***Official Unigine Heaven 4.0 benchmark scores! Thread***

[BallaTheFeared]

Powerful card.

Nvidia has considerably more tess power than AMD.

 

[futurefields]

yeah ive noticed, even with my 7950 at 1050mhz I dont use tessellation at all in Arkham City.

 

[aarontpx]

gtx 680 - 1280/1829

http://imageshack.com/photo/my-images/339/heaven40c.jpg/

 

[Face2Face]

Nvidia has considerably more tess power than AMD.

Balla, here is yours from the other thread if you want to post it as well?

http://forums.anandtech.com/showpost.php?p=34867950&postcount=59

 

[futurefields]

Updated.

Would like to get more scores from the enthusiast Nvidia cards 660, 660 ti, 670 as well as some more AMD crossfire scores.

 

[MTDEW]

I actually run these daily at 925/1250 @1.044v which has proven to be more than enough for any game for me, and keeps the temps low.


Not sure what they will do if i really push the clocks/voltage and temps,.
Someone else with dual 7950/7970's can do that as i'm not willing to go above the default 1.25v since i think its too much to begin with.

Note: They are both SAPPHIRE 100352-2L boost cards so default is 1.25v, so i use Trixx to lower the voltage for daily gaming.

This was just for kicks to see if i could catch the other posted dual GPU scores.
7950 x2 @ 1175/1450 @ 1.25V (+20% PT so no throttling)

 

[(sic)Klown12]

Here's my X6 @ 4Ghz and 7970 @ 1150/1600.

 

[Fastx]

I actually run these daily at 925/1250 @1.044v which has proven to be more than enough for any game for me, and keeps the temps low.


Not sure what they will do if i really push the clocks/voltage and temps,.
Someone else with dual 7950/7970's can do that as i'm not willing to go above the default 1.25v since i think its too much to begin with.

Note: They are both SAPPHIRE 100352-2L boost cards so default is 1.25v, so i use Trixx to lower the voltage for daily gaming.

This was just for kicks to see if i could catch the other posted dual GPU scores.
7950 x2 @ 1175/1450 @ 1.25V (+20% PT so no throttling)

Nice scrore with those 7950's on this test!

MTDEW can you tell me at you stock speeds 950/1250 @ 1.044v CF what are GPU temps on the top card under load?
Also do you have 1 or 2 pci slots between the top & bottom 7950 cards?

TIA

 

[MTDEW]

Nice scrore with those 7950's on this test!

MTDEW can you tell me at you stock speeds 950/1250 @ 1.044v CF what are GPU temps on the top card under load?
Also do you have 1 or 2 pci slots between the top & bottom 7950 cards?

TIA

I'm running the ASUS P8Z68-V PRO mobo, so I have two slots between the cards.

I'm also running a CoolerMaster HAF X Case which has a side GPU fan as well as a GPU fan between the HDD cage and GPU's that has a shroud that directs air over/between the GPU's. (note: in the newegg video, the fan isn't mounted in the shroud, but you can see the shroud and where it goes at 4:07 in the video)

Give me a sec, and i'll run the heaven benchmark here quick and post a pic with temps for ya. :thumbsup:
I usually run a custom fan profile that basically scales 10% higher than the card temp.
ie: 50c = 60% fan spd / 60c = 70% fan spd etc....etc..

Do you want me to use my usual fan profile, or just the Sapphire defaults?

 

[MTDEW]

Nice scrore with those 7950's on this test!

MTDEW can you tell me at you stock speeds 950/1250 @ 1.044v CF what are GPU temps on the top card under load?
Also do you have 1 or 2 pci slots between the top & bottom 7950 cards?

TIA

Info is in the top left of pic.
Info goes like this ...
GPU temp / GPU usage / fan speed / core speed / core voltage
Then below is memory spd, voltage and mem usage.

Here is the info with my fan profile. (925/1250 @ 1.044v)

Same run with no fan profile. (stock fan speeds 925/1250 @ 1.044v)

And since I know someone is going ask eventually, here is the info at 1175/1475 @ 1.25v



I tested both of my cards individually when i first got them at 1200/1500 and both passed with flying colors and were in the mid 60c range @ 1.25v.
But once I x-fired them, i had to back both the memory and core speeds down a notch for stability. (yes i tested mem and core separately)

Even now, if i disable x-fire, a single one can do 1200/1500, but once x-fire is enabled it needs to be backed down for stability.
(and i refuse to add more voltage for fear of damaging my cards just for a benchmark score)

So, going by others overclocks, my memory on my cards aren't the greatest OC'ers (far from it) and my cores are about average.

But, as i said, I really have no need to max the the card's overclocks because my x-fire performance is fine by me at 925/1250 @ 1.044v
If i ever feel the need to push the cards, from my notes, I'd probably run something like 1125/1450 @ 1.188v since that makes the top card max around 73c under load during benching.

 

[Fastx]

Wow that is the best setup/case I have seen to date for running SLI/CF on air w/stock coolers you really have this dialed in. Great temps too, but kind of defeats what those temps are to a degree for me now due to that setup But thanks for all the info/test/temps I appreciate it very nice setup & score with those 7950's.

 

[MTDEW]

Here is a pic of my setup I took when i first built this pc.
Back then i was running two 6950's (unlocked)
But my setup is still the same now just with two 7950's instead.
You still can't see the GPU fan in the shroud, but trust me, there is a fan in there blowing over/between the cards. :biggrin:

 

[Fastx]

Very nice you got me oohing & aahing over here with that case I even saved that pic to my desktop thanks for posting a large pic. When I was looking at the video it look like the cards slide into the shrould but I wasn't positive and how far till you just posted this pic. That case is at the top of my list if and when should I decide to get a new case.

Pretty cool thanks for the pic.

 

[RadiclDreamer]

My sons dual 6670 1GB cards rocking 682 with a 3570k @ 4ghz

 

[AndyE]

rgds,
Andy

 

[MTDEW]

rgds,
Andy

No wonder the lights in my house just went dim, someone is hogging all the power. :thumbsup::thumbsup:
Your GPU's are worth more than my entire setup.

 

[futurefields]

Ding ding ding we have a winner!

 

[cmdrdredd]

If I had unlimited disposable income I could post that too lol. We all could.

What surprises me is the minimums are still 32.

 

[BallaTheFeared]

Given how the bench works, it doesn't surprise me at all.

 

[Kenmitch]

This is the best I could pull with my His IceQ 7870. My card likes the HIS software more than Afterburner for some reason.



Temp topped out at 61C during my runs.

 

[futurefields]

+12 board power? 61C?

That card has a lot more juice in it my friend

 

[cmdrdredd]

Given how the bench works, it doesn't surprise me at all.

You can cycle through the scenes, then run it that eliminates any delay between.

 

[Durvelle27]

This is the best I could pull with my His IceQ 7870. My card likes the HIS software more than Afterburner for some reason.



Temp topped out at 61C during my runs.

i'm coming for you

 

[Kenmitch]

+12 board power? 61C?

That card has a lot more juice in it my friend

Well you'd think but my core doesn't agree. The + power doesn't seem to do anything drastic for my chip at least. I just feel kinda obligated to move the slider as I crank the clocks up.

This looks to be maxed out now



Cooler must be as good as they say as temps peaked at 64C this run.

i'm coming for you

May the overclocking gods smile upon you

 

[BallaTheFeared]

You can cycle through the scenes, then run it that eliminates any delay between.

You can, though we don't know if he did... You can also restart it over and over after looping it several times and get different results on each startup.

There is also the tessellated hallway where my gpu usage would drop considerably with a highly clocked i5 with just 470 SLI.

Could have just been a hitch off the stairs coming out of that room as well, who knows, but I take nothing from Heaven mins, and even as is it's nothing more than a ROP/Tess benchmark with overkill AA and overkill tessellation.