Low score with GTX 980 in Uniengine Benchmark

D

dogbyte_13

[H]ard|Gawd
Joined
Dec 21, 2004
Messages
1,486
So i got a new system with the the following specs.
PNY GTX 980 with latest Nvidia drivers
4690K @ 4.5GHZ Stable only top out at 1.25V
Corsair CS650M and it has 51A on the 12V rail
Rest of the specs in sig.
So i run heaven bench at DX 11 ultra and tessellation at normal and my scores seem off, i get 8.8 FPS min and around 72 Average this is such a large gap and just seems odd to me, i have the latest drivers and such. Could this be a Windows 8.1 issue? Now could this be hecause i use a DVI cable instead of a Display port or HDMI? Now one last question do synthetic benchmarks translate into real world performance? Anything i should look at? I haven't OCed my video card.
 
Looks normal.

Using Unigine Heaven Benchmark 4.0 with:

Ultra quality
x8 AA
1080p resolution

EVGA GTX 980 SC (not overclocked) on i7 4930K @ 4.5 GHz

FPS:
88.8
Score:
2238
Min FPS:
9.1
Max FPS:
160.2

Synthetic benchmarks translate only in the sense that you can perform apples to apples comparison to other systems...otherwise, I play the games and note performance in MSI Afterburner/etc. for "real world" results.
 
Last edited:
Would Kombuster be a better test since it does benchmarking? And seems to be closer to real world performance? Because my last system seemed to do better in this test than heaven.
 
No, because any benchmark that makes a gpu heat up like that is bad. It will make it throttle down while running it because it is staying at the max heat limit while running. The more it heats up the more it down-clocks to remove the heat. That is the last thing you want to do is heat the gpu doing a stress test. I would honestly get "slim drivers" and let it scan to see if you missed something in the drivers area. If not try doing a simple bios reset and load up default settings and see what you get then.

dogbyte_13 said:
Would Kombuster be a better test since it does benchmarking? And seems to be closer to real world performance? Because my last system seemed to do better in this test than heaven.
Click to expand...
 
dogbyte_13 said:
So i got a new system with the the following specs.
PNY GTX 980 with latest Nvidia drivers
4690K @ 4.5GHZ Stable only top out at 1.25V
Corsair CS650M and it has 51A on the 12V rail
Rest of the specs in sig.
So i run heaven bench at DX 11 ultra and tessellation at normal and my scores seem off, i get 8.8 FPS min and around 72 Average this is such a large gap and just seems odd to me, i have the latest drivers and such. Could this be a Windows 8.1 issue? Now could this be hecause i use a DVI cable instead of a Display port or HDMI? Now one last question do synthetic benchmarks translate into real world performance? Anything i should look at? I haven't OCed my video card.
Click to expand...

Actually seems normal to me on those settings.I am guessing its at the default 1080p?
 
Yes 1080P 60Hz, Does any games actually use tessellation? Or is that something that was thrown into DX11 and no one cares to use it? I am going to get a new monitor eventuall hopfeully 144Hz and possibly with G-Sync.
 
dogbyte_13 said:
Yes 1080P 60Hz, Does any games actually use tessellation? Or is that something that was thrown into DX11 and no one cares to use it? I am going to get a new monitor eventuall hopfeully 144Hz and possibly with G-Sync.
Click to expand...

Its just another way to use features of cards that support it. Again, for a apples and apples comparison. The benchmark isn't meant to indicate gaming performance per se. Its only used to compare systems.
 
KILLER_K said:
No, because any benchmark that makes a gpu heat up like that is bad. It will make it throttle down while running it because it is staying at the max heat limit while running. The more it heats up the more it down-clocks to remove the heat. That is the last thing you want to do is heat the gpu doing a stress test. I would honestly get "slim drivers" and let it scan to see if you missed something in the drivers area. If not try doing a simple bios reset and load up default settings and see what you get then.
Click to expand...

I have never heard of slim drivers, what are those?
 
Slimdrivers Will check all your devices and compare drivers to see if you have a up-to-date driver for the device. It is 100% free and it is a good mind to have to just check devices and drivers.
 
You will reach the cards TDP long before it overheats. I believe it was like Max 90 -92° degrees though. But the TDP is 145 watts on the gpu. Power limit is max 110% in MSI Afterburner. So only 10% over factory Power Limit.

This is why it is clocking down so fast in Kumbuster, etc. The gpu doesn't have a lot of wiggle room on it for the TDP. I guess that I should have use TDP in the place of the Heat part in my previous post. It would have been a better explanation.

Vatican said:
How much is too much heat before it starts downclocking on the msi 970's?
Click to expand...
 
Last edited:
i have the same problem, new built pc with 5930K, GTX 980 and gigabyte Gaming 7 mobo. at ultra and 8x AA i get 1908 at the unigine valley benchmark.2500 without 8x AA. score without 8x should be 4800-4900 with GTX 980 normally. no one has a clue what the problem could be.....nice to know i am not the only with this prob, started to doubt the card...
 
You must log in or register to reply here.
Back
Top