Black flicking screen and freezing

kailight

Standard Member
Joined
Dec 28, 2010
Messages
20
Reaction score
1
Points
6
Age
41
My PC seems to have developed a problem over the last 6 months of so.

My computer will randomly freeze, completely unresponsive and the only way to get out of the freeze is to force a power down by holding the power button. It is seemingly random, watching YouTube, playing a game or simply leaving my PC alone for a few hours and coming back to it frozen. YouTube seems to trigger a freeze more than anything but it is by no means consistent.

Once the PC has powered down, when I reboot my PC, it takes an age to log back in (I have an SSD so it's normally very fast and does load very fast when it's not recovering from a freeze) and will spend a number of minutes flicking between a blank desktop and a black screen, sometimes not stopping until a shutdown the PC (by ctrl, alt, del) and then reboot it again.

I have formatted my hard drives (both SSD and HDD) and installed a fresh copy of Windows 10. I have run numerous diagnostic software such as occt and CPU-Z which have found no errors.

Please see attached reliability log and dxdiag details.

What could be wrong with my PC and what can I do to try and troubleshoot/fix the issue.
 

Attachments

  • DXDiag Page 1.jpg
    DXDiag Page 1.jpg
    94.9 KB · Views: 55
  • DXDiag Page 2.jpg
    DXDiag Page 2.jpg
    91.8 KB · Views: 61
  • Reliability History Page 1.jpg
    Reliability History Page 1.jpg
    226.2 KB · Views: 55
  • Reliability History Page 2.jpg
    Reliability History Page 2.jpg
    105.7 KB · Views: 59
My best guesses would be:
Power supply failing - PSU checking devices are cheap and quick to use
Heat - clogged fans, CPU thermal paste replacement
RAM errors

How old is the PC?
Has it been upgraded, ie new GPU that can put more pressure on the PSU?

Mark.
 
Looks like RAM or the FX 8320, if you have any overclock on them I'd try removing it, after that then removing all but 1 RAM module and test again until you have tested multiple modules and both channels.
I'd also download afterburner ( or whichever monitoring software you like) to check the temps of the 8320 (and the 200 series GPU) but that doesn't sound like the issue based on the boot looping.

After that, it will be most likely the 200 series/7900 GPU, again remove any OC and test. If that gives no joy then use DDU and install and an older driver for the GPU from when you know it was working - if you have another GPU to test with you could eliminate it as a possibility. -- You could also try slightly underclocking the GPU as a lot of the 290/390(x) now seem to be happier at slightly lower voltages and speeds.

Saying all this if you are using a cheap 550w or less PSU or even a decent one that's 4-5 years old then that would make sense that its starving/overfeeding the system and causing these hard locks.
assuming the GPU is a 290/390(x) they should peak around 330w and the 8320 should peak around 240w.
 
My best guesses would be:
Power supply failing - PSU checking devices are cheap and quick to use
Heat - clogged fans, CPU thermal paste replacement
RAM errors

How old is the PC?
Has it been upgraded, ie new GPU that can put more pressure on the PSU?

Mark.
I had a new PSU professionally installed about 2 years ago as mine died on me. I am going to run an hour stress test on my PSU today but failing that, I will search for a PSU checker too.

At the same time of replacing my PSU, they applied new thermal paste as a bit of a service and very recently I have recleaned everything with compressed air and checked the thermal paste. I also did a visual inspection of my motherboard (I am by no means an expert but I checked for the obvious issues, blown transistors, burn spots etc etc) and the board looked fine.

Whilst using occt, an hour of stress testing my CPU, GPU (an hour each of physical and VRAM stress testing) and linpack caused no freezes and no error messages however on my first attempt at stress testing my RAM, it froze my PC within 10 mins. I forced shutdown my PC and stress tested my RAM again which caused no freeze or errors. I was hoping stress testing my RAM would cause another freeze to make it highly likely it was my RAM however the issue did not repeat itself. It may have been the 4 hours of stress testing leading up to my RAM stress test or it could have been my RAM. I also ran the RAM diagnostic tool inbuilt to Windows 10 Pro and that found no issues.

The PC is approximately 6-7 years old if I am remembering correctly. It's relatively old now.

The only thing changed is my PSU, everything else has been unchanged since purchase.
 
Looks like RAM or the FX 8320, if you have any overclock on them I'd try removing it, after that then removing all but 1 RAM module and test again until you have tested multiple modules and both channels.
I'd also download afterburner ( or whichever monitoring software you like) to check the temps of the 8320 (and the 200 series GPU) but that doesn't sound like the issue based on the boot looping.

After that, it will be most likely the 200 series/7900 GPU, again remove any OC and test. If that gives no joy then use DDU and install and an older driver for the GPU from when you know it was working - if you have another GPU to test with you could eliminate it as a possibility. -- You could also try slightly underclocking the GPU as a lot of the 290/390(x) now seem to be happier at slightly lower voltages and speeds.

Saying all this if you are using a cheap 550w or less PSU or even a decent one that's 4-5 years old then that would make sense that its starving/overfeeding the system and causing these hard locks.
assuming the GPU is a 290/390(x) they should peak around 330w and the 8320 should peak around 240w.
Everything is at base clocking, I have not overclocked any components.

I have core temps installed on my PC and it shows all temperatures to be within normal ranges (including when I am able to get prolonged use out of it without it freezing)

As my RAM is pretty old, I might buy a cheap 16gb stick and replace my old RAM to see if that makes much difference. I am in the process of getting a new PC built (aim is within the next 6 months) so I am loath to spend too much money on it. I want it working as I was going to give my PC to a friend so we could game together but as you can imagine, I do not want to spend hundreds on this to then spend a thousand on a new PC too.

I will get a DDU and see if I can get a driver from a year or so ago to see if that makes any difference. Naively, I assumed any unstable driver patches that could cause this would be patched out, I have had the issue for over 6 months and updated my drivers a couple of times so didn't think it would be a driver issue.

I'll also look at underclocking and seeing if that has any positive impacts to my issues.
 
Remove everything you can, so you have a bare minimum system (eg only 1 hard drive), and retry. Also, reseat all of the cards and power connectors. Check to make sure all of the fans are working. There is an error log file you can also review. Have a look here
 
I have completed an hour long PSU stress test, this coincidently stress tested the CPU and GPU at the same time. No errors or freezing.

I've attached some temps from just before it finished the stress test and about 3/4 minutes after to show temps during stress and after.
 

Attachments

  • Temps after ST.jpg
    Temps after ST.jpg
    130.8 KB · Views: 45
  • Temps ST.jpg
    Temps ST.jpg
    126.2 KB · Views: 41
Remove everything you can, so you have a bare minimum system (eg only 1 hard drive), and retry. Also, reseat all of the cards and power connectors. Check to make sure all of the fans are working. There is an error log file you can also review. Have a look here
When I cleaned my PC a few weeks ago I removed and re-seated all components.

Will look at removing a stick of RAM and my HDD and see if that makes a difference though.

I will also look at the link and see if that gives me any info
 
After I completed an hour long stress test of the PSU, I decided to do another RAM test and my PC froze again. So out of three RAM stress tests, two have caused freezes.

As suggested above, I will remove a stick of RAM, complete another stress test and then switch the RAM to see if it is one or both sticks potentially causing issues.
 
Remove everything you can, so you have a bare minimum system (eg only 1 hard drive), and retry. Also, reseat all of the cards and power connectors. Check to make sure all of the fans are working. There is an error log file you can also review. Have a look here
here a screen grab of the event log
 

Attachments

  • Event Log.jpg
    Event Log.jpg
    323.2 KB · Views: 44
I would say you have a hardware fault. Just start by removing and replacing what you can. Do it one piece at a time, otherwise you will lose track of what is good or bad. If you are unlucky you could 1 or more duff parts. I assume the faults occurred before you removed and re-seated everything?
 
So I started by removing one stick of my RAM and left my computer running with YouTube (usually triggers a freeze) for hours, no freeze. Loaded up a game and left it on playing for hours, no freeze. So I'm thinking, maybe the other stick of RAM is faulty and that was causing the freeze.

To test that out, the following day, I replaced the RAM stick with the removed one. Left YouTube on for hours, left a game running for hours, still no freeze.

Maybe it's a problem with dual channel on my motherboard? Pop them both back in to see if I can recreate the freeze.

I have not had any freezes at all since removing and testing my RAM.

So I have absolutely no idea what is happening or why but I am going to continue to monitor my PC and see if the problem suddenly reoccurs.

Thank you for everyone who has offered advice but it appears it has magically stopped.
 
Test your RAM with "Memtest 86" that will show up any errors with your RAM it takes a while to run though.
 

The latest video from AVForums

TV Buying Guide - Which TV Is Best For You?
Subscribe to our YouTube channel
Back
Top Bottom