PC is freezing and just restarts

demon8991

demon8991

demon8991

Soldato
Joined
7 Jan 2003
Posts
4,458
Location
Gold Coast, Australia
Hello All,

I am having some problems with the PC in the spec of my signature, PSU is the stock one that came with the case.

PC has worked completely fine for 10 months no matter what has been thrown at it with the following settings in the BIOS:
  • RAM set to DOCP
  • PBO Curve Optimiser set to negative 20
  • PCI Express set to PCIE 3 (due to riser)

All of a sudden a few weeks ago it started freezing in desktop and then just restarting, particularly noticeable when cold and under no particular stress.

I have two Windows 10 installs on the SSD one for work and one for personal, both OSs are having the problems.

Things I have tried:
  • Ran a MEMTEST over night and that came back fine.
  • Changed the PBO Curve Optimiser to negative 15.
  • Latest Nividia Drivers
  • All Windows Updates installed
  • Latest BIOS and subsequently rolled back
Had several lockups this morning including loading windows and twice in the BIOS itself!!! I have now removed the PBO Curve Optimiser and it has been fine since (20 mins)

I still suspect, however, that there is a problem, could it be my PSU, Motherboard or CPU failing?
 
Sorry, I mentioned that in the first line, it is the stock PSU that came with NZXT H1.

Don't know a huge amount about it other than it is a 650W Gold PSU.

Also after running all day, i switched it off after work at 5pm, turned PC back on at 9pm and it froze again.
 
demon8991 said:
Sorry, I mentioned that in the first line, it is the stock PSU that came with NZXT H1.

Don't know a huge amount about it other than it is a 650W Gold PSU.

Also after running all day, i switched it off after work at 5pm, turned PC back on at 9pm and it froze again.
Click to expand...
Ah sometines read to fast.

There was something about the Nxzt h1 gpu riser cable causing problems mite be worth looking into that.

650w for a 5900x and 3090 is close to thd limit 3090 can power spikd yo 400w but mitd not be the cause as problem happens whdn not under load.
 
mickyflinn said:
Ah sometines read to fast.

There was something about the Nxzt h1 gpu riser cable causing problems mite be worth looking into that.

650w for a 5900x and 3090 is close to thd limit 3090 can power spikd yo 400w but mitd not be the cause as problem happens whdn not under load.
Click to expand...
Id say that's under... Min requirement for 3080 is 750w psu, now you have a 3090? with a 5900x...didnt linus use a 1000w psu and swap it out for a 1200w when they stress tested it all. Thats overkill but think 850w to be safe
 
Craig_d1 said:
Id say that's under... Min requirement for 3080 is 750w psu, now you have a 3090? with a 5900x...didnt linus use a 1000w psu and swap it out for a 1200w when they stress tested it all. Thats overkill but think 850w to be safe
Click to expand...
Yes i agree though some people have been getting away wth 650w but i wouldn't want to risk it.
 
mickyflinn said:
Ah sometines read to fast.

There was something about the Nxzt h1 gpu riser cable causing problems mite be worth looking into that.

650w for a 5900x and 3090 is close to thd limit 3090 can power spikd yo 400w but mitd not be the cause as problem happens whdn not under load.
Click to expand...

Thanks, I have the new riser cable in my unit so that shouldn't be a problem.

With regard to the PSU having always ran a negative curve and an undervolted on my GPU I was well within 650w.

Craig_d1 said:
Id say that's under... Min requirement for 3080 is 750w psu, now you have a 3090? with a 5900x...didnt linus use a 1000w psu and swap it out for a 1200w when they stress tested it all. Thats overkill but think 850w to be safe
Click to expand...

mickyflinn said:
Yes i agree though some people have been getting away wth 650w but i wouldn't want to risk it.
Click to expand...

Do you think that's the best place to start or is the motherboard cooked?

As mentioned previously I have had the GPU undervolted since I got it and the curve optimiser at negative 20 so the draw was a reasonable amount less.

In terms of SFX power supplies what do you recommended, I note Coolermaster have an 850w Gold and Corsair have a 750w Platinum.

Joxeon said:
Try turning off the curve optimizer.

Edit: seen you have already done this so I'd see how it goes.
Click to expand...

Yeah I have had two crashes with the curve off.
 
Joxeon said:
Any errors in event viewer in and around the kernel power event which happens when the system crashes?.
Click to expand...

Nothing that really sticks out, this might be the only one, however I don't know much about what these events mean.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffffde04f9515010, 0xfffff807278419e4, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 0e481043-3943-4084-9466-97e690ee39da.

The bit I am finding strange is these freezes only happen when computer is cold and under minimal stress.
 
Do you have another gpu to try ?

The error you qouted could be Gpu driver reset crash but the computer locking up in the bios can be a psu ?

So when you say cold does that mean its fine when warmed up?

Games load and you can play without crashes
 
mickyflinn said:
Do you have another gpu to try ?

The error you qouted could be Gpu driver reset crash but the computer locking up in the bios can be a psu ?

So when you say cold does that mean its fine when warmed up?

Games load and you can play without crashes
Click to expand...

I don't unfortunately, yeah everything works fine once warmed up.

Regarding GPU drivers, I have two different versions installed on each OS as the work one doesn't really need the latest driver, however the personal one is fully updated.
 
Just had another restart this time when the PC was warmed up, been on and used for 5 hours.

3 Errors Event Viewer:
  • Error setting traits on Provider {8444a4fb-d8d3-4f38-84f8-89960a1ef12f}. Error: 0xC0000001
  • The computer has rebooted from a bugcheck. The bugcheck was: 0x0000010e (0x0000000000000037, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: b1e6e509-2092-4d6d-975f-c15a066a5e0d.
  • The previous system shutdown at 11:22:58 AM on ‎1/‎10/‎2021 was unexpected.
It is worth mentioning I have had HW Monitor Running all morning to keep an eye on Power Draw, the CPU never exceeded 140w and then GPU never exceeded 120w. Typically whilst at work its the CPU that gets a work out as I deal with large complex datasets.

It seems my problem is now becoming more frequent.
 
Last edited:
demon8991 said:
Just had another restart this time when the PC was warmed up, been on and used for 5 hours.

3 Errors Event Viewer:
  • Error setting traits on Provider {8444a4fb-d8d3-4f38-84f8-89960a1ef12f}. Error: 0xC0000001
  • The computer has rebooted from a bugcheck. The bugcheck was: 0x0000010e (0x0000000000000037, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: b1e6e509-2092-4d6d-975f-c15a066a5e0d.
  • The previous system shutdown at 11:22:58 AM on ‎1/‎10/‎2021 was unexpected.
It is worth mentioning I have had HW Monitor Running all morning to keep an eye on Power Draw, the CPU never exceeded 140w and then GPU never exceeded 120w. Typically whilst at work its the CPU that gets a work out as I deal with large complex datasets.

It seems my problem is now becoming more frequent.
Click to expand...

Been doing a search on error, codes, seems others had same and always goes back to gpu driver issue, last post I read, guy fixed it by going to an older driver

Did you try WinDbg which will analyse your data dump
Getting Started with Windows Debugging - Windows drivers | Microsoft Docs
 
A few sites can help..have a read through 1st

https://support.robertsspaceindustr...-Nvidia-Graphic-Drivers-Install-and-Configure


Preferably I'd just go down the above route of Uninstall program...add/remove program route 1st

or could use something like below DDU..uninstall in windows safe mode...have a good read 1st, and do exactly as stipulated

https://www.guru3d.com/files-details/display-driver-uninstaller-download.html

link to latest driver for 3090 below..you can download it prior to uninstall of old drivers, but don't run it till other drivers have been removed

https://www.nvidia.co.uk/Download/driverResults.aspx/180342/en-uk
 
Craig_d1 said:
A few sites can help..have a read through 1st

https://support.robertsspaceindustr...-Nvidia-Graphic-Drivers-Install-and-Configure


Preferably I'd just go down the above route of Uninstall program...add/remove program route 1st

or could use something like below DDU..uninstall in windows safe mode...have a good read 1st, and do exactly as stipulated

https://www.guru3d.com/files-details/display-driver-uninstaller-download.html

link to latest driver for 3090 below..you can download it prior to uninstall of old drivers, but don't run it till other drivers have been removed

https://www.nvidia.co.uk/Download/driverResults.aspx/180342/en-uk
Click to expand...

Thanks for that, however is this not the newest driver?

https://www.nvidia.com/Download/driverResults.aspx/180557/en-us
 
Ok so just to let you all know what I have done since the last freeze and restart this morning and then the second one that I had in the afternoon.
  • Opened case up, checked all connections
  • Disconnected and reconnected the GPU
  • Reset CMOS
  • Uninstalled all GPU drivers through "Programs and Features" (On Personal OS)
  • Ran DDU in Safemode to ensure everything was removed.(On Personal OS)
  • Installed the latest NVIDIA driver 472.12 (On Personal OS)
No freezes yet but its very random, fingers crossed this has fixed the issue.
 
mickyflinn said:
@demon8991 everything still stable ?
Click to expand...

8YEwdnz.gif
 
You must log in or register to reply here.
Back
Top Bottom