PC Randomly Crashing

[pete6274]

Hi all, I would greatly appreciate it if anyone has any advice on a weird issue I'm having. I just recently built my second ever gaming PC which has an i7 14700K processor and a Zotac RTX 4070 Ti Super graphics card. The PC runs perfectly most of the time but seemingly at random my screen will just go black, the fans will kick into overdrive, and the PC will become unresponsive forcing me to hold the power button in order to force shutdown. It should be noted that any audio playing from my PC is unaffected. Apps like Spotify will continue playing music uninterrupted despite the screen being black and nothing responding.

 

This has so far only happened while playing a game. I have tried multiple different games and I have not been able to get this issue to occur deliberately. It appears to be random while playing any game. Does anyone have any ideas what might cause this kind of behaviour? I have checked my drivers and all windows updates are done. I have also tried shutting down any apps running in the background. So far the only consistency with this behaviour appears to be the fact that I am playing a game.

[Mumintroll]

What are the rest of the pc specifications?

 

Are there any error and warning messages in the event viewer that could point to something?

Are there any warnings in the device manager?

 

Do you have another pc to swap parts with for troubleshooting?

Or a friend that you can borrow a pc from?

If so start with the 4070, test it in another pc. Then test another gpu in your pc.

Do so with other components aswell.

 

Another note is your cpu.

Do you know about Intel's degradation issues on their Raptor lake series cpus?

If you've bought it second hand then that's a huge risk it has permanently degraded by the previous user.

Sometimes resellers scam people selling those cpus that got returned to them as new, mostly through Amazon it seems.

Did you buy the cpu from Amazon?

 

In any case I hope you've the latest bios and using Intel's default profile.

 

Once you've found the suspicious component use your warranty.

[pete6274]

Thanks for the response.

 

I have been checking event viewer, but the only critical event is from me holding the power button to force shutdown. There is an error event with the description "Session "PerfDiag Logger" stopped due to the following error: 0xC0000188" or error "0xC0000035" but these are the only items that stand out to me.

 

All the parts for my PC are brand new so I doubt degradation would be the issue. Unfortunately I don't have another GPU to test. Do you think it's likely to be a hardware issue?

 

I have checked all my drivers are up to date and I have also updated my motherboards BIOS.

[pete6274]

I haven't been able to confirm a connection with this just yet, but the issue does seem to happen more often when I am streaming. I've noticed that if I am just playing a game normally, this issue has only happened once or twice. whereas if I am streaming, it seems to occur once every couple of hours.

[pete6274]

Sorry, I realised I missed the first part of your message. My full PC Specs are:

• Zotact RTX4070 Ti Super Graphics Card
• Crucial P3 Plus 2TB NVMe SSD
• Corsair Vengeance RGB 32GB DDR5-6000MHz RAM
• Intel i7-14700K Raptor Lake Processor
• Gigabyte EAGLE AX Z790 Motherboard
• Super Flower Leadex III GE 1300W Power Supply
• Corsair iCUE LINK TITAN 360 RX RGB AIO Cooler

There are no errors in device manager that I can see.

[Mumintroll]

Well my first bet would still be the 4070. Can you bring it over to a friend to test it there perhaps?

 

Perhaps it could also be the powersupply? Since it happens during load, when gaming and streaming, when it draws more power.

Perhaps the AIO cooler is not doing its job properly? What's the temps on the CPU and GPU during load?

 

You could also do a memtest to rule out the RAM.

Check the SMART table on that Crucial drive.

 

Also to answer me, use quote, otherwise I don't get a notification if you've replied.

[pete6274]

6 hours ago, Mumintroll said:

Well my first bet would still be the 4070. Can you bring it over to a friend to test it there perhaps?

 

Perhaps it could also be the powersupply? Since it happens during load, when gaming and streaming, when it draws more power.

Perhaps the AIO cooler is not doing its job properly? What's the temps on the CPU and GPU during load?

 

You could also do a memtest to rule out the RAM.

Check the SMART table on that Crucial drive.

 

Also to answer me, use quote, otherwise I don't get a notification if you've replied.

Thanks for the tips. I might have to see if I can send my GPU back under warranty. Unfortunately all my friends are console gamers. None of them have desktop PC's. 

 

I have installed Furmark, Prime95, and HWmonitor and run multiple stress tests but still cannot repeat the issue. I have also plugged my PC into a power meter so that I can see the power draw from the wall outlet.

 

The first time I ran just Furmark on its own and within 5 minutes the screen again went black and the fans spun up and got very loud. For context my PC usually runs very quiet even when under a gaming load. If I leave the PC like this, it does restart on its own after several minutes - I actually don't need to hold the power button to force a restart. What I also found strange was that my power draw immediately dropped when the screen went black. Under just a Furmark load the system draws around 530 watts from the wall. When the screen went black, it dropped to around 210 watts despite the fans sounding like a jet engine. 

 

After this I again ran multiple stress tests including a combined test of Furmark and Prime95 for almost an hour. But nothing happened. The system ran perfectly. HWmonitor shows my CPU getting quite hot and hitting 100 degrees which seems strange for a 14700K hooked up to a 360mm rad. My GPU doesn't even hit 80 degrees and the power draw from the wall is around 760 watts.

 

I did check event viewer after the crash on my first stress test with just Furmark and I have attached a screenshot of all the events I could find that happened at about the same time as the crash.

[Bjoolz]

55 minutes ago, pete6274 said:

Thanks for the tips. I might have to see if I can send my GPU back under warranty. Unfortunately all my friends are console gamers. None of them have desktop PC's. 

 

I have installed Furmark, Prime95, and HWmonitor and run multiple stress tests but still cannot repeat the issue. I have also plugged my PC into a power meter so that I can see the power draw from the wall outlet.

 

The first time I ran just Furmark on its own and within 5 minutes the screen again went black and the fans spun up and got very loud. For context my PC usually runs very quiet even when under a gaming load. If I leave the PC like this, it does restart on its own after several minutes - I actually don't need to hold the power button to force a restart. What I also found strange was that my power draw immediately dropped when the screen went black. Under just a Furmark load the system draws around 530 watts from the wall. When the screen went black, it dropped to around 210 watts despite the fans sounding like a jet engine. 

 

After this I again ran multiple stress tests including a combined test of Furmark and Prime95 for almost an hour. But nothing happened. The system ran perfectly. HWmonitor shows my CPU getting quite hot and hitting 100 degrees which seems strange for a 14700K hooked up to a 360mm rad. My GPU doesn't even hit 80 degrees and the power draw from the wall is around 760 watts.

 

I did check event viewer after the crash on my first stress test with just Furmark and I have attached a screenshot of all the events I could find that happened at about the same time as the crash.

It's BSODing, the screen is just black when it happens which isn't uncommon when the GPU is crashing. 0x133 is DPC_Watchdog_Violation which is usually a driver error, not hardware, but I have seen a lot of cases lately where faulty Nvidia GPUs trip up on ISRs (Interrupt Service Routines) and cause DPC_Watchdog_Violation. So if you could supply the dump files from the crash then I could look at it, but we will of course try fixing the driver before concluding that it's hardware.

 

And just to be clear, there is nothing weird or strange about ISRs. They are done all the time and is a completely normal command. No idea why there is this pattern. 

 

Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

[pete6274]

2 hours ago, Bjoolz said:

It's BSODing, the screen is just black when it happens which isn't uncommon when the GPU is crashing. 0x133 is DPC_Watchdog_Violation which is usually a driver error, not hardware, but I have seen a lot of cases lately where faulty Nvidia GPUs trip up on ISRs (Interrupt Service Routines) and cause DPC_Watchdog_Violation. So if you could supply the dump files from the crash then I could look at it, but we will of course try fixing the driver before concluding that it's hardware.

 

And just to be clear, there is nothing weird or strange about ISRs. They are done all the time and is a completely normal command. No idea why there is this pattern. 

 

Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

There is only 1 file in the minidump. Hope this is correct.

Minidump.zip

[Mumintroll]

2 hours ago, Bjoolz said:

It's BSODing, the screen is just black when it happens which isn't uncommon when the GPU is crashing. 0x133 is DPC_Watchdog_Violation which is usually a driver error, not hardware, but I have seen a lot of cases lately where faulty Nvidia GPUs trip up on ISRs (Interrupt Service Routines) and cause DPC_Watchdog_Violation.

 

Hmm...

Didn't I read somewhere that the 572.xx drivers were causing BSODs some weeks ago, but hasn't that been fixed?

Weren't that mostly on the 50xx cards?

 

I've used the 572 drivers for awhile now on my 4060 and never experienced a crash.

Although I've never installed the Nvidia app, just the drivers if that matters.

 

If that's the case he could try find an older driver from december to try with in meantime, while waiting for Nvidia to fix their driver.

[Bjoolz]

19 hours ago, Mumintroll said:

 

Hmm...

Didn't I read somewhere that the 572.xx drivers were causing BSODs some weeks ago, but hasn't that been fixed?

Weren't that mostly on the 50xx cards?

 

I've used the 572 drivers for awhile now on my 4060 and never experienced a crash.

Although I've never installed the Nvidia app, just the drivers if that matters.

 

If that's the case he could try find an older driver from december to try with in meantime, while waiting for Nvidia to fix their driver.

I've seen 4000 and 5000 series have issues on the latest drivers. Had two people try an older in the last 2-3 days and it got fixed for them (Both on 4000 series), but  don't know exactly which driver they were on before that, they just said it was up to date. 

 

@pete6274 It's crashing exactly like I theorized, on an ISR. Nvlddmkm.sys is the Nvidia driver.

0x133_ISR_nvlddmkm!unknown_function

Try DDU and an older driver. On Nvidia's driver search page, click More Versions at the bottom. I've gone as far as back as August/September with other cases just to be sure it's before any issues with the newer drivers. 

[pete6274]

Really appreciate your guys help @Bjoolz and @Mumintroll.

 

I'll give DDU a go and test an older driver.

 

I'll update this post in a couple days if I haven't had any more crashes.

 

Do you maybe have any ideas how I can test if it works? So far the crashes seem to be random. They have happened more often while streaming, but I've had many streams without any crashes. So it'll be difficult to know for certain if the older driver does the trick.

[Bjoolz]

44 minutes ago, pete6274 said:

Really appreciate your guys help @Bjoolz and @Mumintroll.

 

I'll give DDU a go and test an older driver.

 

I'll update this post in a couple days if I haven't had any more crashes.

 

Do you maybe have any ideas how I can test if it works? So far the crashes seem to be random. They have happened more often while streaming, but I've had many streams without any crashes. So it'll be difficult to know for certain if the older driver does the trick.

With almost every issue, I prefer just using it normally. 

[pete6274]

6 minutes ago, Bjoolz said:

With almost every issue, I prefer just using it normally. 

Fair enough, I'll give it a go for a couple of days and then update this thread.

[pete6274]

@Bjoolz@Mumintroll Well, it's been almost a week and I have tried several older driver versions right back to 560.81 which is the oldest one available on NVIDIA's website. Unfortunately no luck. I still have the same issue on each driver version I tried.

[Bjoolz]

19 minutes ago, pete6274 said:

@Bjoolz@Mumintroll Well, it's been almost a week and I have tried several older driver versions right back to 560.81 which is the oldest one available on NVIDIA's website. Unfortunately no luck. I still have the same issue on each driver version I tried.

Then a faulty GPU is the main suspect. 

[pete6274]

25 minutes ago, Bjoolz said:

Then a faulty GPU is the main suspect. 

That is unfortunate. But at least it's within warranty. Really appreciate your help. I learned a lot with all these troubleshooting steps.