves elsewhere as it may need to be turned down, usually for gaming I turn my GPU's back down, give them the extra power they need and let the watercooling and extra power allow it to self clock all it wants.
    I was always curious about what was different in Unigine that allowed for ridiculous clocks that would run all day long looping their test and be fine that crashes EVERYTHING else.  Wonder what they are not loading  up on that others do.
    #7
    Jump to:
  • Back to Mobile