PC Hard Crashes (Then Reboots) at Random

[Beary]

RTX 4080
14900k (Underclocked Performance Cores to 5.4GHz)
Windows 11 64x
64GB DDR5

As of six days ago, my PC has seemingly at random crash to black and reboot. Not sure if it's a software or hardware issue.

First crash happened while trying to cast to my TV, which I'd done for months without issue. DDU'd my graphics drivers and any related drivers, reinstalled, and it worked for about another hour before crashing again.

Then it started crashing anywhere between every fifteen to two minutes, completely at random, same style of crash (no BSODs), regardless of what was running, before crashing upon booting over and over, until going into disk recovery mode. I tried reseating all the PSU cables, making sure it was plugged into the wall, nothing.

Removed my M.2, formatted it, and put a fresh install of Windows 11 on it. Thought that solved it, because I haven't had a single crash in five days with constant use, until this morning.
 

It crashed twice early in the morning while not in use, rebooted both times. Left it running while at work while under light load to see if it was dead again. Not a single crash in over twelve hours, thought maybe it was just a fluke, until about an hour ago. Four more crashes, back-to-back, within twenty minutes all under different circumstances.

As of right now, it hasn't crashed in about 30 minutes, with everything running as it was before.

Not exactly sure how to diagnose this. Windows Reliability Monitor just lists all of the crashes as "Windows was not properly shut down" rather than a crash, leading me to think its a power issue, but simultaneously why would it suddenly work perfectly fine for almost a week with heavy daily use after simply formatting the drive and reinstalling windows? I'm not sure where else to check for more imformation on the events leading up to the crashes.

Any and all help is greatly appreciated!

[y0ur5h4d0w]

did you try to undo the underclocking of the CPU? sounds like a "hardware issue" but it's hard to diagnose if you put overclocking/underclocking in the mix, put everything to "stock" and then try again, if it crashes then we can start to check for the issue

 

i see that you've underclocked you CPU recently, at least i assume you did recently, did these issues happen BEFORE underclocking or is it something new?

[Beary]

1 hour ago, y0ur5h4d0w said:

did you try to undo the underclocking of the CPU? sounds like a "hardware issue" but it's hard to diagnose if you put overclocking/underclocking in the mix, put everything to "stock" and then try again, if it crashes then we can start to check for the issue

 

i see that you've underclocked you CPU recently, at least i assume you did recently, did these issues happen BEFORE underclocking or is it something new?

No, the this is a very recent issue. I've been underclocking my CPU since about July of last year, because the base clock speed is far too fast for DirectX and other software to the point they're either unlaunchable or CTD. Actually just updated my last forum post about that with the solution after a year lol. Base clock for the performance cores is 5.7Ghz, DirectX is able to function flawlessly at 5.4Ghz. Haven't changed anything hardware-wise or clocking-wise since then

Edit: On top of that, it hasn't crashed since the last crash listed on the reliability monitor. I've been playing The Finals just fine for hours (with w/ underclock, usual daily settings) But that's just the problem. There's seemingly no reason for it, nothing in particular seems to trigger it given the state its in immediately preceding the crash doesn't matter, but somehow a system formatting and fresh install of windows just seems to... stave off the problem completely for 6 days with daily use?

[y0ur5h4d0w]

4 minutes ago, Beary said:

No, the this is a very recent issue. I've been underclocking my CPU since about July of last year, because the base clock speed is far too fast for DirectX and other software to the point they're either unlaunchable or CTD. Actually just updated my last forum post about that with the solution after a year lol. Base clock for the performance cores is 5.7Ghz, DirectX is able to function flawlessly at 5.4Ghz. Haven't changed anything hardware-wise or clocking-wise since then

hmmm ok so there might be a different issue, if you open up the event viewer and check for the windows and application events "error" kind around the time your PC freeze and reboots what can you see? maybe there's some logs there about something that crashed.

 

i was going to suggest you to go for the DISM and the SFC but if you wiped your windows there's another issue somewhere.. and i start to think it might be hardware related since all your tests...

 

how long did you use that PSU for? is it an old one? i had one for 8 years and literally killed my GPU and an HDD storage

 

can you try to change your GPU? maybe simply remove it and try to use your PC with the iGPU to check if that's the issue, i had a faulty GPU some months ago and my system started to freeze and crash randomly, and sadly after a couple months my poor RX6750XT died a very bad death, i don't think this is the case though but it's worth a try.

 

RAM is to exclude, at least from my experience with windows it always lands to a BSOD

 

did you try to change your M.2? how long have you been using it? if it's an old one it might suffer from aging and it's going bad, though if still the PC can boot to the windows checkdisk seems like it's not there... 

[Bjoolz]

Open Event Viewer → Windows logs → System and scroll down to around the timestamp where the previous crash happened. You should have a Kernel-Power event here with the event ID 41. Select it and right above where it gives you info about the event, click the Details tab. Screenshot/copy paste this. 

[Beary]

9 hours ago, y0ur5h4d0w said:

hmmm ok so there might be a different issue, if you open up the event viewer and check for the windows and application events "error" kind around the time your PC freeze and reboots what can you see? maybe there's some logs there about something that crashed.

 

i was going to suggest you to go for the DISM and the SFC but if you wiped your windows there's another issue somewhere.. and i start to think it might be hardware related since all your tests...

 

how long did you use that PSU for? is it an old one? i had one for 8 years and literally killed my GPU and an HDD storage

 

can you try to change your GPU? maybe simply remove it and try to use your PC with the iGPU to check if that's the issue, i had a faulty GPU some months ago and my system started to freeze and crash randomly, and sadly after a couple months my poor RX6750XT died a very bad death, i don't think this is the case though but it's worth a try.

 

RAM is to exclude, at least from my experience with windows it always lands to a BSOD

 

did you try to change your M.2? how long have you been using it? if it's an old one it might suffer from aging and it's going bad, though if still the PC can boot to the windows checkdisk seems like it's not there... 

All of the hardware is brand-new as of December 2023, so about a year and a half old apart from the nearly six months of not really using it because of every game closing at random due to the clock speed issue.

I know the CPU's got integrated graphics so I could try that, I do have a spare M.2 laying around somewhere so if it happens again I'll try that. As of right now it went through the whole night without crashing, still hasn't crashed since I initially made the post.

[Beary]

11 hours ago, Bjoolz said:

Open Event Viewer → Windows logs → System and scroll down to around the timestamp where the previous crash happened. You should have a Kernel-Power event here with the event ID 41. Select it and right above where it gives you info about the event, click the Details tab. Screenshot/copy paste this. 

From the last crash

[Beary]

11 hours ago, Bjoolz said:

Open Event Viewer → Windows logs → System and scroll down to around the timestamp where the previous crash happened. You should have a Kernel-Power event here with the event ID 41. Select it and right above where it gives you info about the event, click the Details tab. Screenshot/copy paste this. 

The rest of them

[Bjoolz]

4 hours ago, Beary said:

The rest of them

I was hoping that the black screen was hiding a BSOD, but it doesn't look like that's the case. Any WHEA events around the time of the crash? If no, my main suspect would be power. The PSU would be the main suspect, but it could be the power cable, if it's plugged into a power strip/surge protector or even unstable power from the wall.

 

And if you have any WHEA events, copy paste the content from the Details tab or right click → save and upload them. I need the long RawData section.

[Beary]

3 hours ago, Bjoolz said:

I was hoping that the black screen was hiding a BSOD, but it doesn't look like that's the case. Any WHEA events around the time of the crash? If no, my main suspect would be power. The PSU would be the main suspect, but it could be the power cable, if it's plugged into a power strip/surge protector or even unstable power from the wall.

 

And if you have any WHEA events, copy paste the content from the Details tab or right click → save and upload them. I need the long RawData section.

Unfortunately nothing from WHEA in the Event Viewer. However, it looks like the exact same events happen at the same few seconds leading up to the crash, though I'm unsure if this is normal as I'm not familiar with Event Viewer. Is that anything of significance?

[Bjoolz]

4 hours ago, Beary said:

Unfortunately nothing from WHEA in the Event Viewer. However, it looks like the exact same events happen at the same few seconds leading up to the crash, though I'm unsure if this is normal as I'm not familiar with Event Viewer. Is that anything of significance?

Informational events are as a rule of thumb ignored because they are just informational. If you get a very clear pattern it could be worth looking into. HAL is the Hardware Abstraction Layer, it works with pretty much all hardware, but is usually more seen with CPU/PCIe device issues (And just to be clear here, this is when you have errors). The Filter Manager makes sure that files are handled correctly. This together with HAL and NTFS could indicate a storage issue. 

 

Again, informational events could mean nothing. We don't have much to go on here so this is just if you want to go looking. 

[Don Ramon]

I have been having a similar issue, when I boot up Marvel rivals my PC just turns off and reboots. The thing is I swapped the PSU right after and then it still did this. I did however have my CPU undervolted and GPU undervolted as well. Not sure if that was the issue but I disabled any sort of undervolt and am gonna try again later to see if that fixed the problem. 

[Beary]

Still no crashes since I made the post despite not changing *any* variables since prior to the crashes... which is really anxiety inducing. But in the event it just magically never happens again, I wanted to thank y'all for your time. Your guidance is appreciated