PC crashes/blackscreens while gaming, but not during stress tests

[maaqu361]

Hey guys, i really need some help.

Roughly 9 months ago, I build my new pc. Since I build it, I have had random crashes/black screens during games or when in idle, but never during stress test or anything like that.

My specs are:

AMD Ryzen 5 7600
GIGABYTE Radeon RX 6800 XT GAMING OC 16G
ASUS TUF Gaming B650-Plus
2x16 GB G.Skill Ripjaws 5600MHz
Noctua NH-D9L
NZXT C850 Gold (850W)
OS: Windows 10

I have tried the demo of 3dMark, all OCCT benchmarks, cinebench r23, memtest86 and some other benchmarks. My PC never crashed during these. The speeds and temperatures stay at normal levels, even after like 3 hours or more.
But most of the time it just randomly crashes in games after like 20-30 minutes. Sometimes it only crashes after like 2 hours or even never . The games which crash it are basically all games that I currently play, so Valorant, Cyberpunk, Fortnite, Forza Horizon 5, GTAV, and some more.

I am running stock speeds on both GPU and CPU and over-/underclocking or changing anything hasn't helped ether. I've tried all DOCP profiles and turning it off has also not helped. I am running the newest chipset and gpu driver, as well as the newest BIOS. (I dont really think that matters scince these crashes exist since I build my PC).

The Windows Event Viewer also doesn't help. I only get:
- Error: Eventlog (ID: 6008)
- Critical: Kernel-Power (ID: 41)
But these 2 show only after the crash, there is no event before my pc crashes.

Currently I really don't know what to do anymore. The crashes are just getting really annoying while im playing with friends and it's also not like the PC was cheap.

Thanks for reading all that, I hope someone can help me.

[Average Nerd]

1 hour ago, maaqu361 said:

Hey guys, i really need some help.

Roughly 9 months ago, I build my new pc. Since I build it, I have had random crashes/black screens during games or when in idle, but never during stress test or anything like that.

My specs are:

AMD Ryzen 5 7600
GIGABYTE Radeon RX 6800 XT GAMING OC 16G
ASUS TUF Gaming B650-Plus
2x16 GB G.Skill Ripjaws 5600MHz
Noctua NH-D9L
NZXT C850 Gold (850W)
OS: Windows 10

I have tried the demo of 3dMark, all OCCT benchmarks, cinebench r23, memtest86 and some other benchmarks. My PC never crashed during these. The speeds and temperatures stay at normal levels, even after like 3 hours or more.
But most of the time it just randomly crashes in games after like 20-30 minutes. Sometimes it only crashes after like 2 hours or even never . The games which crash it are basically all games that I currently play, so Valorant, Cyberpunk, Fortnite, Forza Horizon 5, GTAV, and some more.

I am running stock speeds on both GPU and CPU and over-/underclocking or changing anything hasn't helped ether. I've tried all DOCP profiles and turning it off has also not helped. I am running the newest chipset and gpu driver, as well as the newest BIOS. (I dont really think that matters scince these crashes exist since I build my PC).

The Windows Event Viewer also doesn't help. I only get:
- Error: Eventlog (ID: 6008)
- Critical: Kernel-Power (ID: 41)
But these 2 show only after the crash, there is no event before my pc crashes.

Currently I really don't know what to do anymore. The crashes are just getting really annoying while im playing with friends and it's also not like the PC was cheap.

Thanks for reading all that, I hope someone can help me.

This sounds like there's an isue with the power supply/power delivery system, which is indicated by Kernel-Power 41, which shows up after the system has rebooted without being cleanly shut down before (basically after a sudden loss of power).

Since your power supply exceeds the recommended wattage for your components, I suspect it to be faulty. If possible, test with a different PSU, and see if the issue persists.

[adm0n]

Based on what you say, there is very little anyone can help you. Your PC doesn't produce any kind of logs, so if you have the time, I'd try to at least figure out which device is causing your issues.

 

It sound a bit like a power problem, because your PC just stops. So based on that it could be a PSU/GPU/CPU or Motherboard problem. (Also possible but less likely are SSD/RAM/USB device problems).

 

So if you have any USB devices connected to your PC, that you don't trust, I'd disconnect them. If you get a short circuit on the USB pins, your PC usually turns off to protect itself.

 

Next, I'd take out your GPU and run your PC with the iGPU of your CPU. You said your PC also dies in idle, so if your PC still happens, you can at least rule out your GPU. 

 

The next troubleshooting steps become trickier though. If you have a friend with a similar PC, you could ask them to swap CPUs and see if he experiences the same issues, otherwise you could try to RMA all possible parts (so CPU/PSU/Motherboard). But you'll probably need to make up some kind of issue for the vendors to accept your RMA. Otherwise they'll say it's not their product that is broken.

 

Other things that could help are:

• reinstalling windows
• running with only one RAM stick
• running memtest for 12 or 24 hours.
• check your GPU/CPU temperatures (you probably did this already)

[maaqu361]

I could definitely try another PSU (some 600W be quiet PSU from my old PC, which worked fine). The weird thing is, that the PC doesn't just fully shut off. The PC still runs -> the fans keep spinning, the RGB is still on. The audio in the moment of the crash loops and nothing happens till I manually hold the start button or switch the PSU off to turn the PC off. I personally have no explanation as to why it is crashing in this way. But somehow I have the feeling that it just might be the PSU. Im going to test it when im home in like 4-5 hours. 

[Average Nerd]

1 minute ago, maaqu361 said:

I could definitely try another PSU (some 600W be quiet PSU from my old PC, which worked fine). The weird thing is, that the PC doesn't just fully shut off. The PC still runs -> the fans keep spinning, the RGB is still on. The audio in the moment of the crash loops and nothing happens till I manually hold the start button or switch the PSU off to turn the PC off. I personally have no explanation as to why it is crashing in this way. But somehow I have the feeling that it just might be the PSU. Im going to test it when im home in like 4-5 hours. 

The 600W PSU might not be sufficient for your PC (AMD recommends at least 750W for your GPU), and its overcurrent protection might switch it off. If that happens, the entire PC will just go dark from one second to the other, so you will definitely know.

 

But with that additional info, it doesn't sound like the PSU is the issue (it might still be, but less likely), but rather a faulty motherboard or other component.

[maaqu361]

At this point I also think it might be the motherboard. I could switch the whole combination of motherboard, CPU and RAM, just to test if the problem lies with the GPU or PSU. (Old specs were: Ryzen 5 3600x, some ASUS motherboard and 2x8GB 2666MHz DDR4 RAM). I could also switch the GPU to my old GTX 1660 and see if that changes anything.

[Average Nerd]

4 minutes ago, maaqu361 said:

At this point I also think it might be the motherboard. I could switch the whole combination of motherboard, CPU and RAM, just to test if the problem lies with the GPU or PSU. (Old specs were: Ryzen 5 3600x, some ASUS motherboard and 2x8GB 2666MHz DDR4 RAM). I could also switch the GPU to my old GTX 1660 and see if that changes anything.

Do that, the best way would be to first only swap in one component at a time, then, when that isn't possible anymore (because of compatibility and such) swap in entire combinations.

[adm0n]

42 minutes ago, maaqu361 said:

he weird thing is, that the PC doesn't just fully shut off. The PC still runs -> the fans keep spinning, the RGB is still on. The audio in the moment of the crash loops and nothing happens till I manually hold the start button or switch the PSU off to turn the PC off.

That's different from the PC just turning off. If the audio keeps looping, the components of your PC are basically still executing the last instructions over and over again. So if there is still sound in the buffer, you will just keep hearing the sound that is in there over and over again.

 

I'd say this isn't likely to be a power issue then. It's far more likely for this to be a CPU or Motherboard issue. Less likely would be a RAM or SSD issue, since your PC won't be able to record anything, if your SSD dies while it is on, but you will usually still get bluescreens with info on it. And RAM would lead to random driver crashes and not the same kind of issue over and over again.

 

If it was your GPU, your CPU could still create a memory dump (and write something in the event log). So if you CPU malfunctions (or the board causes it to malfunction), you'd have a situation as you described.