Dec 30, 2020
17
0
10
Fyi, this is a long story. Ever since I had my system I had bsods with the stop code SYSTEM_THREAD_EXCEPTION_NOT_HANDLED. I do have to mention that I did drop it once in the process of building a the pc. It was dropped from a typical linnmon with adils legs table, flat on the backplate. It went away after a few hours and I been enjoy my system except for the few times my games crashed by either the colors changing into a some sort of colorblind colors and freezing and restarting, or having squares on my screen and rebooting. It was mostly Overwatch and I also got errors Rendering device lost! on it. It didnt happen in practice ranges or anything, but as soon as I got in a game I crashed(if it crashed). Usually after the reboot the game would work fine for the rest if the day.

I need to note that I usually had MSI Afterburner minimized for the fan curves. One day I had the now-typical color shifting crash but this time, the freeze before the reboot was longer and when I tried to open any launcher except rockstar, they would just not recognize the internet even though I was able to play an online game and chat on discord. It was the same for the radeon and windows update checkers. Ive chatted with multiple people on forums and the only option was to reinstall windows. I did reinstall it and everything was fine for a few minutes and then I got a bsod loop where I could only access the pc for a few seconds before bsod again. The stop codes were mostly SYSTEM_THREAD_EXCEPTION_NOT_HANDLED what failed: cl.dll(not ci) but there was a CRITICAL_STRUCTURE_CORRUPTION what failed: cl.dll.

I reinstalled it again and this time it was fine until I installed my chipset drivers. In the restart after the chipset drivers, same bsod loop again. Kept happening for multiple reinstalls of windows. I took multiple troubleshooting steps in safe mode such as sfc /scannow and dism commands. Sfc did find corrupt files and fixed it but the bsod kept happening after it. Dism said system health was fine.

Someone said I might need to flash bios so I did to the newest non beta verison(aka the one he told me to download)(another note all the people I talk to are from the coolermaster/nzxt discord server) still kept happening. The only things i downloaded during these reinstalls are discord steam and malwarebytes(only once, i removed it) Weirdly though it didnt crash in safe mode. I thought a 3rd party software was interfering but at one point it was only the chipset drivers and it bsod. I finally decided to not install the chipsets and downloaded the gpu drivers and other utilities. I did noticed that as soon as I finished finish download for msi afterburner, it bsod again. Also it bsod a couple times while downloading gta v on steam.

The pc was fine until I noticed that on some play sessions(much like the other crashes) the game would just freeze, there would be static sound(bzzzzzz) in my headphones, and the game would restart. I noticed it on both games I had installed so I reached out for help and someone said to run OCCT to see if this issue persists here. I already ran gfxbenchmark and it ran fine. So I download OCCT and as soon as I hit benchmark the pc goes back into the bsod loop. He told me to run memtest86 and I got tons of erros on the hammer test only. He linked me a forum saying that if I only get errors in the hammer test then my ram might be fine. I told him the shortened story starting from the big crash and he said the symtoms points to my gpu. He also said that the bsod seemed to be mentioning OpenCL. I did what he said and I sent my gpu out to rma but im now scared that the issue might happen even after the rma.

Whats wrong with my system? Was the guy right or wrong and is there anything that I missed? Im also curious what could be the cause of this if its not the gpu or ram. At this point I dont know what the problem is. My brother has almost the exact same rig(except the his gpu is the sapphire 5600xt and was not dropped) and his system is working smoothly ever since the build date with 20.9 drivers and the factory bios settings.

System specs:
R7 3700x
32gb Team T force rgb ram
Msi gaming x 5600xt(i built it in the worst time possible so yeah)
Team t force vulcan 500gb ssd
Enermax rev d f 650w
A 2 tb seagate hdd
 
Solution
He told me to run memtest86 and I got tons of erros on the hammer test only. He linked me a forum saying that if I only get errors in the hammer test then my ram might be fine.

its not windows, if you get same errors on multiple installs, you can't be that unlucky to get same software every time.

How many errors in Memtest as max number you want is 0. Any higher is too many. I would have replaced ram before thinking its GPU.

cl.dll isn't opencl. it stands for Code Integrity, which makes sense if you got a CRITICAL_STRUCTURE_CORRUPTION error.
bad Microsoft having 2 dll that look the same in text. cl.dll is open dl.

OCCT is for testing overclocks, and among other things, it hits ram hard -...

Colif

Win 11 Master
Moderator
He told me to run memtest86 and I got tons of erros on the hammer test only. He linked me a forum saying that if I only get errors in the hammer test then my ram might be fine.

its not windows, if you get same errors on multiple installs, you can't be that unlucky to get same software every time.

How many errors in Memtest as max number you want is 0. Any higher is too many. I would have replaced ram before thinking its GPU.

cl.dll isn't opencl. it stands for Code Integrity, which makes sense if you got a CRITICAL_STRUCTURE_CORRUPTION error.
bad Microsoft having 2 dll that look the same in text. cl.dll is open dl.

OCCT is for testing overclocks, and among other things, it hits ram hard - https://www.tomshardware.com/reviews/stress-test-cpu-pc-guide,5461-3.html

i would fix the ram situation before you really can believe any bsod you get as bad ram can cause perfectly good drivers to not work.

only way to know if its GPU is either test GPU in another PC or use another GPU in yours, since ryzen won't work without GPU
 
Last edited:
Solution
Dec 30, 2020
17
0
10
its not windows, if you get same errors on multiple installs, you can't be that unlucky to get same software every time.

How many errors in Memtest as max number you want is 0. Any higher is too many. I would have replaced ram before thinking its GPU.

cl.dll isn't opencl. it stands for Code Integrity, which makes sense if you got a CRITICAL_STRUCTURE_CORRUPTION error.
bad Microsoft having 2 dll that look the same in text. cl.dll is open dl.

OCCT is for testing overclocks, and among other things, it hits ram hard - https://www.tomshardware.com/reviews/stress-test-cpu-pc-guide,5461-3.html

i would fix the ram situation before you really can believe any bsod you get as bad ram can cause perfectly good drivers to not work.

only way to know if its GPU is either test GPU in another PC or use another GPU in yours, since ryzen won't work without GPU
Thanks for the reply, I had a few questions about your answer.
Would the bad ram be able to cause these crashes?

View: https://imgur.com/a/2pJZQCd

View: https://imgur.com/a/5JmBkc4

View: https://imgur.com/hPAsgjJ

View: https://imgur.com/a/3Of7wdc

View: https://imgur.com/a/6SrlL9f


Also the person who told me to rma my gpu linked me this article, https://forums.passmark.com/memtest86/42114-errors-in-test-13-row-hammer-what-does-it-mean is this forum correct?
 

Colif

Win 11 Master
Moderator
well, i hadn't seen the pictures. Given those i would lean towards GPU. I would try using another GPU in machine and see if it has same problems. Or test that GPU in another PC and see if it happens there.

Its normally not a good idea to ignore the tests as the advice in the link he links to says
There are several actions that can be taken when you discover that your RAM modules are vulnerable to disturbance errors:
  • Do nothing
  • Replace the RAM modules
  • Use RAM modules with error-checking capabilities (eg. ECC)
Depending on your willingness to live with the possibility of these errors manifesting itself as real problems, you may choose to do nothing and accept the risk. For home use you may be willing to live with the errors. In our experience, we have several machines that have been stable for home/office use despite experiencing errors in the Hammer Test.
https://www.memtest86.com/troubleshooting.htm#hammer
 
Dec 30, 2020
17
0
10
well, i hadn't seen the pictures. Given those i would lean towards GPU. I would try using another GPU in machine and see if it has same problems. Or test that GPU in another PC and see if it happens there.

Its normally not a good idea to ignore the tests as the advice in the link he links to says

https://www.memtest86.com/troubleshooting.htm#hammer
well, i hadn't seen the pictures. Given those i would lean towards GPU. I would try using another GPU in machine and see if it has same problems. Or test that GPU in another PC and see if it happens there.

Its normally not a good idea to ignore the tests as the advice in the link he links to says

https://www.memtest86.com/troubleshooting.htm#hammer
well i did send my gpu to RMA so if that doesnt do it then I i guess its my ram.
 
Dec 30, 2020
17
0
10
well, i hadn't seen the pictures. Given those i would lean towards GPU. I would try using another GPU in machine and see if it has same problems. Or test that GPU in another PC and see if it happens there.

Its normally not a good idea to ignore the tests as the advice in the link he links to says

https://www.memtest86.com/troubleshooting.htm#hammer
Sorry for the extra reply, i was wondering if it’s an possibility that it could be a bad slot on the motherboard.could it be a bad dimm slot?
 

Colif

Win 11 Master
Moderator
Sorry for the extra reply, i was wondering if it’s an possibility that it could be a bad slot on the motherboard.could it be a bad dimm slot?
sorry for slow reply. Only way to test slots would be if you know ram is okay, and then you run good ram in the slots you think are problem. If known good ram have problems with the slot, it could be the motherboard.
 
Dec 30, 2020
17
0
10
sorry for slow reply. Only way to test slots would be if you know ram is okay, and then you run good ram in the slots you think are problem. If known good ram have problems with the slot, it could be the motherboard.
So I received my gpu and everything is fine until I tried to open event viewer and it bsod. Im testing with memtest86 1 at a time and Im having 0 errors even with the hammer test so far. Im suspecting the second slot because both ram sticks worked when I tried to recreate the problem on the 4th slot
 

Colif

Win 11 Master
Moderator
what motherboard is it? I don't see it in listing unless I am blind. Since PC seems new you might be able to RMA the motherboard and get it replaced. Or just avoid ram slots unless you got 4 x 8gb. I would try to get board repaired/replaced.
 
Dec 30, 2020
17
0
10
what motherboard is it? I don't see it in listing unless I am blind. Since PC seems new you might be able to RMA the motherboard and get it replaced. Or just avoid ram slots unless you got 4 x 8gb. I would try to get board repaired/replaced.
I found the error. One of the ram sticks had the same hundreds of erros on the hammer test. I tested 2 slots with one working ram and they all seemd fine. I took it out and its running great now. Bummed that Im losing a chunk of frames in games. I would send it to RMA but I need to ship it all the way to Taiwan and ship both of my sticks so Id rather have only 16x1 than no sticks for weeks. Thanks for all the help!