Bsod after playing demanding games for a while

vandatene · 09 Sep 2017

I've never done over or under clocking so that couldn't have damaged the GPU chip I assume? Is there a way to test any damage to the GPU chip or GPU at all with some kind of test ?

vandatene · 13 Sep 2017

It's been a few days. Could I maybe have a update? Would like to know if there is any way to test and find out if a GPU is faulty or damaged ?

vandatene · 17 Sep 2017

So I assume no-one has a answer for my question about the GPU testing ?

axe0 · 17 Sep 2017

Sorry for the delay, been busy past few days.

vandatene said:

The Prime95 test, does that test GPU? I've had twice now that For Honor crashed when alt tabbed. The popup said the GPU crashed, and I have never had this before. My GPU is only 1,5 year old, but since I have freezes in Ark Survival Evolved which others do not have, I'm considering a faulty or almost broken GPU ?

Prime95 tests the RAM and CPU, not the GPU.

To test your GPU, run Furmark.

Diagnostic test

GPU TEST

Run Furmark to test your GPU. Furmark tutorial

Note

Run Furmark for around 30 minutes

Warning

Your GPU temperatures will rise quickly while Furmark is running. Keep a keen eye on them and abort the test if temperatures rise too high

When finished, make a photo/screenshot of Furmark and post it.

In Furmark you can find a tool called 'GPU-Z' (if you want to install GPU-Z is up to you), when GPU-Z is open click on the 'Sensors' tab, on this tab you can monitor

the frequency of the GPU,
the GPU memory clock,
the temperature,
the power,
the load on the GPU,
Etc.

I would like you to make a screenshot of GPU-Z after you have either cancelled Furmark for any reason or Furmark is finished, post this screenshot in your next reply with the result of Furmark.

vandatene · 18 Sep 2017

I've run the test twice for at least 30 minutes. First time was with case closed, but since I've no more bsod's since several weeks (maybe because the weather/room temperature has been going down?). The second test was with the case closed. Also, with the second test I enabled the logging in a file (Didn't noticed it the first time). See attachments.

To me, everything looks normal. I've decided to keep the pc case closed again to try and get a bsod on purpose as it happens faster with a closed case.

GPU_Z allowed me to make a file that kept track of the status of several value's (GPU-Z Sensor Log_Case_Open in attachments). Would there be a program that would do the same, but then for CPU, GPU etc so if I get a bsod, we can see the values that are getting disrupted and most likely caused the bsod ?

axe0 · 18 Sep 2017

The results look good, let us know if you experience any more BSOD crashes :)

vandatene · 19 Sep 2017

First day with a closed case. Yesterday evening went fine. Today I crashed (bsod) 1 hour and 20 minutes after playing a demanding game (Ark Survival Evolved). Temperature is normal, and pc CPU temperature after restarting was only 50 degrees celcius. Error given with the bsod was WHEA_UNCORRECTABLE_ERROR. Something in my pc must be broken as it seems to be hardware related but I just cannot find it and all tests are looking fine except that Prime95 test has bsod's within a minute after the start.

axe0 · 19 Sep 2017

Please run memtest86+ again, the version you used is outdated.
Use Memtest86+ - Advanced Memory Diagnostic Tool to download the latest version, choose the downloader with "*NEW!*".

Crashes with prime95 may point to the CPU or RAM, with memtest86+ we should be able to rule it out or point it as the cause.
Could you try to run prime95 using small ffts and in-place large ffts.

Could you also check in the BIOS for the 3.3v, 5v and 12v and the CPU voltage.

vandatene · 19 Sep 2017

Thanks again for your help. It is very much appriciated.

I will run Memtest86+ all night, and test prime95 tomorrow morning.

I've already included some pictures from my BIOS with contact that looks important. They had to be uploaded in pairs of 2 or 3 else they were too big to be uploaded (limit is 5). About the screenshots, maybe there is something off which you can see instantly (also saw something with limits that was set on auto?). See attachments.

vandatene · 20 Sep 2017

And here are the other test results.

For Memtest86, I didn't have the time to do 8 passes. After nearly 6 hours, 3 passes have been done successfully. See attachment.

For Prime95, The small test went successfull. You can find the result in the screenshot (see attachment).

The in-place large ffts test went less successful. Within a minute in the test, mouse, screen, everything froze.
Seconds later I received a bsod with the message: CLOCK_WATCHDOG_TIMEOUT. As happened before, the
collecting information was stuck on 0%, even after waiting for nearly a hour.