Unigine Superposition Benchmark.

[straxusii]

Tried pushing the mem, re-ran 1080ti @2ghz core, 3078mhz ram. 3930k still at 4.4ghz.

1080p again:

 

[Kaapstad]

1070 @2088/2375
6950X @4.444
Drivers 381.65

So close to beating Humbug but missed out by 2 points.

 

[LtMatt]

Hi Kaap,

I don't know if this counts for an entry, if not i can rerun. I used 1080P Extreme preset and recorded it using Relive at the same time.

There is no MGPU support for CrossFire in this title and i don't think any is planned, so i used 1x1 Optimize which appears to be offer almost perfect scaling.

I will rerun again for a proper entry as per the OP requirements regarding Custom.

Spoiler

 

[LtMatt]

1800X @4.1Ghz
1x Radeon Pro Duo @1150Mhz/500Mhz CrossFire disabled
Crimson 17.4.2
1080P

Spoiler

 

[Kaapstad]

Hi Kaap,

I don't know if this counts for an entry, if not i can rerun. I used 1080P Extreme preset and recorded it using Relive at the same time.

There is no MGPU support for CrossFire in this title and i don't think any is planned, so i used 1x1 Optimize which appears to be offer almost perfect scaling.

I will rerun again for a proper entry as per the OP requirements regarding Custom.

If you can re run please.

 

[Kaapstad]

Here is my little air cooled 980 Ti on stock volts.



980 Ti @1560/2103
6950X @4.444
Drivers 381.65

 

[The Artfull Bodger]

1080 TI (FE) @2126/+620 mem

4790K @4.8

Drivers 381.65

 

[Kaapstad]

Scoreboard updated.

 

[fenchu]

actually starting to feel better in referance to my 2016 titan given scoreboard. Was close to ebaying and purchasing 2017 but sanity prevailed

 

[humbug]

1070 @2088/2375
6950X @4.444
Drivers 381.65

So close to beating Humbug but missed out by 2 points.

Don't make me stick my PC in the fridge

 

[jklc]

Anyone else having trouble running the VR ready benchmark? It just goes to a black screen for a second then bombs out back to the main menu.

 

[SupraWez]

I dropped back to older NVidia drivers and ran some more benchmarks, at stock levels on a 1080FE the older drivers scores better and is stable at higher overclocks but the newer driver scores higher, but less stable at same overclocks, interesting indeed.

Will upgrade again and run some more tests.

 

[humbug]

I dropped back to older NVidia drivers and ran some more benchmarks, at stock levels on a 1080FE the older drivers scores better and is stable at higher overclocks but the newer driver scores higher, but less stable at same overclocks, interesting indeed.

Will upgrade again and run some more tests.

Are you not finding the FX-8350 a bottleneck to the 1080?

 

[FlyingScotsman]

Eventually managed to break 6200 !!!

EVGA 1080ti @2025/6066
[email protected]

Going to play with older drivers.

 

[SupraWez]

Are you not finding the FX-8350 a bottleneck to the 1080?

During the benchmark the CPU sat around 20%

 

[FlyingScotsman]

Another 4K run
EVGA 1080ti @2050/6066
[email protected]

 

[Quartz]

This thing doesn't like SLI! Anyway, I tried it in the 4K optimised mode:

 

[Kaapstad]

This thing doesn't like SLI! Anyway, I tried it in the 4K optimised mode:

Do you have any GPU and CPU clockspeeds please.

I can only use the single card run as SLI is not supported yet.

 

[Razor Time]

EVGA FE on water 2088/6200, 6800K @4.2Ghz
Drivers: 381.65

 

[AiiR]

The 1070 looks faster on paper, in reality its not, the thing with the 1070 is it boosts out of the box between 1800Mhz and 1900Mhz, some even a little higher than that, mine starts at 1911Mhz, but after a few minutes once its warmed up it drops to about 1870Mhz, when its already running at those sorts of speeds out of the box there isn't a lot of room left for overcooking, the average being about 2050Mhz, its 200Mhz above out of box boost clocks, its 10%.

Thats the average real overclock, the 980TI often gets 30% above its reference clocks, so while in reviews the GTX 1070 looks 10 to 15% faster than the 980TI in reality they are about even if not a slight win overall to the 980TI.

Interesting to know!