• 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.

Black Screen issues

humbug

humbug

humbug

Caporegime
Joined
17 Mar 2012
Posts
49,766
Location
ARC-L1, Stanton System
Powercolor PCS+ R9 290, 18 Months old.

I have never had any Black Screen issues with this card in the past, over the last 4 to 6 weeks its become a problem.

Whats happening is the screen will go black and then into standby, the computer is also frozen and i have to hard reset, alarmingly for about 1 in 10 of this happening i get a Motherboard error beep and a GPU error light. when that happens it will keep doing it until i physically remove the GPU and put it back, after than it will start again and run fine.

It Black screens like this very randomly during gaming, at idle and browsing, it doesn't happen often.

Any ideas whats wrong?
 
CEUOTC said:
Had something very similiar with my first 290, had to RMA it in the end. Could be GPU Vram issue?
Click to expand...

Its clocked at 1350Mhz out of the box, it did it at 1300Mhz. i might just test that properly by running it at 1200Mhz?

I don't want to RMA it yet because because it doesn't happen often and i can't get it to do it by doing anything specific to it, like stressing it. it just does it at random.
 
This is the most worrying thing to me, has anyone else with black screen issues also have this happen to them?

humbug said:
the computer is also frozen and i have to hard reset, alarmingly for about 1 in 10 of this happening i get a Motherboard error beep and a GPU error light. when that happens it will keep doing it until i physically remove the GPU and put it back, after than it will start again and run fine.
Click to expand...
 
Have you tried removing the overclock altogether to see if that helps? that would always be my first guess in the middle of summer.

Having to remove the GPU before it will work again seems a bit power supply related to me though, it sounds like there's some sort of electrical protection mechanism kicking in. That could just be a result of crashing/faulty card though.
 
Kelt said:
Have you tried reseating the card and the pci connectors, just to rule it out?

Is your PSU OK as well?
Click to expand...

Many times yes, sometimes i have to take it out and put it back in the get the PC to start again, but its doesn't fix it.

mmj_uk said:
Have you tried removing the overclock altogether to see if that helps? that would always be my first guess in the middle of summer.

Having to remove the GPU before it will work again seems a bit power supply related to me though, it sounds like there's some sort of electrical protection mechanism kicking in. That could just be a result of crashing/faulty card though.
Click to expand...

I'm now running the Memory at 1200Mhz, if it does it again i will try at 1000Mhz.

I'm thinking PSU could be the problem but i'm no more sure of than that i am Motherboard or GPU, the PSU is a 4 year old XFX 750 Pro.
 
I had an issue with similar symptoms with a 7850 (black screen under load, screen goes to standby, hard reset needed).

I could stop the black screen from happening by massively underclocking the core (to around 550MHz), but ultimately the card was RMAd.
 
Sounds like for the idle and browser black screening, that it's unstable when switching between the p-states of idle to 2d clocks. Hence under load stress testing it won't do it as you are constantly in 3d clocks with enough voltage.

As the memory will bump up from 150mhz to 300 to 1300 for your card.
The voltage for the core and the memory is shared and maybe as it cycles through the states it doesn't have enough voltage for the memory when switching.

As for locking up in gaming, is your cpu still stock due to the cooler rma.
If so the gpu might not being fully utilised now that your cpu isn't clocked.
So it might not be holding the voltage as reliably as if it was being used to it's fullest.

I havent played with Hawaii so I don't know if there's ways of modding the 2d/3d clocks and voltage like you can if you play with Vbe for example
 
A batch of these cards were very suspect at doing this OP. No matter what you do/replace it will probably still happen occasionally very difficult to reliably reproduce.
 
Yeah i have two duff 290x myself. If i try and overlcok the memory at all they black screen :( shame as they are both under water and the core clocks fine.
 
I'm running the memory at 1200Mhz, it seems fine so far but it will take at least a week for me to know for sure as it can be days between it working fine and then not.

I will give it a week, if it doesn't crash i will put it back to its 1350Mhz, if it does it again at that i will contact the retailer.

I have also striped the computer down, cleaned it up and had a close look, everything looks in good order, i also took the PSU apart, cleaned that out, again it all looks healthy.

I don't get it, i have had it for 18 Months and its been solid, only now is it doing this.
 
I would try a more recent bios for the card. I would only get black screens if I overclocked the memory to far and the benchmark would randomly black out causing me to need to hit the reset button. Newer bios version just gives be a bit of gfx corruption instead telling me I need to back off so maybe the latencies have been tweaked for more stability with the memory in newer bios versions.

This is the one I am using https://www.techpowerup.com/vgabios/162939/powercolor-r9290-4096-140814.html which is the last one supporting elpida ram as far as I can tell.
 
MadMatty said:
I would try a more recent bios for the card. I would only get black screens if I overclocked the memory to far and the benchmark would randomly black out causing me to need to hit the reset button. Newer bios version just gives be a bit of gfx corruption instead telling me I need to back off so maybe the latencies have been tweaked for more stability with the memory in newer bios versions.

This is the one I am using https://www.techpowerup.com/vgabios/162939/powercolor-r9290-4096-140814.html which is the last one supporting elpida ram as far as I can tell.
Click to expand...
Thats a 15.047 BIOS, a lot newer, i'm running a 15.042.

How old is your PCS+? they got a revision about 8 months ago, mine is the old PCS+ and i don't think the Rev-2 BIOS works on mine.

For example the Rev-2 PCS+ has 6 Phase VRM's to my 5.
 
I've had my card a little over a year. I will have to pull the card out sometime to see if I can spot any revision numbers without taking it apart.

Edit: Without dismantling the card, all I can see is 'LF R9FA ver 1.0 4g' scribed above the pcie header.
 
Last edited:
4 Days trouble free in Cryengine at 1050/1200, clocked it up to 1250Mhz Blackscreen.

The problem is it seems to be fine at its factory clocks in gaming and benchmarking apps, apart from that one time in world Of Warship's, so outside of Cryengine its extremely difficult to get it to Blackscreen.

 
You must log in or register to reply here.
Back
Top Bottom