***Official Unigine Heaven DX11 3.0 Benchmark Scores***

[Besty]

7970 1050p Single run.

CPU @3930k 4.7ghz memory 9-10-10-21 1T

GPU @ XFX 7970 1365mhz/1900mhz Trixx.


http://img507.imageshack.us/img507/7025/heaven20120708123808019.png

http://www.techpowerup.com/gpuz/g5cec/

you son of a bitch !

 

[piggin]

Results at 1680x1050

CPU 5125Mhz
RAM 2333 Mhz 9-11-10-28
GPU (1) MSI +185 / +540 slot 1 @ 16x
GPU (2) Zotac +130 / +480 slot 1 @ 8x
GPU (3) MSI +170 / +550 slot 1 @ 16x

Original links

http://i1100.photobucket.com/albums/g419/piggin25/00010.jpg

http://i1100.photobucket.com/albums/g419/piggin25/validation.jpg

 

[piggin]

Results at 1920x1080

CPU 5125Mhz
RAM 2333 Mhz 9-11-10-28
GPU (1) MSI +185 / +540 slot 1 @ 16x
GPU (2) Zotac +130 / +480 slot 1 @ 8x
GPU (3) MSI +170 / +550 slot 1 @ 16x

Original links

http://i1100.photobucket.com/albums/g419/piggin25/00011.jpg

http://i1100.photobucket.com/albums/g419/piggin25/validation.jpg

 

[Locky]

you son of a bitch !

Got to admit these 12.7 really pull out some performance from these cards, I can for now forgive them for really poor 3dmark scores - as everything else flies.

they inform me - they are working on the 3dmark problem.

 

[8 Pack]

3D mark 11 with single card is improved around 5%

 

[piggin]

Could someone have a quick look at my scores and tell me if I have to add anything to the screen grabs to make these scores count.

Thanks

 

[F1aw1ess]

1050p run

Link To Full http://personal.chandlerit.co.uk/heaven/heaven1050p.jpg

GPUz Validation http://www.techpowerup.com/gpuz/ukadx/

Details as below except change GTX670 to 1357mhz / 7658mhz

 

[Locky]

3D mark 11 with single card is improved around 5%

trifire has UN-improved by 25%

 

[F1aw1ess]

Could someone have a quick look at my scores and tell me if I have to add anything to the screen grabs to make these scores count.

Thanks

You need a GPUz validation link showing full clocks, otherwise should be fine.

 

[piggin]

You need a GPUz validation link showing full clocks, otherwise should be fine.

Thanks m8 for the pointer
Original posts now changed

 

[GoogalyMoogaly]

Could anyone with a GTX 670 tell me what your stock speed score for 1080p is please? (not worried about screenshots or anything)

 

[Bradmax57]

Could anyone with a GTX 670 tell me what your stock speed score for 1080p is please? (not worried about screenshots or anything)

around 1850 I think mine was

 

[GoogalyMoogaly]

around 1850 I think mine was

Thanks. I was getting high 1700s when I tried mine. Still seems a bit slow. Maybe it'd work better if I physically removed the second card rather than disabled SLI.

I just wanted to see how much of a boost I was getting with SLI.

 

[pgi947]

It also depends on the stock boost of the 670, not all are the same, some windforces can vary up-to and over by 100mhz.

 

[GoogalyMoogaly]

My main boosts to 1168 (or something like that) the second one boosts to 1188. At some point I'll swap them over (or sell them both).

 

[F1aw1ess]

Could anyone with a GTX 670 tell me what your stock speed score for 1080p is please? (not worried about screenshots or anything)

1914 @ stock for me, boosts to 1215mhz

 

[Kaapstad]

i7 3960x @5.0

2 x GTX 690 in Quad sli

1680 x 1050

http://imageshack.us/photo/my-images/96/1680x1050fc.jpg/

1920 x 1080

http://imageshack.us/photo/my-images/201/1920x1080fc.jpg/

http://imageshack.us/photo/my-images/269/settings2.jpg/

http://imageshack.us/photo/my-images/846/settingst.jpg/

http://www.techpowerup.com/gpuz/6gepb/

I am running up the white flag to the red camp I surrender.

Is there anything else I need to include to get these on the score boards

Whats interesting is I seem to be hitting a wall just below 370 for max frames whichever resolution im running. I also noticed this with the run by piggin in the post earlier.

 

[Bradmax57]

It also depends on the stock boost of the 670, not all are the same, some windforces can vary up-to and over by 100mhz.

Yeah like my new one versus old, I wonder what can cause such a difference? Is it the chip or is the way the card is made?

 

[Tonester0011]

Thought i'd test my new 7970 before i do my watersetup at weekend, this was the card on the 24hr special.

[email protected]
7970@1260/1550

1680x1050.





http://www.techpowerup.com/gpuz/bm3k6/

 

[GoogalyMoogaly]

i7 3960x @5.0

2 x GTX 690 in Quad sli

...snip...

Interesting that 2 x 690s are so far behind 3 7970s.
I know Heaven seems to favour AMD cards and I know that multi GPU scaling gets worse as you add more cards, but still, that's a pretty huge difference considering the Nvidia setup costs ~£1680 and the faster AMD setup costs ~£900 (and also gives you an extra 1GB VRAM).

However I'm guessing if you have either of those setups you maybe well use higher levels of AA & AF than these benchmark settings and possibly at a higher resolution (multi-monitor?)
If run at higher settings does the Nvidia setup gain or lose ground on the AMD setup?
I guess this thread might not really be the place for it, is there currently a thread that deals with these really high end setups using the really high end settings?