Unigine Heaven 4 benchmark

[SS-89]

That Crossfire one is obvious, there is no tessellation on the floor.

Stealing my thunder with your eagle eyes , I knew there was something suspect but couldn't see anything in the picture to explain it

 

[Kaapstad]

No chance on the Unigine Benchmarks. They favour Nvidia too heavily.

If someone wants to top the charts on the Valley bench three 290Xs should have a good chance but four will be totally useless.

 

[Geeman1979]

With the pic being so blurry, it distracts the eye away from the obvious.

 

[SS-89]

If someone wants to top the charts on the Valley bench three 290Xs should have a good chance but four will be totally useless.

Valley or Heaven?

I found that I got zero scaling past two GPU's in Valley which was odd :/ However Heaven did scale for 3 GPU's

 

[Kaapstad]

Valley or Heaven?

I found that I got zero scaling past two GPU's in Valley which was odd :/ However Heaven did scale for 3 GPU's

Valley

Valley @1080p is a useless bench for multi GPU setups as it bottlenecks them really bad. What I have noticed with the 290Xs is they are not as fast as the Titans and 780ti's on a one on one basis but they handle a bottleneck better. This means in theory that 3 x 290Xs against say 3 x 780ti's on the Valley bench could come out on top but you are going to need a fast CPU (over 5ghz) to test it.

 

[SS-89]

Valley

Valley @1080p is a useless bench for multi GPU setups as it bottlenecks them really bad. What I have noticed with the 290Xs is they are not as fast as the Titans and 780ti's on a one on one basis but they handle a bottleneck better. This means in theory that 3 x 290Xs against say 3 x 780ti's on the Valley bench could come out on top but you are going to need a fast CPU (over 5ghz) to test it.

It still struck me as odd that I got absolutely no scaling on the third GPU, my score was near enough the same as 2 gpu's (or even the same can't rememember exactly)

7*** series cards seem to scale better on trifire in Valley.

 

[Silent_Scone]

Valley is a weird benchmark. I might do a run at 5Ghz in it but the volts needed for my Ivy-E chip are cringe worthy.

 

[Kaapstad]

Valley is a weird benchmark. I might do a run at 5Ghz in it but the volts needed for my Ivy-E chip are cringe worthy.

It is not worth it on NVidia cards, all that happens is you bounce off the GPU bottleneck.

4 Titans score about the same on Valley @1080p or 1600p which is a joke.

 

[Silent_Scone]

That really is a joke

 

[Kaapstad]

That really is a joke

My best @1600p for Valley was 5740 and @1080p I can not remember the exact score but it was less than 5800 lol.

 

[roadie]

I just ran a quick benchmark on my newly installed 670. CPU @ 4.5Ghz. 670 @ stock clocks (980/1502Mhz).

Does everything look OK?

 

[Kaapstad]

I just ran a quick benchmark on my newly installed 670. CPU @ 4.5Ghz. 670 @ stock clocks (980/1502Mhz).

Does everything look OK?

Yes it looks fine.

If you do an in bench screenshot I will put you on the scoreboard.

 

[Devious]

3770K @ 5.0Ghz
SLI GTX 670's @1220/1800

Bit low really

 

[Kaapstad]

3770K @ 5.0Ghz
SLI GTX 670's @1220/1800

Bit low really

It is actually a good score for a pair of GTX 670s.

And on the scoreboard.

 

[RetrogamerX]

Stock clocks. Don't know if this is a good score or not with this 780?

3770k @4.3
Inno3D 780 Ultra DHS Edition

 

[thefogo]

^Looks spot on

 

[Silent_Scone]

Seems about right for a stock run. Need a screenshot in Heaven to go on the leaderboard though dude .

 

[thefogo]

Here's my Classified at stock, as you're looking to get one. A touch slower.

 

[RetrogamerX]

Seems about right for a stock run. Need a screenshot in Heaven to go on the leaderboard though dude .

I knew this, but I thought Kapp stopped doing that?

 

[RetrogamerX]

Here's my Classified at stock, as you're looking to get one. A touch slower.

Your right.....and yours is OC too isn't?