Question Windows 10 Freezing and Crashing Without BSOD

[ActuarialJudgment]

Hi,

Around 4 months ago I was having issues with my PC crashing to BSoD frequently and I came here for help. Ultimately we realized it was a faulty CPU so I got a new part and had no issues at all for 4 months straight.

Just last week though my computer has started freezing randomly. It's happened 3 times so far (about every 3 days or so). I've been doing various things each time and it just completely locks up - I'm not able to input anything and if I don't turn it off, it'll restart after a minute or two.

I tried running Prime95 for a few hours and had no errors at all, so I think my CPU/RAM are working fine.
I've also tried doing sfc /scannow in the command prompt. This found corrupted files and repaired them, but it didn't fix the crashing.

What else should I try to diagnose or fix this issue? I could reinstall Windows if I have to, but I'd prefer not to do that if possible.

Specs:
CPU: Ryzen 5 5600
Mobo: B450 AORUS Elite
GPU: Nvidia GeForce GTX 1660 SUPER
DDR4 3200 RAM 2x8GB
Samsung EVO 970 1TB NVMe m.2 SSD
PSU: Seasonic FOCUS GX-550
Windows 10 Pro

Thanks a bunch in advance!

EDIT: I looked through my minidump folder and it looks like one of the crashes did create a dump file.
Here's a link to it: https://drive.google.com/file/d/1hNFBXbJH4wkgBu5cOoaiYai43E6T-3pi/view?usp=share_link

 

[Colif]

Converrsion of Dump

Report - click run as fiddle to read (Mostly for me)


File: 010623-7500-01.dmp (Jan 6 2023 - 22:35:16)
BugCheck: [DPC_WATCHDOG_VIOLATION (133)]
*** WARNING: Unable to verify timestamp for nvlddmkm.sys
Probably caused by: nvlddmkm.sys (Process: System)
Uptime: 0 Day(s), 0 Hour(s), 17 Min(s), and 44 Sec(s)

The nvlddmkm.sys file is a NVIDIA graphics card driver. There are a few things you can do to fix this problem. First off, try a full uninstall using DDU in Safe Mode then re-install the driver (more information). Or try getting the latest version of the driver. Or try one of the 3 most recent drivers released by NVIDIA. Drivers can be found here: http://www.nvidia.com/ or you can allow Windows Update to download the driver for you, which might be a older/better version.

 

[ActuarialJudgment]

Thanks for the quick response! I just installed the latest driver from nvidia's website. I didn't bother with the DDU for now. If I'm still having issues I'll try the DDU installation.

Since the crashes were fairly infrequent it may be a few days before I have an update, but I'll let you know how it goes. Thanks again!

 

[ActuarialJudgment]

Wow, that was fast. I already got another crash and this time it didn't create a dump file. In the event viewer there's an error event that says "Dump file creation failed due to error during dump creation."

I guess I'll try the DDU instead now..

 

[ActuarialJudgment]

Okay, about 10-15 minutes after completing the DDU install with the newest GPU driver I got another crash that didn't create a dump file...
Any other ideas for what I should try next?

This is concerning that the crashes are really frequent today..

 

[ActuarialJudgment]

I checked reliability monitor and this is the error I'm getting for the crash that happened after the DDU.

One more piece of information that might be helpful: I started having these issues after a power outage a couple weeks back. I had my PC plugged into a surge protector, but I wonder if that has something to do with it.

 

[johnbl]

stack was corrupted, raw stack shows a call to this driver:
rt640x64.sys Tue Jun 14 23:03:01 2022

 

[ActuarialJudgment]

Hmm, does that mean there's an issue with my Realtek driver? Should I reinstall the newest one if I'm still having issues?

I was poking around more in the reliability monitor and noticed around the time I started getting crashes (and periodically since they started) there's an error that says Runtime Broker stopped working.
I looked online and found this was related to the Windows Store. I tried resetting the microsoft store's data, so hopefully that'll fix that issue.

 

[johnbl]

sorry, I got interrupted when I was looking at the memory dump.(Just got back)
your system was running for
System Uptime: 0 days 0:17:44.141
the system spent
Time: 17 Min 34 Sec 187.50 mSec

making calls to
nvlddmkm.sys Thu Jul 21 17:04:08 2022

the corrupted stack did show the real tek driver but it might be random chance.
the bugcheck just indicates that all together (all drivers) took too long to complete. Most likely it will be nvlddmkms.sys

update the driver directly from nvidia. I would turn off functions that are not required like nvidia share if you have it running.
make sure you do not have any bit coin mining running.

more debugging can be done if you change the memory dump type to kernel and provide memory.dmp file. This can look for software locks that are being held by stupid programs that will cause the nvidia driver not to be able to process its queue of work items. generally, it will be some screen video/audio capture software or streaming software.

what ever was running started 10 seconds after your system booted up. (at least the queue of work items was backed up right after bootup)

 

[ActuarialJudgment]

Okay, thanks! I updated my nvidia driver earlier to the latest version using DDU and have changed my memory dump type to kernel. Unfortunately most of the time my PC is crashing, it doesn't provide a dmp file but we'll see how it goes.

Since you mentioned video capture software, I think I'll go update my OBS to the latest version too.

 

[johnbl]

Okay, thanks! I updated my nvidia driver earlier to the latest version using DDU and have changed my memory dump type to kernel. Unfortunately most of the time my PC is crashing, it doesn't provide a dmp file but we'll see how it goes.

Since you mentioned video capture software, I think I'll go update my OBS to the latest version too.

most likely the problem starts at boot but does not bugcheck. This means you can google "how to force a memory dump using a keyboard" make the registry changes. then run your system for a while say 20 or 30 minutes then force a kernel memory dump on the running system. I can take a look and see if something is starting to go wrong. with a kernel dump I can maybe see what is making the driver calls or delaying the processing.

 

[ActuarialJudgment]

most likely the problem starts at boot but does not bugcheck. This means you can google "how to force a memory dump using a keyboard" make the registry changes. then run your system for a while say 20 or 30 minutes then force a kernel memory dump on the running system. I can take a look and see if something is starting to go wrong. with a kernel dump I can maybe see what is making the driver calls or delaying the processing.

I tried all this and after forcing the BSoD my computer didn't create a dump file.. Should I turn off the setting that makes it automatically reboot so that it has time to create the dump file or something?

Okay, second time it worked. Uploading the file now - Even zipped the dump file is almost 3 GB..

Here's the dump file:
https://drive.google.com/file/d/1cgF_rEcxG9ZBQsUCslHYWr8NTgs-l5kV/view?usp=share_link

 

[Colif]

I will let John do it as he can get more info out of dumps than I can. 3gb dumps give it just as much as a minidump for me.

 

[johnbl]

Looking at the kernel dump now:
rt640x64.sys is waiting for a wake up signal. This is the driver that was on the corrupted stack on the first dump. I would go into windows control panel, device manager find the device and right click to bring up properties and find the power management tab and tell windows not to turn it off to save power.

in total 6 devices are waiting for a wake up signal. most are usb devices but 2 are PCI. Will try to figure out what they are.

UcmCxUcsiNvppc USB Type-C port controller driver

USB3hub
USBXHCI
HidUsb


these two I would turn off sleep
storahci (this is your storage driver)
rt640x64.sys (realtek ethernet driver?)

I would turn off sleep for the storage driver only due to not getting saved memory dumps when the system crashes.

for the most part it looks like the system is running well. it is not backing up on processing requests yet.

will look at some internal logs and add some notes if I see anything of interest.
notes:
UcmCxUcsiNvppc if this is the usb 3.1 controller on the nvdia graphic card and you do not actually use it. then it can be disabled and would remove it as a factor in messing up your GPU.
(the gpu drivers would not have to have handle the processing of the devices)

if you have usb devices attached to the nvidia card, I would move them off onto a motherboard usb hub.

Greenshot.exe screen shot software running.
the last calls to the storage adapter were to power it down.
one looks like a RAID call. Is the storage set up for RAID mode?
this was the call: storport!RaidAdapterPowerDownDevice
looks like you have another storage device that is working without problems on \Driver\secnvme

a few minor errors in the internal usb logs for
USB\VID_046D&PID_C07D - G502 Mouse | Device Hunt
maybe see if there is a firmware update for your mouse?
internal error was:
Error code: (NTSTATUS) 0xc0000004 (3221225476) - The specified information record length does not match the length required for the specified information class.
logitech g502 mouse.

got to run for now. No real big problems found.

 

[ActuarialJudgment]

Looking at the kernel dump now:
rt640x64.sys is waiting for a wake up signal. This is the driver that was on the corrupted stack on the first dump. I would go into windows control panel, device manager find the device and right click to bring up properties and find the power management tab and tell windows not to turn it off to save power.

in total 6 devices are waiting for a wake up signal. most are usb devices but 2 are PCI. Will try to figure out what they are.

UcmCxUcsiNvppc USB Type-C port controller driver

USB3hub
USBXHCI
HidUsb


these two I would turn off sleep
storahci (this is your storage driver)
rt640x64.sys (realtek ethernet driver?)

I would turn off sleep for the storage driver only due to not getting saved memory dumps when the system crashes.

for the most part it looks like the system is running well. it is not backing up on processing requests yet.

will look at some internal logs and add some notes if I see anything of interest.
notes:
UcmCxUcsiNvppc if this is the usb 3.1 controller on the nvdia graphic card and you do not actually use it. then it can be disabled and would remove it as a factor in messing up your GPU.
(the gpu drivers would not have to have handle the processing of the devices)

if you have usb devices attached to the nvidia card, I would move them off onto a motherboard usb hub.

Greenshot.exe screen shot software running.
the last calls to the storage adapter were to power it down.
one looks like a RAID call. Is the storage set up for RAID mode?
this was the call: storport!RaidAdapterPowerDownDevice
looks like you have another storage device that is working without problems on \Driver\secnvme

a few minor errors in the internal usb logs for
USB\VID_046D&PID_C07D - G502 Mouse | Device Hunt
maybe see if there is a firmware update for your mouse?
internal error was:
Error code: (NTSTATUS) 0xc0000004 (3221225476) - The specified information record length does not match the length required for the specified information class.
logitech g502 mouse.

got to run for now. No real big problems found.

Thanks for all the info/detail!

I just got another crash this afternoon, so something is definitely still wrong, but no dump/minidump was created.

I disabled the nvidia GPU usb driver and changed the power management settings on several of the drivers you listed. For the storachi driver, it didn't have a power management tab when I went into properties, so I wasn't able to update that one.

I don't believe my storage is set up to use RAID mode. Is Greenshot causing issues? I'd be fine getting rid of it if so.

I just updated my mouse firmware too.

Let me know if there's anything else I should look into or try and thanks a ton for the help!

 

[johnbl]

Thanks for all the info/detail!

I just got another crash this afternoon, so something is definitely still wrong, but no dump/minidump was created.

I disabled the nvidia GPU usb driver and changed the power management settings on several of the drivers you listed. For the storachi driver, it didn't have a power management tab when I went into properties, so I wasn't able to update that one.

I don't believe my storage is set up to use RAID mode. Is Greenshot causing issues? I'd be fine getting rid of it if so.

I just updated my mouse firmware too.

Let me know if there's anything else I should look into or try and thanks a ton for the help!

I did not see any problem with greenshot, just noted it because it was a capture program that could cause a problem.

the raid mode would be set on the sata controller inside of the BIOS.
you might also run crystaldiskinfo.exe and read the SMART data about your drive(s) if it reports a firmware version you might look up the drive and see if there is a firmware update.

see how the system runs with the changes you have made. if it starts to run poorly then you can force a memory dump and maybe I can see something wrong. if the system does a freeze, try to force a kernel dump.

I think windows will tell a drive to sleep after 5 minutes (or maybe 10)

i just looked up greenshot it looks like a simple app without any device drivers of its own. You would of had to do a screen capture 10 seconds after the system booted for it to have an effect on the minidump I looed at. (unlikely to be the root cause)
did you happen to have any devices connected to the nvidia usb 3.1 port? I am thinking that port is the most likely source of a problem for a gpu.

all of the audio devices queue were idle in the kernel dump.
(audio devices tend to mess up the GPU also)

only thing that looked wrong was the system process memory usage.

 

[ActuarialJudgment]

I did not see any problem with greenshot, just noted it because it was a capture program that could cause a problem.

the raid mode would be set on the sata controller inside of the BIOS.
you might also run crystaldiskinfo.exe and read the SMART data about your drive(s) if it reports a firmware version you might look up the drive and see if there is a firmware update.

see how the system runs with the changes you have made. if it starts to run poorly then you can force a memory dump and maybe I can see something wrong. if the system does a freeze, try to force a kernel dump.

I think windows will tell a drive to sleep after 5 minutes (or maybe 10)

i just looked up greenshot it looks like a simple app without any device drivers of its own. You would of had to do a screen capture 10 seconds after the system booted for it to have an effect on the minidump I looed at. (unlikely to be the root cause)
did you happen to have any devices connected to the nvidia usb 3.1 port? I am thinking that port is the most likely source of a problem for a gpu.

No, I only have my 3 monitors connected to my GPU and nothing else. Hopefully disabling that in the device manager will fix the issues I've been having. I'll keep you updated!

 

[johnbl]

3 monitor would each be running via a different interface on this card.
dvi, hdmi 2, and some version of display port.
you might look to see if the card has a vbios firmware update.
NVIDIA GeForce GTX 1660 Firmware Update Utility
I see one for dell system.

card info:
Display outputs include: 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a. GeForce GTX 1660 SUPER is connected to the rest of the system using a PCI-Express 3.0 x16 interface.
looks like the update helps in detecting the proper interface when you plug in the monitor to prevent a blank screen
dell vbios update for their laptop was released 24 Apr 2020

 

[ActuarialJudgment]

3 monitor would each be running via a different interface on this card.
dvi, hdmi 2, and some version of display port.
you might look to see if the card has a vbios firmware update.
NVIDIA GeForce GTX 1660 Firmware Update Utility
I see one for dell system.

card info:
Display outputs include: 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a. GeForce GTX 1660 SUPER is connected to the rest of the system using a PCI-Express 3.0 x16 interface.
looks like the update helps in detecting the proper interface when you plug in the monitor to prevent a blank screen
dell vbios update for their laptop was released 24 Apr 2020

I have 2 monitors plugged in with display port and 1 with hdmi.

I tried getting the dell firmware update but it said my card wasn't compatible and gave me an error when I tried to run it.

 

[ActuarialJudgment]

Just got another crash this time without any freezing at all, it just shut off and restarted all of a sudden... Again, no blue screen or minidump to be found.

Any other ideas? I also noticed recently that my computer notified me I'm eligible for a windows 11 upgrade. Do you think it's worth upgrading to Windows 11 to see if that helps?

 

[johnbl]

Just got another crash this time without any freezing at all, it just shut off and restarted all of a sudden... Again, no blue screen or minidump to be found.

Any other ideas? I also noticed recently that my computer notified me I'm eligible for a windows 11 upgrade. Do you think it's worth upgrading to Windows 11 to see if that helps?

without knowing the root cause of the problem you would just migrate the bad driver/settings over to the new windows. Best to do a clean install of windows and reinstall the current version of any apps you need. ie do a clean install of windows 11 rather than a update of window 10 to windows 11.

 

[ActuarialJudgment]

without knowing the root cause of the problem you would just migrate the bad driver/settings over to the new windows. Best to do a clean install of windows and reinstall the current version of any apps you need. ie do a clean install of windows 11 rather than a update of window 10 to windows 11.

Then what are next steps for me? Should I try forcing another blue screen and uploading another log or should I just do a clean install of windows now?

I already got a crash this morning, so I just pulled out my spare PSU and swapped it out. I'm hoping that'll fix it.

 

[johnbl]

Then what are next steps for me? Should I try forcing another blue screen and uploading another log or should I just do a clean install of windows now?

I already got a crash this morning, so I just pulled out my spare PSU and swapped it out. I'm hoping that'll fix it.

I would do a clean install and see if the system is stable. Then add programs and see if the stability changes.

 

[ActuarialJudgment]

Okay, that'll be my next step if the new PSU doesn't help. Thanks!

 

[ActuarialJudgment]

After a couple of days of no crashes, I already got 2 crashes this morning with the new PSU.

So you think I should definitely do a fresh install of windows next? Should I try testing out my GPU or RAM first? I'd prefer not to wipe computer and reinstall windows if I don't have to.

Do you know a good way to test if the GPU is the problem?

Edit: Ran Memtest86 for 3 hours today and there were no errors on all 4 passes. Then I downloaded the Heaven GPU stress test and ran it with max quality/tessellations and no issues.

I'm gonna try rerunning prime95 for a few hours and if there are no errors then, I guess reinstalling windows is the next step, huh?