PC Restarting Under Load

[NameIsGreg]

Hi everyone, first time poster here who has been dealing with this issue for a while and am now officially stuck and am feeling a bit hopeless.

When playing games my PC sometimes restarts, it is extremely erratic and sometimes won't crash for days but usually when I get a crash, their will be a flurry of them. I recently bought a new PSU as this is usually the main culprit for an issue like this but my PC has restarted again. I exchanged all the main power cables when installing my new PSU but the fault still happens. The thing that prompted me to buy a new PSU was that my PCie cables that attach to my gpu (2x8pins), 1 of them had been getting very hot and had melted a bit of the plastic (ill put some imgs in the google drive). I'm in the process of getting some Isopropyl alcohol to try and clean the connection to see if that does anything but I'm not sure it will.

I have tried lots of things to try and get to the route of the issue but to no avail. Any testing software i use like Furmark or Heaven do not do anything, I've ran both for multiple hours at 100% load and it didnt crash. I've also tried memtest and had 0 errors after a 4-5hr test. I have factory reset my computer multiple times to try and erase any possibility of it being a software issue as well as performing DDU and driver verifier. Another thing to note could be that I don't get minidump files when the system crashes and there isnt anything (at least that i can see) in event viewer before the crash. Sometimes the crashes are coupled with certain errors like vol mgr 161.

It has recently started crashing playing games like Factorio or Football Manager which are not demanding at all, Football Manager can be a bit CPU demanding but Factorio does not use up much computing power at all especially on GPU side.

I did a couple of HWINFO Tests to see if i could find anything whilst crashing but the only thing that stood out was the 12V sometimes going around 11.4-11.5 but it didnt seem to happen frequently enough for that to be the culprit. I'll attach the logs but not sure if that will be of much use: https://drive.google.com/drive/folders/1VOpzkUzIJ1ouj-fUKIedpGJ9lp8Hbo7l?usp=drive_link

I'd really love some help from the community here as I just feel a bit lost and am not sure what to do. More than happy to run some sensor tests or anything you need to try and get to the bottom of this.

Edit: One thing I forgot to add is in the last few hours I enabled DOCP in my bios to try and get my RAM to work a bit faster and it crashed a lot (as soon as I open a game and sometimes get various BSOD when idling), all I did was turn on DOCP and leave all the default settings.

Specs:
PSU: both were Corsair RM850x
Mobo: ROG x570e gaming wifi ii
RAM: 4 x DDR4 Corsair Vengeance Pro (3600mHz)
CPU: Ryzen 9 5950x
GPU: RTX 3080 iChill
Storage: 2x1TB and 1x2TB

 

[BlindZenDriver]

The PSU would also have been my first swap, but when that didn't solve the issue it get harder to pick. Also it sounds like you have done what is possible with just software.
Here is something more you can try:

• Try running with only two sticks of ram.
• Check the paste on your CPU, make sure it is only a thin layer and that it covers the whole CPU. And also that the cooler is level so it coves the whole CPU.
• Also check the paste on your GPU including its ram, however I would say that is a long shot unless you have a lot of hours on the GPU or the cooling is some custom thing.
• Try running just the main storage drive if possible.

 

[NameIsGreg]

The PSU would also have been my first swap, but when that didn't solve the issue it get harder to pick. Also it sounds like you have done what is possible with just software.
Here is something more you can try:

• Try running with only two sticks of ram.
• Check the paste on your CPU, make sure it is only a thin layer and that it covers the whole CPU. And also that the cooler is level so it coves the whole CPU.
• Also check the paste on your GPU including its ram, however I would say that is a long shot unless you have a lot of hours on the GPU or the cooling is some custom thing.
• Try running just the main storage drive if possible.

I just made an edit to my main post:

Edit: One thing I forgot to add is in the last few hours I enabled DOCP in my bios to try and get my RAM to work a bit faster and it crashed a lot (as soon as I open a game and sometimes get various BSOD when idling), all I did was turn on DOCP and leave all the default settings.

Thanks for the information, I will give some of these things a go now and will let you know how I get on.

 

[Contra]

Disable CPU and RAM OC

 

[xtreemchaos]

mine dos the same i do a all core overclock of 4.1 to sort it. its a pain in the ass but a lot 5950 suffer from it.
edit- ive tryed it in 3 diff x570 boads and just the same but works norm in a ROG b550e gaming i think i4s down to driver but not sure.

 

[Athlonite]

but the only thing that stood out was the 12V sometimes going around 11.4-11.5

That's not a good sign you shouldn't be seeing that much V-Droop on the 12V rail try another PSU

 

[micropage7]

have you tried resetting the bios first then check it again, it may from the psu or could be from processor or GPU that draws power too much more than the psu can handle it

 

[NameIsGreg]

That's not a good sign you shouldn't be seeing that much V-Droop on the 12V rail try another PSU

I bought a new PSU and swapped out the GPU PCie cables and it still crashed.

 

[NameIsGreg]

have you tried resetting the bios first then check it again, it may from the psu or could be from processor or GPU that draws power too much more than the psu can handle it

I've updated bios a few times to try and help the issue but never reset it. It cant be a PSU issue because I got a new one and it still crashed but I've stress tested my CPU and GPU quite a lot and no crashes so would be odd if it was one of those things.

 

[NameIsGreg]

Disable CPU and RAM OC

I have no OC on, everything is default to try and mitigate the issue as much as possible.

mine dos the same i do a all core overclock of 4.1 to sort it. its a pain in the ass but a lot 5950 suffer from it.
edit- ive tryed it in 3 diff x570 boads and just the same but works norm in a ROG b550e gaming i think i4s down to driver but not sure.

I've updated my chipset drivers a few times as well but nothing changed. I've never really messed around with any overclocking, could you give me a bit more info on how I can do this?

 

[FreedomEclipse]

I would try two sticks of ram.

I wonder if its the memory controller thats having a fit with 4 sticks of ram. Alternatively. You can run 4 sticks but run them at their slower than rated speed. So instead of 3600. You run them at 3200 instead and see if that changes anything.

For higher performance systems or people that like to tweak their machines its better to run with 2 memory modules instead of 4 because it stresses the memory controller out less... Unless you are running a HEDT platform that uses quad channel. You should be sticking with 2x8GB or 2x16/2x32Gb etc etc to reach the ram capacity that you want/need etc.


This is just how the cookie crumbles unfortunately. Its been this way for a long time.

 

[NameIsGreg]

I would try two sticks of ram.

I wonder if its the memory controller thats having a fit with 4 sticks of ram. Alternatively. You can run 4 sticks but run them at their slower than rated speed. So instead of 3600. You run them at 3200 instead and see if that changes anything.

For higher performance systems or people that like to tweak their machines its better to run with 2 memory modules instead of 4 because it stresses the memory controller out less... Unless you are running a HEDT platform that uses quad channel. You should be sticking with 2x8GB or 2x16/2x32Gb etc etc to reach the ram capacity that you want/need etc.


This is just how the cookie crumbles unfortunately. Its been this way for a long time.

Thanks for the reply,

I had my ram at default speeds (2133) for a long time and the issue was still happening but yesterday I enabled DOCP and went up to 3600 and it started crashing a lot. I have just taken two sticks out as someone else said this and I'll do a test at 3200 to see what happens.

I had some BSOD which I never get but I also had one of my normal crashes which just restarts the PC, could that be an indication that maybe the ram is at fault?

 

[Waldorf]

@NameIsGreg
please fill out the system info in your account profile, make sure to tick the box to show it publicly.

having swapped psu already, and given it happens with DOCP, its either from the board using AUTO (too much/not enough V for something; applying to tight timings etc),
or because you are using 4 sticks, which on asus can mean lower clocks to be stable.
you can use something like memtest 64 (not 86/86+)/ TM5 or HCIK to test for ram stability with docp enabled, or (better), set clock/primary timings and dram/soc voltages manually,
which is the preferred method for Ryzen (at least <AM5).

 

[NameIsGreg]

@NameIsGreg
please fill out the system info in your account profile, make sure to tick the box to show it publicly.

having swapped psu already, and given it happens with DOCP, its either from the board using AUTO (too much/not enough V for something; applying to tight timings etc),
or because you are using 4 sticks, which on asus can mean lower clocks to be stable.
you can use something like memtest 64 (not 86/86+)/ TM5 or HCIK to test for ram stability with docp enabled, or (better), set clock/primary timings and dram/soc voltages manually,
which is the preferred method for Ryzen (at least <AM5).

Thanks for the reply,

I have just filled in my system specs so you should be able to see them now .

Do you think you could give me some extra info on what I should set those two? I've never really messed around with bios before as its quite daunting to a newcomer.

Is it just these things on my ram sticks: 3600MHz, 18-22-22-42, 1.35 V for dram voltage?

 

[FreedomEclipse]

Thanks for the reply,

I had my ram at default speeds (2133) for a long time and the issue was still happening but yesterday I enabled DOCP and went up to 3600 and it started crashing a lot. I have just taken two sticks out as someone else said this and I'll do a test at 3200 to see what happens.

I had some BSOD which I never get but I also had one of my normal crashes which just restarts the PC, could that be an indication that maybe the ram is at fault?

are you running mixed ram sets? mixed speeds or brands?

 

[NameIsGreg]

are you running mixed ram sets? mixed speeds or brands?

Nope, all are Corsair Vengeance Pro 8gb with a max speed of 3600MHz

edit: also in their correct pairs as well.

 

[Superzuber]

Had similar issue, with random resets, it looked like PSU problem, no issues found with stress test's and in the end was my RTX 3070ti
It was hard to diagnose or isolate the issue I had put the card in my friend's computer and he had no issues for 2 days and 3rd day he had 2 random resets with it

 

[_roman_]

Edit: One thing I forgot to add is in the last few hours I enabled DOCP in my bios to try and get my RAM to work a bit faster and it crashed a lot (as soon as I open a game and sometimes get various BSOD when idling), all I did was turn on DOCP and leave all the default settings.

Stable test your platform first. Check your temperatures externally. I would instantly get rid of that power supply unit.


Stable testing on windows is difficult and expensive.

Windows 10 pro with my previous ryzen 5800x and 6600XT graphic card will run for 10 hours fine with unstable DRAM settings. It crash very fast with the gnu userspace when i compile code. The best and fastest way to determine bad settings in the mainboard is compiling libreoffice with the gnu toolchain - not in windows. Windows - epic games - gaming stable means nothing.

PSU: both were Corsair RM850x

Note: i had a stable system which was long term compiler stable. After a while the system had one or two random crashes per week. For several months.

I had a faulty Corsair RM750 after 18 months of usage. It was very hard to determine the error with my previous ryzen 5800x and 6600XT graphic card. I had for months issues.

I picked up a enermax revolution D.f. 750w and loosely wired from the glass side panel side the new power supply unit with all the new cables. I tested for two weeks and the issue was gone.


Note: Amazon De displays it wrong - one page for all CORSAIR PSU - but with wrong text on the order view page.

You last purchased this item on 17 Aug 2021
Size Name: 750 Watt | Style Name: RM Series | Colour Name: black

Corsair RM750 PSU = Trash = Garbage

edit: 750 Watts was far too much for my previous build or my current build. The box crashed with one www-client open and basic gnu desktop. I assume the load on the psu less than 100 Watts. I only changed the PSU and got rid of that issue. Never an issue again.

 

[phanbuey]

DDR5 ram stability is so much more difficult to test than prior gens. It can stay stable for weeks and even pass extended stress tests and still crash randomly.

 

[NameIsGreg]

DDR5 ram stability is so much more difficult to test than prior gens. It can stay stable for weeks and even pass extended stress tests and still crash randomly.

I have DDR4 RAM

 

[phanbuey]

I have DDR4 RAM

what are you using to test it? - have you tried Karhu Ram Test?

 

[NameIsGreg]

what are you using to test it? - have you tried Karhu Ram Test?

No I haven't, I'll give that a go. I've been using Memtest64 and Prime95.

Also did a memtest86 test a while back but it reported 0 errors.

 

[phanbuey]

No I haven't, I'll give that a go. I've been using Memtest64 and Prime95.

Also did a memtest86 test a while back but it reported 0 errors.

give it a try - kahru ramtest and GpuMemTest - https://www.programming4beginners.com/gpumemtest -- just to rule these out.

volmgr - is volume manager - its usually just windows whining that the volume has data corruption from a crash.

It really sounds like the rig is stable at full load (prime, furmark) but not in the transient states - where there's either instability somewhere along the curve - or from something else - bad ram / incompatibility (Corsair Vengance strikes again) -- i've also seen bad SATA and NVMe devices do this - one time it was as simple as a loose sata cable.

 

[3x0]

Are there any .dmp files in C:\Windows\LiveKernelReports ? If yes, then archive them and upload them here.

Also, just to make sure, when you were swapping PSUs/cables, you only used cables that came with their respective PSU, not interchanging them?

 

[NameIsGreg]

Are there any .dmp files in C:\Windows\LiveKernelReports ? If yes, then archive them and upload them here.

I have two folders within LiveKernelReports - USBHUB3 and WATCHDOG but both have nothing inside

Also, just to make sure, when you were swapping PSUs/cables, you only used cables that came with their respective PSU, not interchanging them?

I only used the power cables that were provided with the new PSU, no interchanging of old and new power cables.