• 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.

Screen goes black, GPU fans go loud, have to reboot

Ice213

Ice213

Ice213

Associate
Joined
30 Mar 2012
Posts
427
Hi everyone,

I've been having this problem off and on since just after I built my new PC. Randomly, every now and then, the PC screen will go black, and my GPU fans spin to (what I assume is) 100%, and I need to hold the power button and reboot the PC.

Now, I assume this is a graphics card issue because of the fans spinning up whenever it happens (although I could be wrong), and I have been unable to actually replicate it happening. I overclock it when gaming sometimes, and have gamed for really really long periods with no crashing. But other times (like just a few minutes ago), I'll be idling, just browsing the web, and the thing will go black.

When it happened a few weeks back I overclocked the card and stressed it in a hope of a crash (to try and find out if it was definitely at fault), and it survived that for a good few hours with no issues, and then survived more gaming straight afterwards with no problems.

I've got a 7970 with a custom cooler (which keeps it cool, even during all the gaming/stressing, so I don't think it's a heat issue), i5 3570k at stock, 8gb of corsair vengeance ram and a Gigabyte UD5H mobo.

If anyone could offer any help I'd really appreciate it :).

Thanks everyone! :D
 
Possibly a power issue.

In my experience with AMD cards the fans spinning up to full speed is an indication the card isn't getting enough power or a cable is loose or disconnected.

Check the power cables to the graphics card (both ends if it's a modular PSU).

Make sure they're firmly connected.

If it is a modular PSU you could try swapping the cables/PSU connectors if you have spares.
 
Last edited:
Thanks for the reply :).

I recently added a new PSU to my rig actually (ax850), but it did it a few times before with my old PSU too :(. I've just quickly pushed them all in and they all seem pretty secure.

Could the slots for the power cables on the card be a bit iffy at all?
 
I had this with my psu witch is a corsair hx850w semi modular. If I use the two pci per cables fixed inside psu, my screen will black out since changing them to another two out the box its never happened again.
 
shankly1985 said:
I had this with my psu witch is a corsair hx850w semi modular. If I use the two pci per cables fixed inside psu, my screen will black out since changing them to another two out the box its never happened again.
Click to expand...

I did try this, and it did it again, though it took a while, but as I said, seemingly completely at random times, so it's a pain to actually test if the problem is fixed because...well I can't test it :p.

todosmentira said:
This happened to me - caused by flash - the solution was to disable hardware acceleration in flash settings here:

http://helpx.adobe.com/flash-player/kb/find-version-flash-player.html

right click on the red flash version number
flash is officially ****e
Click to expand...

Hmmm, that sounds promising. I assume this doesn't disable flash completely or anything? I can't actually find anything about disabling hardware acceleration on that link you posted, will definitely give it a shot though cheers :)
 
Actually, I've found the hardware acceleration bit and disabled it, lets hope it works for me too. Just a quick additional question though, what does the hardware acceleration in flash actually do?
 
I have recently suffered this issue with a gtx670 ftw

It turned out to be because my modular power cable was very loose in the upper power block of the gpu.

I could recreate the problem by either giving the cable a bit of a wobble where it connects to the card or by just banging on my desk.
 
I started getting random black screens after replacing my Sapphire 6950 with a Sapphire 7950. It never happened when gaming or stressing the card. It always happened when browsing or running 2D apps or even just sitting on the desktop. Sometimes it resulted in a BSOD. Other times the screen came back after 10 minutes or so with a "Driver stopped responding" error. Obviously when the problem first started I didn't wait 10 minutes, I just hit the reset button. It was only after it had happened half a dozen times that I decided to leave it and wait to see if it sorted itself out.

Check your Windows Event Log. Every incident appeared in the Event Log (Windows Logs|System folder) as a "Display driver amdkmdap stopped responding and has successfully recovered" - Event ID 4101

I disabled the flash hardware acceleration but it made no difference.
I reinstalled the drivers (and cleaned up with Driver Sweeper) more times than I care to remember but it made no difference.

I only managed to fix it by creating a custom preset in the Catalyst Control Centre and setting the idle clock speeds/voltage (using Notepad to edit the xml file) to something slightly lower than the stock speeds/voltage.
It seems that (for whatever reason) the idle settings used by the CCC are too low for the card. Definitely an issue with the drivers. (And I have tried 12.3, 12.4, 12.6 and now I am on 12.8) Lots of people have had the same problem.
 
Last edited:
wapcatlet said:
I started getting random black screens after replacing my Sapphire 6950 with a Sapphire 7950. It never happened when gaming or stressing the card. It always happened when browsing or running 2D apps or even just sitting on the desktop. Sometimes it resulted in a BSOD. Other times the screen came back after 10 minutes or so with a "Driver stopped responding" error. Obviously when the problem first started I didn't wait 10 minutes, I just hit the reset button. It was only after it had happened half a dozen times that I decided to leave it and wait to see if it sorted itself out.

Check your Windows Event Log. Every incident appeared in the Event Log (Windows Logs|System folder) as a "Display driver amdkmdap stopped responding and has successfully recovered" - Event ID 4101

I disabled the flash hardware acceleration but it made no difference.
I reinstalled the drivers (and cleaned up with Driver Sweeper) more times than I care to remember but it made no difference.

I only managed to fix it by creating a custom preset in the Catalyst Control Centre and setting the idle clock speeds/voltage (using Notepad to edit the xml file) to something slightly lower than the stock speeds/voltage.
It seems that (for whatever reason) the idle settings used by the CCC are too low for the card. Definitely an issue with the drivers. (And I have tried 12.3, 12.4, 12.6 and now I am on 12.8) Lots of people have had the same problem.
Click to expand...

Driver stopped responding error I got this when I OC my card with not enough voltage. I was OC to 1000/1500 on stock volts and this would driver crash, I increased my voltage 6+ and it not happened since.
Same how you have lowered the voltage works both ways.
 
I had this exact same problem, my thread is on here somewhere, i had 6870s in CF i sent the card for testing and they were fine, i tested out my rig with a different psu and it was fine, i stuffed all my gear back in the case and it started doing it again.
So i was convinced it must be my psu right? NOPE, my mate used my psu for over a week with no problems, i was about to go mental and throw the whole rig out the window when i had an epiphany, what if its not the psu but the cables supplying the juice???
Now the side of my case where all the goodies are looked awesome, take the other side and it looked like a rats nest. I spent a quiet afternoon doing some serious cable management and i swapped out the four pci-e cables, i also downloaded msi afterburner and tweeked the settings, set msi and the overclock to start up on boot, disabled Overdrive in CCC and VOILA!!! never happened since that was three to four weeks ago.
Since then i have changed my 6870s for a 7950 and sold those cards to the same friend who tested my psu for a week and neither of us is having a problem.
Im not sure what the problem was but after doing all that stuff it seems to be cured.
 
nick.lidgett said:
I spent a quiet afternoon doing some serious cable management and i swapped out the four pci-e cables, i also downloaded msi afterburner and tweeked the settings, set msi and the overclock to start up on boot, disabled Overdrive in CCC and VOILA!!! never happened since that was three to four weeks ago.
Im not sure what the problem was but after doing all that stuff it seems to be cured.
Click to expand...
I forgot to mention that I also tried MSI Afterburner, set the Constant Voltage setting and StartUp on Boot then disabled overdrive in CCC. It made no difference. I still had the problem. Creating my own CCC preset/profile was the only way I could fix it.
 
Just registered to say I had some very similar issues with my HD 4870. Driver would often crash over and over again (4 times in a row), then would crash completely with black screen + 100% fan. I was thinking it might be the card since onboard graphics worked fine when card was removed but this makes me think it might have been the PSU and/or cables. Given wapcatlet also had similar issues with his 7950 (I was planning on getting an MSI TF3 7950 to replace the 4870 imminently), this really makes me hesitant to get one :( Would have thought AMD had got the hang of drivers by now especially after I had to manually delve into XML files to keep my 4870 from frying itself on release.
 
Thanks for all the responses!

Well so far since the hardware acceleration being disabled it hasn't happened (only been a day or 2, but still), but in the past it hasn't happened for a while and it just suddenly does it at a random point, sometimes gaming, sometimes doing nothing. I've been downstairs eating before, and come back upstairs and the PC sounded like a jet engine and it had happened when totally idle. So because of this I can't really say that it's fixed or not at the moment.

My cable management behind the nice neat facade is a bit of a mess, but all the cables are definitely securely in place, and the cables were swapped a few weeks ago because of this issue, but no luck in that department. And it has done it with 3 (I think) different PSUs. An XFX one (which I RMA'd for a separate issue), and 2 corsairs. I haven't disabled overdrive in CCC, and used to use afterburner (stopped because I thought it might have been the issue, but clearly it wasn't :p), but I will go back to using afterburner and disable overdrive in CCC, worth a shot :).

In terms of drivers, I did a reinstall when it first happened, and another one the other day, but until it happens again (or hopefully doesn't), I can't really tell you whether it helped :(.

I also don't think the cables are loose into the card itself, as wiggling and moving the case about has done nothing (while it was on obviously). However I did think about it possibly being the actual slots for the power cables on the card, although I would have no idea how to actually test that, I guess a wiggle of the wire would set it off again, which it hasn't, so it may not be that either.

And if it happens again (which again I will say, I hope it doesn't ;)), I will definitely leave it and see if it claims some driver issues and it comes back on. Voltage wise, I wouldn't know if that was the problem or not, I mean as I say, it seems totally random, so I can't effectively test to see if it is a voltage problem due to not being able to recreate the situation :(.
 
You must log in or register to reply here.
Back
Top Bottom