Unigine Valley Benchmark Scores Thread + SUPERPOSITION ***Over 1000 Submissions!***

[done12many2]

1 minute ago, jeppouza said:

 

Thanks, post fixed.

 

Yep, they do not run at 100%, the benchmark just can't push them to the limit.

 

Actually, after looking at your revised submission post, it's your CPU clock speed that holding you back.  Valley at 1080p is pretty CPU bound and unless you crank your clock speed up, it's not really going to improve.  

 

Then there's the whole Valley hates Pascal SLI thing.  

[Jumper118]

1 minute ago, done12many2 said:

 

Actually, after looking at your revised submission post, it's your CPU clock speed that holding you back.  Valley at 1080p is pretty CPU bound and unless you crank your clock speed up, it's not really going to improve.  

 

Then there's the whole Valley hates Pascal SLI thing.  

Its cos he's running too low res and too high fps.  Anything is a cpu bench at super high fps. 

[Dive5885]

8 hours ago, Jumper118 said:

 

Your cpu is broken

My AIO is my only exhaust fan, and those 390X cards put out fire so I rather not OC at the moment.. lol

[Jumper118]

6 minutes ago, Dive5885 said:

My AIO is my only exhaust fan, and those 390X cards put out fire so I rather not OC at the moment.. lol

lol. would still under volt by 0.25v and still get 4ghz  

[JoshM813]

Ok here is my latest run with GPU-Z up this time haha

 

 

[Rickard.Shrimpy]

CPU: Intel i5 6600K 3,5Ghz-3,9Ghz boost
GPU: Asus 980ti Reference +190 Core, +250 Vram, With after market whater cooling solution from Corsair.   HG10 N980 And H60 mounted on that. Hitting 66 degrees.
Mem:16Gb 2133Mhz DDR4 From Kingston

SSD: Samsung 850 EVO

Powa!: EVGA Supernova 750 G2

Das Motha:GA-Z170M-D3H

 

 

[VagabondWraith]

17 hours ago, Rickard.Shrimpy said:

CPU: Intel i5 6600K 3,5Ghz-3,9Ghz boost
GPU: Asus 980ti Reference +190 Core, +250 Vram, With after market whater cooling solution from Corsair.   HG10 N980 And H60 mounted on that. Hitting 66 degrees.
Mem:16Gb 2133Mhz DDR4 From Kingston

SSD: Samsung 850 EVO

Powa!: EVGA Supernova 750 G2

Das Motha:GA-Z170M-D3H

 

 

66 degrees is awfully high for water cooling.

[DildorTheDecent]

19 minutes ago, VagabondWraith said:

66 degrees is awfully high for water cooling.

It is a H60 though. 

[VagabondWraith]

31 minutes ago, DildorTheDecent said:

It is a H60 though. 

I had an H55 on my 980 Ti and never went past 43c. 

[Rickard.Shrimpy]

16 hours ago, DildorTheDecent said:

It is a H60 though. 

Well kinda, the solution from Corsair is not optimal.
It´s hard to apply the block, and for the info the fan is on low rpm. Dont hear it ^^ 

and the onley "guide" is some dude from Corsair that apply it in less than 2 minutes. Not so good info on that one ^^

[Rickard.Shrimpy]

15 hours ago, VagabondWraith said:

I had an H55 on my 980 Ti and never went past 43c. 

Well, I might take it apart and try again but I never expected this "bracket" to be so badley optimized to 980ti . 
It really was bad to hear your card crack and bend  

So I would love to get an better guide on it than what Corsair have on Youtube. 
Well, I will try to remeber to write the new results when I´m done with it ^^

[DildorTheDecent]

6 minutes ago, Rickard.Shrimpy said:

Well kinda, the solution from Corsair is not optimal.
It´s hard to apply the block, and for the info the fan is on low rpm. Dont hear it ^^ 

and the onley "guide" is some dude from Corsair that apply it in less than 2 minutes. Not so good info on that one ^^

Could be an issue with block contact then. 

 

Not really a massive fan of AIOs. Although that's because i've a 1260mm radiator though. 

[Rickard.Shrimpy]

Just now, DildorTheDecent said:

Could be an issue with block contact then. 

 

Not really a massive fan of AIOs. Although that's because i've a 1260mm radiator though. 

Can be. 
I really tested this one because no one on Dustin had done it. 
So I wanted to try it out. But from 89 to 66 under load its way better so for the time I´m happy.
But will try to make it better soon. Maybe tomorrow. 

[Masada02]

On 10/3/2016 at 8:29 AM, Jumper118 said:

Yes if he want to do It I would change the thread. 

I'm a little late on this, but idk how useful it would be. I mean at this point we already have a huge amount of data for Valley; and, with DX12 and Vulcan coming around, is it really useful to add another soon-to-be outdated DX11 benchmark to the list? If Unigine or someone was to release a DX12/Vulcan bench I'd be totally down to start working on that, but I'm a little hesitant to add Heaven at this point in time.

[Jumper118]

2 minutes ago, Masada02 said:

I'm a little late on this, but idk how useful it would be. I mean at this point we already have a huge amount of data for Valley; and, with DX12 and Vulcan coming around, is it really useful to add another soon-to-be outdated DX11 benchmark to the list? If Unigine or someone was to release a DX12/Vulcan bench I'd be totally down to start working on that, but I'm a little hesitant to add Heaven at this point in time.

i would agree. wait for some new unigine engine to come out.

[done12many2]

4 minutes ago, Masada02 said:

I'm a little late on this, but idk how useful it would be. I mean at this point we already have a huge amount of data for Valley; and, with DX12 and Vulcan coming around, is it really useful to add another soon-to-be outdated DX11 benchmark to the list? If Unigine or someone was to release a DX12/Vulcan bench I'd be totally down to start working on that, but I'm a little hesitant to add Heaven at this point in time.

 

I don't think it's any different than what's happening with 3DMark.  Time Spy is the new DX12 benchmark, but the majority of users continue to use Fire Strike over Time Spy for some reason.  I personally prefer Time Spy as I usually go with the latest and greatest, but I am far from the majority.  

 

Heaven remains very relevant on LTT.  On top of that, it isn't a CPU benchmark like Valley is (at least at 1080p, which everyone runs).

[Jumper118]

2 minutes ago, done12many2 said:

 

I don't think it's any different than what's happening with 3DMark.  Time Spy is the new DX12 benchmark, but the majority of users continue to use Fire Strike over Time Spy for some reason.  I personally prefer Time Spy as I usually go with the latest and greatest, but I am far from the majority.  

 

Heaven remains very relevant on LTT.  On top of that, it isn't a CPU benchmark like Valley is.

if people ran the right resolution for there gpu it wouldn't be a cpu benchmark  

 

hey look at me running 1080p on my titan xpp sli trololol 

[done12many2]

Just now, Jumper118 said:

if people ran the right resolution for there gpu it wouldn't be a cpu benchmark  

 

Notice I edited my post.  

[Masada02]

5 minutes ago, done12many2 said:

 

I don't think it's any different than what's happening with 3DMark.  Time Spy is the new DX12 benchmark, but the majority of users continue to use Fire Strike over Time Spy for some reason.  I personally prefer Time Spy as I usually go with the latest and greatest, but I am far from the majority.  

 

Heaven remains very relevant on LTT.  On top of that, it isn't a CPU benchmark like Valley is (at least at 1080p, which everyone runs).

I think a lot of it comes down to available DX12 games/engines. Until there are solid DX12 engines that the smaller studios can use there won't be a major change to DX12 or Vulcan and people won't be as excited to run the benchmarks for the new tech.

 

The other part is, adding Heaven is going to fragment the benches that are run even more. With both benches being free, I think it's easier just to specify which one to run so that everyone can be comparable. If there was a noticeable demand for Heaven I think I'd be more open to it, but I'm not sure the demand is there.

[done12many2]

4 minutes ago, Masada02 said:

I think a lot of it comes down to available DX12 games/engines. Until there are solid DX12 engines that the smaller studios can use there won't be a major change to DX12 or Vulcan and people won't be as excited to run the benchmarks for the new tech.

 

The other part is, adding Heaven is going to fragment the benches that are run even more. With both benches being free, I think it's easier just to specify which one to run so that everyone can be comparable. If there was a noticeable demand for Heaven I think I'd be more open to it, but I'm not sure the demand is there.

 

I think the Leaderboard took us in the right direction with regards to trying to get people to focus on resolutions higher than 1080p.  

[Jumper118]

33 minutes ago, done12many2 said:

 

Notice I edited my post.  

yes you ninja'ed me. 

 

19 minutes ago, done12many2 said:

 

I think the Leaderboard took us in the right direction with regards to trying to get people to focus on resolutions higher than 1080p.  

agreed 1080p is console peasant res these days even a 4gb gtx680 or hd7970 can run 1440P on most older titles as long as they arent maxxed out. 

[done12many2]

4 minutes ago, Jumper118 said:

agreed 1080p is console peasant res these days even a 4gb gtx680 or hd7970 can run 1440P on most older titles as long as they arent maxxed out. 

 

It's easier to do.  It's really that simple.  I'd rather see 60 or 100 quality frames passing over my screen than 144 low res ones.  That's just my preference.  

[Jumper118]

1 minute ago, done12many2 said:

 

It's easier to do.  It's really that simple.  I'd rather see 60 or 100 quality frames passing over my screen than 144 low res ones.  That's just my preference.  

same, but the games that require 144fps are all easy to run away. you dont need a titan xp to play csgo at 1080p 144fps  then overwatch and things like this arent hard to run either. 

[done12many2]

Just now, Jumper118 said:

same, but the games that require 144fps are all easy to run away. you dont need a titan xp to play csgo at 1080p 144fps  then overwatch and things like this arent hard to run either. 

 

Exactly.

[Marinatall_Ironside]

CPU: Intel Core i7-5930K CPU @ 4.2 GHz 1.200000
GPU(s): EVGA GeForce GTX 960 4GB FTW ACX 2.0+ @ 1,510 MHz Core, 4,001 MHz Memory, and 1.250 volts.
Average FPS (1080p): 38.9 
Score (1080p): 1,629
Average FPS (1440p): 23.1
Score (1440p): 965

Average FPS (4K): 10.7
Score (4K): 447

 

I had to enter a custom resolution of 3840 x 2160 because that resolution was not available at all. The 4K run was definitely painful 

 

Can you also please put these scores onto the average FPS leaderboard 

 

Here are the pictures. Note that the CPU-Z shot was without the CPU turboing to its 4.2 GHz overclock.

Spoiler