• 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.

SLI

That is a shame you are having issues as you should be seeing some nice performance improvements running a second HD 5870.

Can I confirm that each of the cards are securely connected to two PCIE 6 pin power cables from the PSU?

Also, are you certain that you enabled CrossfireX in the catalyst control panel?

May I ask what game you are playing where you don't see any performance improvement?
 
Black ops / Bad company 2 / homefront / and others and all getting around 90 fps with or with out the second card ..

Yes i have crossfire Ticked of in driver settings .
Not really sure what kind of frames i should be seeing to be honest but i would think more then one card at 90 :( .
 
90fps is a really high framerate - especially for those games. I assume everything is very smooth, right?. Perhaps your CPU is bottlenecking performance (albeit at a high, very playable level).

If so, then I would whack up your graphical setting and see how the framerates fare.

I really should have asked this before, but what is the resolution of your main monitor?
 
1680x1050 I dont think CPU you bottlencking it ,Yeah just done test and put all on high in bc2 and there was no frames never drop what so ever infact the frames went up was getting 90 to 120 most of the time odd, But you think by setting all low just as test you would see massive jump in frames i never see any of that odd really .

But to be honest i feel alittle dispointed was hopping for more frames than that ..
 
Ah, well when you set the in-game settings to all low then the load is taken off the graphics card and it is able to render a much higher FPS - however, the load is still pretty constant on the CPU despite the lower resolution - so in the case where you drop the graphics settings then you will be encountering a CPU bottleneck (fortunately with your nice CPU that point is so high that it doesn't affect real-world performance on your 60Hz monitor).

Therefore with a crossfire system you don't want to test how good it is by turning down the graphics (as that introduces a CPU limit) instead turn up all the settings and see how lovely and smooth games play at really high detail setting (using high AA, AF). When you set AA to x4 the games really do look nice as you don't really notice the jaggies so much. Most of the time x8 AA isn't required, but you can see how it looks and what kind of performance hit you get.
 
Ah ok i see what your saying ,Its just when looking round the net i keep seeing people saying that there getting better fps in games and now brain is lock on that :D
 
Ok back yet again done some testing with GPU-Z running on another screen at the same time ,

And found that both cards are not getting fully used by the games that are running

Homefront for both Gpus hits 45% each and never went any higher .
BC2 Main Gpu hit around 75% to 85 % and second card only used 4% and no more .
Black ops both Gpus were around 69% and then i started getting flickering on screen so dropped it to window mode and flickering stopped ,but then i noticed that main gpu was doing all the work at 85% and the second one at 3% very strange that .

will test more games tomorrow .

And btw this is with all settings in games set to max
 
Last edited:
Ok back yet again done some testing with GPU-Z running on another screen at the same time ,

And found that both cards are not getting fully used by the games that are running

Homefront for both Gpus hits 45% each and never went any higher .
BC2 Main Gpu hit around 75% to 85 % and second card only used 4% and no more .
Black ops both Gpus were around 69% and then i started getting flickering on screen so dropped it to window mode and flickering stopped ,but then i noticed that main gpu was doing all the work at 85% and the second one at 3% very strange that .

will test more games tomorrow .

And btw this is with all settings in games set to max

hi ya, just a thought. im in the process of running two 6870's in cross fire on my alienware R3 desktop, and have done a lot of research,

it has been suggest that initially it is best to uninstall the graphics card drivers, and then use driver cleaner to make sure all evidence of the drivers are gone

then shut down the pc and install the 2nd card, then reboot and install the latest drivers, and go from there,
 
1680x1050.......

Someone should have asked about screen resolution before speccing a new card :rolleyes:

At that resolution a single 5870 is plenty, which is why you are seeing no increase in performance with a second card. You need to save up for a 1080p monitor (at least) to make the most of those cards. Also turn up all the settings to full detail. You've got the power for it :)
 
ok guys back again :D

Look at this video that i made

And look at the glitching well that glitching is in all my games when i have crossfire on ,With it off its fine or in window mode its fine ..

So what i am asking here is whys it doing this ? .

cheers
 
That is very strange.

Are the temperatures ok for both GPUs when under load (in a game)?

Also, new catalyst drivers (11.5) just came out - might be worth an install.
 
And look at the glitching well that glitching is in all my games when i have crossfire on ,With it off its fine or in window mode its fine ..

So what i am asking here is whys it doing this ? .

cheers

Have you installed the ATI Catalyst Application Profiles from AMD?:
http://sites.amd.com/us/game/downloads/Pages/radeon_win7-64.aspx

If these don't stop the xfire glitching then:

It may be a lack of power from your psu, or maybe the new card is faulty, as you said when xfire is off or in window mode(which also turns off xfire as it only works in full screen) it is fine.

First try taking your old card out and using only the new card and see if there is any glitching. If there is then the cards faulty, if not, then maybe psu is running out of power.

To test your cards you should try using something like heaven, you can get it HERE, it will give better results than any game as it will stress your cards more.

You can also install 'radeon pro' as this will give you on screen xfire usage while ingame or benchmarking, also lets you overide profiles for better performance, you can get itHERE

Have you installed the ATI Catalyst Application Profiles from AMD?:
http://sites.amd.com/us/game/downloads/Pages/radeon_win7-64.aspx
 
Xfire usage on screen display with radeon pro:

48221bc9693479107adddd81d03b402c.jpg
 
Windowed mode won't use xfire, so thats one problem out the question. That could be the reason BC2 isn't using xfire, though I can't recall if that game had xfire/sli issues to begin with? Haven't used xfire in a few months now.


In terms of power, just going back to what people were saying earlier, if you ran furmark or something it could potentially use close to 700W, but as you could see from the actual gaming power, it was down closer to 400W though that was only in Warcraft, I'd expect probably between that and 500W in "most" games.

Furmark is useless, its not even a benchmark its literally called a power virus by AMD and Nvidia, don't run it and your xfire setup won't come close to maxing out your PSU.

The one thing I'd suggest maybe is, other than whacking up graphical settings and seeing if performance stays the same(without xfire you'd expect them to stop) is to overclock your CPU if you hadn't. Most quads are MORE than capable for pretty much any gaming computer setup you can find, however at stock many aren't.

The thing is, the first Intel quads from 4-5 years ago now, 2.4Ghz stock is a little close for comfort and in quite a few games would be a limit, but at 3-3.4Ghz will basically get past the limit and most quads released since then can all overclock easily to what you'd consider "fast enough" speeds.

So you might find a small overclock could make a pretty big difference here.
 
Furmark is useless, its not even a benchmark its literally called a power virus by AMD and Nvidia, don't run it and your xfire setup won't come close to maxing out your PSU.
I only used the furmark picture for reference as it's the only decent picture I could find from google to show how xfire usage is displayed by Radeon pro.;)
 
Ok no matter what i do for games i get that glitching all ways in menu ,But when i enter the game no more glitching ,I have tryed all kinds of drivers and caps to fix that ,

I did try all so heaven benchmarking tool and both GPU's were flat out@99% and not one glitch in sight ,Out come score was from what i remember 1759 is that good i dunno i never do benching , Only when wight lifting :D

Any way all so tryed the radeon pro tool after tommyboy said about that ,And games was playing at about 50/50 on both gpus and no improvement on frames in sight stayed around 50 to 70fps in most games that i tryed .

I have never had crossfire or sli in my life so i dont no how games should be running and what frame i should be getting out of them ,And not sure if they should be both flat lineing in the game to give the most frames .I have no clue about all that .

But i can say one thing if i take out the second card and try one its just the same there is no benfits from having two ,Because the one card is flat lineing at 100% where as two cards are 50/50% .So whats the point me having two if i see no benfits from them i want to cry :(

I clocked my CPU to 3.4ghz upped the FSB to 400 from 333 .and that did not help :( i have no clue why the frame are not getting any better just that my pocket is lighter now :D
 
Last edited:
So your CPU is overclocked, and no difference. The PSU should be fine powering both cards.

I have noticed however the games you've listed are more CPU intensive. Maybe you should consider a CPU upgrade soon?
 
Back
Top Bottom