• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

**Official Unigine Valley Leaderboard**

Caporegime
OP
Joined
17 Mar 2012
Posts
47,628
Location
ARC-L1, Stanton System
As I said the other day Valley does not report the clocks correctly for all NVidia cards.

He gave me a clock speed of 1224 Mhz, his score was 3200~

Thats the same as the average 780 with the same reported clock speeds, 10~% slower that the average GTX Titan with similar reported clock speeds, 20% slower than the fastest GTX Titans, given the 780TI has 10% more cores than the Titan i would have thought clock for clock it should be 7% faster.

Not the same performance (clock for clock) as the 780 which has 25% less cores.

The Titan beating GPU can't even beat the 780. something is not right.
 
Man of Honour
Joined
21 May 2012
Posts
31,940
Location
Dalek flagship
He gave me a clock speed of 1224 Mhz, his score was 3200~

Thats the same as the average 780 with the same reported clock speeds, 10~% slower that the average GTX Titan with similar reported clock speeds, 20% slower than the fastest GTX Titans, given the 780TI has 10% more cores than the Titan i would have thought clock for clock it should be 7% faster.

Not the same performance (clock for clock) as the 780 which has 25% less cores.

The Titan beating GPU can't even beat the 780. something is not right.

Your maths for starters.:D

The 780ti has 7.14% more cores

The rest would be setup and drivers, the recent drivers from NVidia are not a lot of fun for benching.

And as I said earlier Valley does not report the clocks correctly.:D

Edit just looking at this again, MjFrosty has beaten the GTX 780s running with normal volts and bios. The only ones above him will be using big volts and/or modded bios (you also have Kaktus69 in as a double entry @No.10 and No.12).

I have had a quick look on overclock.net and some people are not getting the performance in this bench they were expecting so the remainder of the problem at a guess I would put down to the launch drivers (780ti users can not load up one of the 320.xx family of drivers like Titan and GTX 780 users can).
 
Last edited:
Man of Honour
Joined
21 May 2012
Posts
31,940
Location
Dalek flagship
Yep these drivers are pants for Heaven. There's plenty of life to be sucked out of these yet. Going to give the other two a try later.

Nothing a new driver won't fix (should not take too long).

On this bench don't worry if you get a bad score using 3 cards, @1080p the CPU will get a horrible bottleneck. If you can run at a higher rez like 1440p or 1600p this will enable your cards to run with a smaller/no bottleneck.
 
Associate
Joined
26 Jan 2012
Posts
88
1080p score 3308, GPU GTX 780 Ti @ 1101/1900, CPU 2500k @ 4.55GHz

kRGPyDZ.jpg

1440p score 2072, GPU GTX 780 Ti @ 1101/1900, CPU 2500k @ 4.55GHz

k4YrnMD.jpg
 
Man of Honour
Man of Honour
Joined
18 Oct 2002
Posts
4,422
Location
2nd City - Manchester.
How do I stop my intel gfx form showing up at the end? I've disabled it in the device manager but it still shows up >.<

set it to PCIe or better to Auto in the BIOS, then it will get disabled when the PCIe card is detected and only enable if you remove the add-in card. You've probably got it set to "enabled" or "on".

Probably re-enable it in device manager first or it would probably only work in safe-mode if you removed the add-in card
 
Associate
Joined
26 Oct 2013
Posts
365
R9 290 1100/1250 (+30% max power)-I have only gotten some sort of error on Firestrike and this, where there is one frame where a few pixels seem to be flashing, normally in line with a rock or something. Should I increase power, or decrease clock speed?
2600k@4300

Run 1: 1080p 4xAA High Quality
24d108l.jpg

Run 2: 1080p 8xAA Ultra Quality
mc3gab.png

Not re-running again now sorry, test rig is now dismantled. It is legit, done at the right settings.

R290 Pro at 1220 / 6400 on a Haswell 5GHz CPU.

I am not here to try and get the best scores, the result is there to merely just show users what the new card can do.

So it helps the users to see it in the chart, so please add it, your decision in the end, but I am the last person who is gonna be trying to forge/fake results.

I am guessing that's on +50% max power? I am afraid my 650W psu will blow up
 
Last edited:
Back
Top Bottom