Hello all, sorry if I post this in the wrong area, my friend recommended I come here with this issue!
Basically, seemingly randomly, my PC will just turn off both of my monitors and my GPU fans will turn to max speed, and the computer becomes completely unresponsive until I turn it off manually.
This is actually an issue I had with my previous PC, which I have since gotten rid of and replaced it with a brand new one. However I did take a 500GB SSD and a 3TB HDD, along with 2x8GB of RAM. Currently, the HDD holds some games that I have transferred over as I didn’t want to reinstall them and the SSD is currently wiped but not formatted.
Since I had been having issues with my previous build this is everything I did to try to fix it:
Run memcheck on the 2x8GB of RAM from my previous PC (no errors)
Unplug from a UPS and plug directly into the wall outlet
Plug the PC into a different wall outlet
Update Bios
Install and Update drivers
Reinstall graphics drivers
Checked Event Viewer for any errors (none found)
Run Chkdsk on the 3TB SSD (no errors)
Completely unplug the SSD and HDD
My Specs are:
Intel Core i7-13700KF 3.4 GHz 16-Core Processor
Deepcool AK620 68.99 CFM CPU Cooler
MSI Ventus GeForce RTX 4070 Ti 12GB GDDR6X PCI Express 4.0 Video Card RTX 4070 Ti VENTUS 3X 12G OC
Corsair RM850e 850 W 80+ Gold Certified Fully Modular ATX Power Supply
Windows 10 Pro Version 22H2 Build 19045.2006
I am currently at a complete loss for what the problem is and I’m willing to try just about anything, any advice is appreciated!
Update #1:
I’ve tried a lot of what was suggested here with no results, the issue still keeps happening. Here is a list of what I’ve tried so far in addition to above:
Remove the extra 2x8GB of RAM
Uninstall GPU drivers using DDU for a clean reinstall
Try one monitor at a time
Double checked my monitor’s refresh rates to make sure they are standard
Ran memcheck on the new RAM
Ran a sfc scan to repair system files
Ran an antivirus scan
Additionally, I have a memory dump file that was generated at one of the crashes, you can read it at: pastebin(dot)com/CzMQqVLt
Update #2
Figured I’d share an update of everything I have tried today to no success:
Switching the GPU to a different PCIe port
Using Nvidia’s tools to uninstall everything and then reinstalling it
Used a second PCIe cable to connect to the power adapter for the GPU
Ran a burn in test and monitored temperatures and power draw, both were within expected levels
Ran a heavy-duty extension cord across my house to the other side to see if it was an issue with power draw.
Obviously none of these have worked, however I have learned that I can cause the issue to happen fairly consistently within 1-3 minutes by running FurMark.
Do you have a screen saver configured? Check if this is set to start a custom screen saver.
When you say your computer becomes completely unresponsive, can you go into more detail?
E.g. it takes a moment (or a couple of moments) for the sceen to show again vs. screen never wakes up again.
When I say it “becomes unresponsive” I mean that the monitors go dark and my GPU fans start maxing out. I can still hear whatever I had playing for a few seconds (4-5), then that cuts out as well. After this the computer doesn’t seem to respond to anything I do as well as the GPU fans stay at max.
I have run a virus scan through windows defender, but I am currently running another one with malwarebytes.
Very odd that this is happening with two different builds. To clarify, are the ONLY pieces of hardware that you carried over from the previous system the SSD, HDD, and RAM? Was it built into the same case?
Yes, the only thing that I carried over was the SSD the HDD and the 2x8GB of RAM. The case and all other parts are all new. I do use the same monitors, keyboard and mouse if that makes any difference.
saying as you have a drive spare. pull the other 2 old ones and install windows on the wiped one.
the less hardware in the system for testing the better.
in uefi make sure hpet is enabled. along with all your other prefrences.
why?
right now it sounds like you have a hardware fault.
but it could be a simple thing like windows cant find the hpet hardware clock and locks up.
so check it, enable it.
hpet disabled:
amd gpu systems lock up with a coloured screen and repeating audio
intel with nvidia lock up with black screens and repeating audio.
when running an o.s. that needs hpet enabled.
if that seems to keep your system stable add in the other drives and wait for a fail.
Is the installation of windows the same from the previous system? If so, there may be an OS level issue afoot. Maybe try a fresh installation on a different drive?
While you are testing, I would recommend removing the 2x8 GB sticks and just run the 2 x 16 GB sticks. The less possible sources of issues the better. Also run memtest on your new memory, that’s just a good habit to get into whenever you buy new memory. Better to prove that it’s working correctly while you are still in the return period.
Out of curiosity, this previous system that had the issue, was this also a brand new build in which this issue would happen? Or was it a system that was working well for a period of time and then started having this issue?
This is a different installation of windows, as suggested above I’m going to try a fresh installation on a new drive tomorrow, along with running a memcheck on all of my RAM.
The previous system that had the same issue was actually fine for a long time (4ish years) and then started having this issue. I figured it was time for an upgrade anyway so I did so and the issue seems to have followed me
It sounds like a kernel lockup, maybe from a bad GPU driver. Did you try using older drivers on the previous system? It could also be a Windows related problem; some software you’re running may cause a windows component to misbehave and lock up the system.
Have you checked event viewer? There may be useful information, such as if the drive you’re using has a flaky controller that’s failing to read out the correct data sometimes.
Memtest also isn’t a very thorough test of memory stability, due to the complicated nature of memory stability. I wouldn’t assume memory is fine just because memtest couldn’t find something wrong. Try using cpu burntests with a heavy memory component as well.
Is the computer plugged into a UPS, or into a surge protector? Your post says UPS, but then you mentioned plugging it straight into a wall outlet rather than a surge protector. Something to try if nothing else works is a dedicated power filter; The UPS will pass through wall power as long as it’s within a certain spec, but a sensitive component might not like a small voltage fluctuation on the AC coming through and making the PSU deliver some less than clean power. I would think a bespoke power filter would have stricter input and output power requirements than a UPS typically would, unless said UPS was designed with power filtering as a feature.
It does sound like a GPU driver crash, have you also tried another cable for your monitor(s) and only connect one at the time? I would also rip out your old 2x8Gb sticks since speed is severly mismatched and while it should you might experience interesting issues. For the sake of troubleshooting, make sure your new set of RAM runs at JEDEC specs (3200) before continuing to troubleshoot.
An issue I’ve noticed myself is that Widewine does not like “odd” refresh rates, in my case 75Hz which will make drivers go bonkers (there are quite a few reports about this known issue). You can easily test this by trying to view something on for example Netflix. It doesn’t hang the computer but it takes a while (10 minutes or so in my case) before the driver recovers. I’m using Intel video but this seems to affect all brands.
It looks like ASUS doesn’t support enabling/disabling HPET and from the basic research I’ve done it seems like its enabled by default with no way to change it.
I am not using any older drivers, and It is plugged into a surge protector/battery backup.
However, checking the event viewer after letting the “crash” go on for a little longer caused it to generate some errors.
Specifically it seems like DWM.exe is crashing for some reason and is unable to restart.
Additionally, one of these times a dump file got created, I’ve made a pastebin for it’s contents but I can’t puts links yet so its here: pastebin(dot)com/CzMQqVLt
If you are having the exact same very specific issue on two different OS installations in two completely different computers with no components in common, then I’m inclined to believe that this isn’t an issue with PC components at all. I understand the feeling that it’s as if bad luck is following you, but this doesn’t really make much logical sense.
Can you possibly hook up the PC to a different power circuit somewhere in your home? Or if you still have your old PC, can you take it to a friend or family member’s house and hook it up and see if the same issue happens there? I’m wondering if you might have some sort of electrical issue.
Update #1:
I’ve tried a lot of what was suggested here with no results, the issue still keeps happening. Here is a list of what I’ve tried so far in addition to above:
Remove the extra 2x8GB of RAM
Uninstall GPU drivers using DDU for a clean reinstall
Try one monitor at a time
Double checked my monitor’s refresh rates to make sure they are standard
Ran memcheck on the new RAM
Ran a sfc scan to repair system files
Ran an antivirus scan
Additionally, I have a memory dump file that was generated at one of the crashes, you can read it at: pastebin(dot)com/CzMQqVLt
How well did it go to have Windows on a totally different drive?
Can you try older drivers that would have been used when your previous system was stable?
