** Single, Dual & Triple Card 7970 - BATTLEFIELD 3 - HIGH RESOLUTION BENCHMARKS **

[rjk]

Yo!

Been thrashing these new cards today, Heaven benchmark scores are all well and good but what we ALL want to know is how these babies run BF3!

When all said and done, there isn't any other game worth playing so...

X79 motherboard
8 Core 2011 Xeon with HT
16GB Dominator GT 2133
1200W Antec PSU

benchmarks taken in game with fraps
60 second recording time
first person gameplay [not using vehicles]
Map Played - Caspian Boarder
min,max,avg recorded


Running at 2560X1600 with all settings on maximum.

System was quiet using a single card and fairly quiet using two cards but very loud with three cards.
being used to water-cooled graphics cards, I was shocked at how quiet the stock cooler is. I really like the way that the idle cards turn off when on desktop too.
That, is smart.

Another thing that was surprising was the smoothness of the game, even when fraps was reporting the fps at below 40
Tri fire decimated everything though. as we expected.

I did three runs at 2560X1600 with one, two and three cards

I then did the benchmark on a single card at 1080p as well using the same settings.

here are the results:

GTX 580 3GB Testing with a marginal overclock [similar to the clock on the 7970s, slight increase of the speeds on stock volts] 850 core on both cards.

 

[rjk]

Hi Guys

we only have limited time for our benchmarking runs.

we have done 2560X1600 as that is what this card is designed to work on.

I have done a 1080p benchmark on the single card as that will be what most people have done.

I haven't ran any power usage tests.
1. I don't have time
2. There are plenty of official reviews out there with indications of power usage
3. If I was buying a high end flagship GPU, then I probably wouldnt be caring too much about my electricity bill [buy a Corvette, ask about fuel efficiency]

 

[rjk]

Please show 1680 by 1050 on ultra please!!

not sure if serious...

you read the part of the title that said "HIGH RESOLUTION BENCHMARKS" in big bold capital letters, right?

 

[rjk]

27" Hazro WC 2560x1440?

that's not high resolution

 

[rjk]

 

[rjk]

erm at 1080p a 6970 get same or higher than single card in your benchie using same settings.

link?

same settings?

HBAO?

 

[rjk]

i run a GTX 590 at 2560X 1600

using all the settings as above but with HBAO off and 2X aa and 2X af

this card is about equal to it performance wise, the 590 does pip it slightly though.

 

[rjk]

Hi

Interesting,
No reason not to trust your results but a few quires,
You state "Map Played - Caspian Boarder" it was. thats where all the benches were taken.
But in the screen shot its clearly "Strike at Karkand" i was messing around trying to find a server i liked to test in. just took the screeny there to show the settings
Also you state first person gameplay [not using vehicles]
but in the 2nd screenshot your clearly in a jet/heli that was just me flying around again, noticed a free jet and wanted some kills on my stats
Also you missed the ssd you used from your system spec list presuming you used an ssd yep, standard crucial 64gb m4
1 last thing, single card max fps 41 while crossfire is more than double at 86fps.
Single card performance at 1080p does look abit disappointing but hey, thanks for taking the time out to do this.

no problem, playing bf3 on unreleased hardware is a really boring job but i know its for the greater good

GTX 590 is VRAM limited as has been noted before - 2 x 1.5G versus 3G at this res is definitely going to have an effect. Crossfire 7970's versus SLi 3G 580's would be a much more direct comparison, which in turn would give people seriously considering 2 of these cards (such as myself) a good idea where to plonk our hard earned. Can you do it?

it is but in fairness my ram usage never goes above 1100mb. sure, 3gb useable would be epic but i dont think it would be entirely game changing.
the only thing i have the option to increase on my system is AA & AF and lets face it, they make little to no difference on that res.

HBAO is a big difference though and i cant get that running on my 590 whereas this card has it enabled with ease, everything looks better too.

GTX 590's are VRAM limited at 2 x 1.5G versus a 7970 with 3G so it's not really a very good comparison at all, *especially* at 2560 x 1600 resolution.

as mentioned above, my 590 doesn't use more than 1300mb of VRAM unless i put more than 2XAA and 2xAF on. I can see the need for the 3GB card, and for eyefinity it will be epic, but its too close for me to consider and an overclocked 590 leaves a single 7970 for dead

Im not selling the card [thats Gibbo's job] and I'm not an nvidia fanboy. so im posting from as much of an unbiased POV as I can

 

[rjk]

also how many people were in each server, its no good running round with no action to test

server was 40/60
benchmarking was activated when I was around capture points so quite a similar action scenario on each run.

its impossible to be subjective without a proper time demo for the game but i did the best I could.
I had my smoke launcher on

still have a bit of work to do with the 7970s before we test the 580s [Gibbo is testing the system on Quake 3]

and Cupra, you are not allowed to buy 7970s because I refuse to listen to you moaning about the drivers in vent

 

[rjk]

cards were at

1100 core
1450 on the mem

these are a smidge above stock tbh. not put them higher. grand scheme of things? this will make about 2-3fps difference.

580 3GB edition comparison coming tomorrow

 

[rjk]

thanks

no worries babe

 

[rjk]

only gone and done some testing with a pair of 580s haven't I!

 

[rjk]

graphs are up

bit of a rush job but I havent got any more time to dedicate to it today guys.