• 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.

AMD ULPS Is Killing Multi GPU Performance & Smoothness

Caporegime
OP
Joined
12 Jul 2007
Posts
40,581
Location
United Kingdom
Amen to that. I love the idea of Zerocore, its superb. But the negative effect it has on performance and to a lesser degree smoothness means its a must for being disabled.
 
Soldato
Joined
7 Aug 2012
Posts
2,643
Matt, I get similar results to you on Sleeping Dogs at same settings on my 2 7970's, but I am not getting 100% (more like 60-70%) usage on both gpu's.

Now I know I am going to be somewhat bottlenecked due to running a 920@4ghz, but does that seem right to you? Just having a wee play around with my cf set-up and wondering why the disparity between our respective gpu usage?
 
Caporegime
OP
Joined
12 Jul 2007
Posts
40,581
Location
United Kingdom
Matt, I get similar results to you on Sleeping Dogs at same settings on my 2 7970's, but I am not getting 100% (more like 60-70%) usage on both gpu's.

Now I know I am going to be somewhat bottlenecked due to running a 920@4ghz, but does that seem right to you? Just having a wee play around with my cf set-up and wondering why the disparity between our respective gpu usage?

Tommy ran those benchmarks in the OP. He has a 3770k at around 4.5ghz and 2x7950's. Not sure if they were stock or overclocked though. Do you have Hyper threading enabled on your cpu? I would have thought you might be a slightly bottlenecked, but nothing drastic. If anything i guess maybe you might be a tad bottlenecked of Mhz speed. You can always try going further and see if usage improves. Generally though i find full usage on one gpu and lower usage on the other.

I will run a benchmark myself at 1440p and report back. Note at this res im much less likely to suffer any forum of cpu bottleneck. Saying that though i do have a 2700k so i do have faster IPC than you.

I'm currently running [email protected], ddr3 @2233mhz, BLK 104.7, and 7950's @1093/1563. Will report back shortly with some figures and see what my gpu usage is like with ULPS on.
 
Soldato
Joined
7 Aug 2012
Posts
2,643
Tommy ran those benchmarks in the OP. He has a 3770k at around 4.5ghz and 2x7950's. Not sure if they were stock or overclocked though. Do you have Hyper threading enabled on your cpu? I would have thought you might be a slightly bottlenecked, but nothing drastic. If anything i guess maybe you might be a tad bottlenecked of Mhz speed. You can always try going further and see if usage improves. Generally though i find full usage on one gpu and lower usage on the other.

I will run a benchmark myself at 1440p and report back. Note at this res im much less likely to suffer any forum of cpu bottleneck. Saying that though i do have a 2700k so i do have faster IPC than you.

I'm currently running [email protected], ddr3 @2233mhz, BLK 104.7, and 7950's @1093/1563. Will report back shortly with some figures and see what my gpu usage is like with ULPS on.

Cool no worries, like I say still experimenting a wee bit with various settings in game(s) and with radeonpro. I would say I'm reasonably pleased atm. Should really just play some games lol.
 
Soldato
Joined
7 Aug 2012
Posts
2,643
I made mistake anyway, didn't have ambient occlusion on high - that increased my gpu usage although my results decreased quite a bit. Still aroundabout the ballpark of tommys results.

Still wasn't getting constant 90% upwards gpu usage though. My max fps was higher than 120 though which I thought was strange since I have vsync set to always on via radeonpro? :/ Got a max of 145 - an anomaly?

Edit just to add hyperthreading is enabled. Old cpu, so to be expected that it will be weakest link at the moment.
 
Associate
Joined
16 Aug 2011
Posts
1,532
Location
Ireland
So I dont have a 120hz panel so I couldnt check if op was running at stock or not, so I ran 2560x1440 on extreme with my cpu at 4.6 and the gpus at stock (didnt have cpu at stock in case ti bottle-necked the crossfire) with ULPS disabled:

 
Soldato
Joined
13 Aug 2012
Posts
4,277
I really can't understand why AMD has crap like this enabled :confused:

Same sort of thing with me having to setup up every game in radeon pro and turn the flip queue size to 1 and limit the frames.



Anyway thanks for the info and I'm glad AMD have improved on latest drivers. I think it's about time for me to get out my other 7950 and stick it in my pc and try out some crossfire :)
 
Soldato
Joined
30 Mar 2010
Posts
13,059
Location
Under The Stairs!
I would like to see AMD give an option in CCC to disable ULPS. I also believe AMD should reccomended all review sites to disable it before benchmarking their multi gpu solutions. Single gpu's are unaffected.

Thanks to Tommy for compiling some results for me.

Np, matt, and good thinking with the thread to get it out there for all the AMD brothers to eek even more performance out of these monsters.:D

Is AMD aware of the ulps dilema in terms of the hit ulps is causing in regards to SLi/CrossFire shootouts for reviews?

On the one hand, it's out the box that counts to the review sites, but if AMD were to give the option to disable ulps(preferably a right click in the system tray if your listening Thracks) and put it down on their testing methology requirements then, evidently, there is a lot more performance available in the tank over out the box settings as you will see with my TR testing below(although gains would be less@stock).

Tommy ran those benchmarks in the OP. He has a 3770k at around 4.5ghz and 2x7950's. Not sure if they were stock or overclocked though.

[email protected] 7950 CrossFire@1100/1350/+35% in those benches.

SD +~17% average and +~14% minimum difference is quite a hit though with ulps enabled.

TR FULL MAX on the above clocks:

d5428c8896719d87653802f8100ebb22.jpg

+~17% max and +~16% average.

BUT, I also tested 3DMark with ulps on/off and there was zero difference, I don't know if 3DMark has been given the frame pacing treatment though in the 13.8 beta.

I really can't understand why AMD has crap like this enabled :confused:
It's an AMD usp, a very good one at that too when you take the performance hit out of the equation.

It must have been a god send for AMD users running their rigs 24/7 through these very hot weeks with their gpus depositing zero waste heat not adding more ambient temps to their living space, not to mention the noise(if that grates you like it does me).

Then think CrossFire, that second gpu dumps zero waste heat when your on the desktop browsing working etc.

I dread to think what the temps would have rocketed too if I had my rig belting out both gpu's on high clocks struggling with temps when I was in Italy considering our heatwave is nothing in comparison to temps there.

It was bad enough over there watching a movie at night with a set of cans on your lugs lying back sweating like a mad ****** as the temps don't let up even at night.:eek:
 
Associate
Joined
16 Aug 2011
Posts
1,532
Location
Ireland
[email protected] 7950 CrossFire@1100/1350/+35% in those benches.

Cut your post down to the bit I'm interested in lol, so you're running at +35% power, so modified bios I'm guessing, do you need the +35% power for your clocks? did you have issues when at 20% for example? I'm running into a problem with amd driver not responding etc atm, which looks like its not enough vcore, my first card I know isnt the problem as I havnt had this problem in the year ive got it, and I've slowly been down clocking my 2nd card (down to 1100 and 1500 on it so far) even though I ran through bioshock and then benches (1150/1575) without problem using only that card and have been considering running a modified bios on it (but not too comfortable with it hence why ive just been slowly down clocking the card lol)

And I'm surprised 3D Mark showed no difference seeing as theirs quite a big difference in game, but as stated it could be something with the new beta who knows
 
Soldato
Joined
30 Mar 2010
Posts
13,059
Location
Under The Stairs!
Been running higher power limits since I found out how to apply it due to the boost bios being a pile of ****, if you don't want to flash your bios, you can hack the power limit and it will keep your clocks from spiking, which can give quite a hit to fps, details here:

http://forums.overclockers.co.uk/showthread.php?p=24212468#post24212468

It's very simple, I was just thorough with the instructions.

It's maybe just a lazy card though as the voltage is 1.25 in the first place.

Bios flashing is where it's at, there is an Ice-Q tbios thread here somewhere, maybe somebody can put up the link.

Btw, put your better gpu in the top slot for better performance in non CrossFire titles-if you play any.
 
Back
Top Bottom