• 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.

Why is AMD quiet about black screens on 290x?

Well had that round with you Matt on locker, then jumped on a Silk road server and got a blackscreen after about 6 mins.

clocks were 1150/1500 vcore +100mv.

Looking at my temps everything seems fine, core max was 82c, vrm1 72c vrm2 64c.

it's confusing but to me honest, at the moment it doesn't bother me to much, i only ever get that when pushing the memory to hard and as you say, outside of benchmarking, memory makes little difference to fps.

The card will happily bench at 1250/1625 and 24/7 oc is 1150/1400 which gives fps over 100 on all maps on BF4, so i'm happy for now.


Watercooling the lot on Friday ;)

Were you mate? I didn't even realise. Apologies if you said hello in chat but i didn't even realise.

I can't say for sure that this is 24/7 stable, but i have no reason to believe its not. I would take a guess that you either need more vcore, or to give aux voltage another bump. Or it could just be the fact that the memory can't go any further. After a while you reach a point of diminishing returns. However it does seem strange than my gimped 290's can run on much lower voltage. Both my cards are fairly high ASIC cards, 76.9% and 77.9% so maybe thats why.

You certainly have better temps than me. As you noted though, 1400vs1500 on memory clocks in game will make 0.1fps difference so what does it matter? You can bench 1625 (as can i) which will annihilate any bandwidth clock a titan or a 780TI can manage thats for sure.
 
Here you go:
ltlj.jpg

Are you really going to make me sit through 10 runs on the benchmark at 1080P :(

For the love of god man...
 
Just do 5. The 10 was just what I used the last time I was stress testing. The runs are pretty short anyway.

Where do i start, ok. I am so cpu bottlenecked this is not even funny. I need more demanding settings. For 15 seconds + of the benchmark both my gpu's are sitting at 0% usage. See the space between the black marks? 0% gpu usage. You need to let me use 1440P and crank up the settings as high as they can go (minus Physx) otherwise my gpu's are not going to put in any work. This is with my cpu at 4.8ghz. The creators of this game should be ashamed of its optimization performance.


I'm going to run it again now at max settings. Stand by.


YPfJ2gC.jpg.png
 
Just do 5. The 10 was just what I used the last time I was stress testing. The runs are pretty short anyway.

Where do i start, ok. I am so cpu bottlenecked this is not even funny. I need more demanding settings. For 15 seconds + of the benchmark both my gpu's are sitting at 0% usage. See the space between the black marks? 0% gpu usage. You need to let me use 1440P and crank up the settings as high as they can go otherwise my gpu's are not going to put in any work. This is with my cpu at 4.8ghz.

I'm going to run it again now at max settings. Stand by.


YPfJ2gC.jpg.png
 
lol you need to go to specsaver :D

Ok so i ran it again, no stability problems (gpu wise) but my psu crapped out so i had to knock my cpu down to 4ghz.

Completed the run ok. I did however have some artifacts, a few purple squares here and there so thats either a fault of the driver/benchmark or more than likely i need more voltage at these clocks. THe funny thing was BF4 had no artifacts.

I'm still cpu bottlenecked at this res and these settings. AAx4 +1440P would have helped. No black screens though. Stable!

2700k@4ghz
290 Pro Crossfire@1150/1500
13.12 WHQL


OWl2Xph.jpg



EDIT

Updated screenshot was missing some of the benchmark runs.
 
Last edited:
Were you mate? I didn't even realise. Apologies if you said hello in chat but i didn't even realise.

I can't say for sure that this is 24/7 stable, but i have no reason to believe its not. I would take a guess that you either need more vcore, or to give aux voltage another bump. Or it could just be the fact that the memory can't go any further. After a while you reach a point of diminishing returns. However it does seem strange than my gimped 290's can run on much lower voltage. Both my cards are fairly high ASIC cards, 76.9% and 77.9% so maybe thats why.

You certainly have better temps than me. As you noted though, 1400vs1500 on memory clocks in game will make 0.1fps difference so what does it matter? You can bench 1625 (as can i) which will annihilate any bandwidth clock a titan or a 780TI can manage thats for sure.

Yeah i was there pal :D like a ninja though lol.

Seems strange that it would be Vcore, +100mv can take me to 1220 without any artefacts etc, the only reason i've left it at +100mv was because temps don't seem a problem.

Like you say though it's not an issue for gaming and the card benches fine:D

Looking forward to watercooling it and really seeing what it can do.

I'll download Metro 2033 again now and test like you have.
 
Ok so i ran it again, no stability problems (gpu wise) but my psu crapped out so i had to knock my cpu down to 4ghz.

Completed the run ok. I did however have some artifacts, a few purple squares here and there so thats either a fault of the driver/benchmark or more than likely i need more voltage at these clocks. THe funny thing was BF4 had no artifacts.

I'm still cpu bottlenecked at this res and these settings. AAx4 +1440P would have helped. No black screens though. Stable!

2700k@4ghz
290 Pro Crossfire@1150/1500
13.12 WHQL


OWl2Xph.jpg



EDIT

Updated screenshot was missing some of the benchmark runs.
Just noticed...why are you using dx10? :confused:
 
Seems strange that it would be Vcore, +100mv can take me to 1220 without any artefacts etc, the only reason i've left it at +100mv was because temps don't seem a problem.
.

Memory clock is linked to vcore with Hawaii. Why this is i don't know.
 
Back
Top Bottom