New pc - random reboots

[FreedomEclipse]

Problem is the retailer will tell me they have to start an RMA process and then it takes like 3 weeks minimum to get information if its approved or not... They wont give me money back immediately.

depending on what part of the world you are from. some retailers will give you a two week grace period in which you can ask for a refund no questions asked. You should check if your retailer does the same thing. Most will give you a week, unless you are past that threshold.

 

[Vya Domus]

It's probably not the PSU, no display output after a reboot it's just not something you get with that type of issue, something else is causing that.

 

[dashful]

depending on what part of the world you are from. some retailers will give you a two week grace period in which you can ask for a refund no questions asked. You should check if your retailer does the same thing. Most will give you a week, unless you are past that threshold.

Im in croatia. I already RMA'd a product through them and they told me every product has to start an RMA no matter if i want a replacement or a refund. Something has to be wrong with it.

It's probably not the PSU, no display output after a reboot it's just not something you get with that type of issue, something else is causing that.

Any ideas?

 

[Vya Domus]

Like I said RAM most likely, run it without XMP and use memtest86.

 

[dashful]

I haven't mentioned. My friend also bought a power extension cord with surge protection. Now it was fairly cheap so could this also be a possibility?

Like I said RAM most likely, run it without XMP and use memtest86.

depending on what part of the world you are from. some retailers will give you a two week grace period in which you can ask for a refund no questions asked. You should check if your retailer does the same thing. Most will give you a week, unless you are past that threshold.

 

[Laurijan]

I haven't mentioned. My friend also bought a power extension cord with surge protection. Now it was fairly cheap so could this also be a possibility?

try without. some tips for playstation 5 shutting down by itself tell to plug it directly into the wall.

 

[evernessince]

You can always run memtest86+ from an USB-stick to test ram. Wont hurt to check harddisk health with crystaldiskinfo (really only need S.M.A.R.T data).

^ This. Plus get a log of your CPU, GPU, and Mobo temps as recommended earlier.

Does the system only crash under load or is it more persistent? Does the system crash if you were to limit the FPS?

I haven't mentioned. My friend also bought a power extension cord with surge protection. Now it was fairly cheap so could this also be a possibility?

It's possible. I'd try the system directly plugged into the wall just to make sure.

 

[QuietBob]

@dashful
What are the temperatures like under heavy load? The board has no VRM heatsinks, maybe thermal protection is kicking in?

 

[eidairaman1]

Problem is the retailer will tell me they have to start an RMA process and then it takes like 3 weeks minimum to get information if its approved or not... They wont give me money back immediately.

Start saving up

 

[Solaris17]

Kernel power 41 is actually usually not the PSU. It’s simply an indicator the kernel itself has crashed and forced CATERR.

I wrote about it here on the forums years ago.

Anyway run GPU-Z in logging mode and watch VRAM and RAM memory usage. You already posted your kernel error you just don’t seem to have known it.

RADAR_PRE_LEAK_64

Something is leaking memory. So bad windows memory leak detection has caught it. Instead of filling page file (maybe it’s off) it appears to be instead corrupting the memory map of whatever program is doing it.

My guess heap corruption.

Given it’s this game specifically I’m willing to bet it’s the problem. But graphs will tell.

In either case updating drivers, the software would be first order. Then probably file system check along with game file validation. Finally, compatibility mode may help.

It goes without saying but if you are using any mods or third party software with the game these are also massive culprits.

GL

 

[dashful]

Kernel power 41 is actually usually not the PSU. It’s simply an indicator the kernel itself has crashed and forced CATERR.

I wrote about it here on the forums years ago.

Anyway run GPU-Z in logging mode and watch VRAM and RAM memory usage. You already posted your kernel error you just don’t seem to have known it.



Something is leaking memory. So bad windows memory leak detection has caught it. Instead of filling page file (maybe it’s off) it appears to be instead corrupting the memory map of whatever program is doing it.

My guess heap corruption.

Given it’s this game specifically I’m willing to bet it’s the problem. But graphs will tell.

In either case updating drivers, the software would be first order. Then probably file system check along with game file validation. Finally, compatibility mode may help.

It goes without saying but if you are using any mods or third party software with the game these are also massive culprits.

GL

I left the game on for about 2 hours at 400 fps, now he's been on the game for like 3 more hours and thats like 5 hours with no crashes. Im honestly not sure what to think anymore. I already scanned the pc cor corruptions, the driver is the latest amd driver for 7600xt. It also happened while he was playing Ironsight (same error ID).

 

[Solaris17]

I left the game on for about 2 hours at 400 fps, now he's been on the game for like 3 more hours and thats like 5 hours with no crashes. Im honestly not sure what to think anymore. I already scanned the pc cor corruptions, the driver is the latest amd driver for 7600xt. It also happened while he was playing Ironsight (same error ID).

Idling isn’t the same as what you were doing though. For all you know the issue is in the active player function.

 

[dashful]

Idling isn’t the same as what you were doing though. For all you know the issue is in the active player function.

Will try plugging it to the wall as the power extension cord could actually be the problem. Though, i will also use the GPU-Z and HWinfo logging to try and see what causes this.

 

[eidairaman1]

Will try plugging it to the wall as the power extension cord could actually be the problem. Though, i will also use the GPU-Z and HWinfo logging to try and see what causes this.

Extension cords are never suggested for use.

 

[dashful]

Extension cords are never suggested for use.

Where else will you plug three different power cables if you only have one outlet available from the wall?

 

[eidairaman1]

You are talking about an extension cord right?

Where else will you plug three different power cables if you only have one outlet available from the wall?

 

[Dr. Dro]

Blue screen of death (STOP error) information in dump files.

Displays information about blue screen crashes occured on your system. (MiniDump Reader)

www.nirsoft.net

Download this and see what driver is responsible for the crash.

1. If it's ntoskrnl.exe, likely it's bad RAM. In this case, try booting your machine with the memory speed at standard (XMP off), and see if the problem is reproduced. If so, go to step 3.
2. If it's amdkmdag.sys, atikmdag.sys, aticfx64.sys or similar - this is the Radeon GPU driver which is notorious for crashing like this. If this is the case, then go to step 4.
3. Download OCCT from https://www.ocbase.com/ - restart your machine, re-enable XMP, increase voltage on the memory in 0.05 steps (max 1.45 V) and run its memory test until the problem goes away. You can use more specialized tools to ensure stability later, but this will do for the time being.
4. Downgrade the Radeon driver to an earlier release. I suggest 24.5.1, which is from an older development branch. See if this stops your machine from acting up.

https://www.amd.com/en/resources/support-articles/release-notes/RN-RAD-WIN-24-5-1.html

After troubleshooting is completed, fix system corruption by opening cmd as administrator and typing in:

dism /online /cleanup-image /restorehealth
sfc /scannow
shutdown /r /t 0 (this command will reboot your machine cleanly!)

 

[dashful]

Blue screen of death (STOP error) information in dump files.

Displays information about blue screen crashes occured on your system. (MiniDump Reader)

www.nirsoft.net

Download this and see what driver is responsible for the crash.

1. If it's ntoskrnl.exe, likely it's bad RAM. In this case, try booting your machine with the memory speed at standard (XMP off), and see if the problem is reproduced. If so, go to step 3.
2. If it's amdkmdag.sys, atikmdag.sys, aticfx64.sys or similar - this is the Radeon GPU driver which is notorious for crashing like this. If this is the case, then go to step 4.
3. Download OCCT from https://www.ocbase.com/ - restart your machine, re-enable XMP, increase voltage on the memory in 0.05 steps (max 1.45 V) and run its memory test until the problem goes away. You can use more specialized tools to ensure stability later, but this will do for the time being.
4. Downgrade the Radeon driver to an earlier release. I suggest 24.5.1, which is from an older development branch. See if this stops your machine from acting up.

https://www.amd.com/en/resources/support-articles/release-notes/RN-RAD-WIN-24-5-1.html

After troubleshooting is completed, fix system corruption by opening cmd as administrator and typing in:

dism /online /cleanup-image /restorehealth
sfc /scannow
shutdown /r /t 0 (this command will reboot your machine cleanly!)

Thank you so much. I will try this tomorrow and will follow up with an update.

 

[dashful]

Blue screen of death (STOP error) information in dump files.

Displays information about blue screen crashes occured on your system. (MiniDump Reader)

www.nirsoft.net

Download this and see what driver is responsible for the crash.

1. If it's ntoskrnl.exe, likely it's bad RAM. In this case, try booting your machine with the memory speed at standard (XMP off), and see if the problem is reproduced. If so, go to step 3.
2. If it's amdkmdag.sys, atikmdag.sys, aticfx64.sys or similar - this is the Radeon GPU driver which is notorious for crashing like this. If this is the case, then go to step 4.
3. Download OCCT from https://www.ocbase.com/ - restart your machine, re-enable XMP, increase voltage on the memory in 0.05 steps (max 1.45 V) and run its memory test until the problem goes away. You can use more specialized tools to ensure stability later, but this will do for the time being.
4. Downgrade the Radeon driver to an earlier release. I suggest 24.5.1, which is from an older development branch. See if this stops your machine from acting up.

https://www.amd.com/en/resources/support-articles/release-notes/RN-RAD-WIN-24-5-1.html

After troubleshooting is completed, fix system corruption by opening cmd as administrator and typing in:

dism /online /cleanup-image /restorehealth
sfc /scannow
shutdown /r /t 0 (this command will reboot your machine cleanly!)

Ok so one issue here is... I dont get a blue screen so I dont see how this software would ever read the minidump files when there arent any. I ran the program but it doesnt show any minidump files anyway.

 

[Dr. Dro]

Ok so one issue here is... I dont get a blue screen so I dont see how this software would ever read the minidump files when there arent any. I ran the program but it doesnt show any minidump files anyway.

Some blue screens don't show (PC just reboots), happens sometimes. But if you don't have any minidump files, proceed with the memory testing anyway

 

[Laurijan]

Here a guide how to stop Windows from rebooting upon bluescreen How to Disable Windows Automatic Restart on System Failure (lifewire.com) so that you have time to check error code

 

[dashful]

Here a guide how to stop Windows from rebooting upon bluescreen How to Disable Windows Automatic Restart on System Failure (lifewire.com) so that you have time to check error code

Tried, no difference. The PC now shuts down, no errors. No minidump files either.

Some blue screens don't show (PC just reboots), happens sometimes. But if you don't have any minidump files, proceed with the memory testing anyway

Ran memory testing for 2 hours, nothing happened. Played a game for 20 mins (Marvel's Spiderman this time), pc just shut down. Also uninstalled drivers with DDU and installed the older driver u mentioned.

Edit: Ran a DxDiag report right after the shutdown, this is the last event:



Windows Error Reporting:
+++ WER0 +++:
Group that caused the error: 1573987189374387487; Type: 5
Event name: RADAR_PRE_LEAK_64
Response: Not Available
ID CAB-a: 0

Problem signature:
P1: Spider-Man.exe
P2: 1.812.1.0
P3: 10.0.22631.2.0.0
P4:
P5:
P6:
P7:
P8:
P9:
P10:

Datoteke u privitku:
\\?\C:\Users\User\AppData\Local\Temp\RDRDDF3.tmp\empty.txt
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.07243b73-7439-4462-bb45-d9fcace79d97.tmp.WERInternalMetadata.xml
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.d82dcf9b-b45a-4f43-b303-b50914db5b4c.tmp.csv
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.977cb560-6ebc-4aac-845d-ffe21ddfa083.tmp.txt
\\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER.060cfe94-b45a-4c73-9247-8bda5bf3595d.tmp.xml

These files may be located here:
NULL

 

[Dr. Dro]

If the memory test did not catch anything in 2 hours, considered your memory is not overclocked or overtightened, it's probably OK. Try maybe an even older version, maybe the one from March (24.3.1). Shooting blind here, really. Seems like you got some memory heap corruption issue that's nasty and persistent.

I'd try taking one memory stick off, if it stops, trying the other one. Reinstall Windows as a last resort.

 

[dashful]

Iw

If the memory test did not catch anything in 2 hours, considered your memory is not overclocked or overtightened, it's probably OK. Try maybe an even older version, maybe the one from March (24.3.1). Shooting blind here, really. Seems like you got some memory heap corruption issue that's nasty and persistent.

I'd try taking one memory stick off, if it stops, trying the other one. Reinstall Windows as a last resort.

I will probably just reinstall windows... I installed it weirdly as I didn't have a USB stick so i made an empty partition and made it bootable and went from there. Memtest showed nothing so I literally have no idea what to check. If only the PC didnt shut down atleast I wouldnt have to bother so much. I reseated the ram sticks, plugged in a different pcie power cable into the gpu, rechecked all cable connections, checked disk health, scanned windows for corruptions etc.
My only shot is just taking it to a repair shop and asking them to test it out or something... I have no other clue what to do. My friend knows very little about PCs (hes literally technologically illiterate) about every single thing so im trying my best to fix it for him.

 

[Laurijan]

Sound more and more like a powersupply issue to me, if the PC shuts off without staying at bluescreen (if its set up like I linked)