Question PC crashing with various symptoms ?

[Dobut]

***Update - GPU was the dying part. Putting it into another PC and running games on it produced the same error. Stress testing could not replicate the error even in another PC. I ordered a new GPU. Hope this thread helps others.

Hi All,

You are my last resort as i think i have done everything to test my PC for error causes. Let me describe the issue. Apologies but it will be a longer post. because my PC crashes with various symptons.:

1. Sometimes but most commonly, both monitor shuts off and the whole computer is restarting. In some cases this is followed up by GPU artifacting, and in device manager i can see the driver has a yellow triangle next to it. If this happens the resolution is 1920x1080. If i change to 1024x768 there is no artifacting. It solves itself after restarting a couple times. It may be the case that i need less restart if i leave the PC rest for a little bit, no power on.

2. Earlier i had BSOD with TDR Failure, but i dont get those anymore

3. I had pink screen of Death today, only restart helped.

4. I had both monitors turned off, but i was talking to my friend on discord in the meantime. I was waiting to see if anything happens, but i had to reset to get back my monitors.

5. Sometimes Diablo 4, F1 Manager 22, Alien Dark Descent, Ancient cities are just crashing and dropping me back to desktop. In a few cases it gives me a random crash message related to the game, but nothing that i could research for a solution. League of legends never crashed... (I dont know if this adds anything to the picture )

6. All the errors come after one of the above games i play. If there is no issue with them, i can play the whole day without an error, but it comes back after next boot, next day. (dont tell me to never shut it down on the no error days please )

My Setup.:
- ASUS TUF GAMING B460-PLUS
- Intel Core i7-10700 2.9GHz Socket 1200 (BX8070110700)
- Kingston HyperX Fury 16GB / 3200MHz CL16 KIT, And +32 GB Same
- Western Digital 3TB Purple
- 500 GB Samsung 970 EVO NVMe SSD (M.2, 2280, PCIe)
- Be Quiet Dark Rock 4
- Corsair TX850M
- Chieftec GS-01B-OP Gaming Hunter
- Nvidia 2080 Super

Link to pictures about type of errors i had. I hope it works.
View: https://imgur.com/a/JFMnzRp

https://imgur.com/a/JFMnzRp

Things i did.:
- Fresh windows 11 install
- Latest driver installed from NVIDIA site. I tried different versions to test, but the problem comes back everytime. I always reinstalled drivers with clean DDU setup.
- No overclock, everything is default
- BIOS is updated with latest update
- Both SSD and HDD checked, no failure
- RAM Tested, no failure
- Stress test for CPU and 3 different for GPU, no failure, no overheat
- I even went to a PC store to have them check it. In theory (I wasnt there) they tested the GPU in a different PC and even in mine, no failure. Again, inbetween a failure and this, the PC rested for a while without Power. It can be the reason why they faced no issue.
- I used a PC monitoring program and MSI Afterburner also to check for heating, no failure, everything is in normal ranges
- I changed the cable from the PSU to the GPU. First 1 cable, switched with another, than both into the 2 different pin slots, no change in functioning.

So basically i would really love to know which hardware is malfunctioning, because this is not a software issue i think anymore. Whatever i type into google it gives me the possibility that all my hardware can faulty... so that is not really helping. I am looking for someone who had similar issue and solved it, or someone who has a damn good idea what can be the problem

Thank you in advance for your patience reading this.
Denes

 

[Metal Messiah.]

Sometimes but most commonly, both monitor shuts off and the whole computer is restarting. In some cases this is followed up by GPU artifacting, and in device manager i can see the driver has a yellow triangle next to it

Judging from the entire description of your problem, the most likely culprit could be the GPU. Those artifacts, pink screen, checkboard signs on Monitor are usually the symptoms of a faulty GPU, which can happen if the VRAM is corrupted or there is some other issue with the card's PCB.

Does it also say Code 43 error under Device Manager settings for the GPU, sometimes ?

 

[Dobut]

Judging from the entire description of your problem, the most likely culprit points out to be the GPU. Those artifacts, pink screen, checkboard signs on Monitor are usually the symptoms of a faulty GPU, which can happen if the VRAM is corrupted or there is some other issue with the card's PCB.

Does it also say Code 43 error under Device Manager settings for the GPU ?

It did say yes. After a couple restart it goes back to normal, driver is recognized again and no Code 43 error. Just like now, i am typing from the same computer everything working fine. 2 hours ago i restarted 6 times...

 

[Metal Messiah.]

Okay, if there was a Code 43 error then mostly the problem lies with the GPU. But before jumping on any conclusion, can your test the GPU by using some other power supply/PSU if possible ?

Or, maybe check the GPU on some other system as well ? Although, the TX850M PSU model is a decent unit, but you never know when a PSU starts failing. But I'm almost certain, this is a faulty GPU, and not the power supply, IMO.

 

[Avro Arrow]

I think that Metal Messiah is correct. The symptoms that you're experiencing seem to be consistent with physical problems with a card's VRAM. I'm afraid that your card may be toast.

 

[BwwwJ1st]

Judging from the entire description of your problem, the most likely culprit could be the GPU. Those artifacts, pink screen, checkboard signs on Monitor are usually the symptoms of a faulty GPU, which can happen if the VRAM is corrupted or there is some other issue with the card's PCB.

Does it also say Code 43 error under Device Manager settings for the GPU, sometimes ?

After reading the OP, I was going to post something very similar. Quite possibly bad VRAM.

 

[Dobut]

Update for you all. I could get someone to test the GPU with gaming. It died after 5 minutes of World of Tanks in another computer...

So in a nutshell stress test or benchmark did not kill it, but gaming somehow did it. Its a VRAM issue. I had to order a new GPU.

With that being said, thank you all for your suggestions and help. I wanted to leave this answer here so maybe it helps someone else too.