Aug
2nd
2023
I'm an amateur techie, have some troubleshooting questions. · 7:17pm Aug 2nd, 2023
So the past few weeks I've had some issues with my PC randomly restarting in the middle of a gaming session, and I'm not ready to attribute it to the summer heat.
So, I would love some clarity on the following:
- My closed loop CPU cooler (NZXT Kraken 240mm) is four years old. Is this old enough to worry about leaks and/or coolant evaporation?
- Can an overheating GPU cause restarts? The games I play don't always detect that my 4K monitor is underclocked at 60hz, and will instead render the game at 120fps, which of course puts it in the furnace (85-88°C).
Just wondering if I have to double check all my games for frame rate locks, or if I have to look at replacing my CPU cooler.
Thank you for any advice.
Check Event Viewer and look at applications or system and find timestamps that match up for when the restarts happened.
Then trouble shoot the problems there using Google.
To answer your questions specifically, I have no idea, but Event Viewer should give you an idea of what's going on
If nothing is there, then it's not program related but something with your hard ware
5740410
definitely can agree with this user they are absolutely right. sometimes it can be software but if it's something related to that stuff it should show in the event viewer. sometimes hardware errors can appear there too but it's not as common as software.
Quick n' dirty
Yes, but if that had been the case you'd have noticed by now
(the big selling point of AIOs is that they're much less likely to leak). Having said this, if it has been installed for 4 years, a good ol' cleaning (around the case and in the radiator fin stack) and repasting is in order. Also, the pump may not be pumping as hard, which causes loss of efficiency.
Yes, although it should be less likely on Windows since Vista/7 (reworked graphics stack that can reset on its own).
I'd test, in order:
If all this fails to fail, you're in parts cannon territory
(try to bum a PSU from a friend would be my suggestion)
5740416
Leak would be obvious. Coolant evaporation would not be.
Techie friend of mine also said to make sure the cooler is seated. Really hope that's not the problem, because the tube connections are in contact with the RAM sticks because of space issues.
Start by getting your proper temps. If you have a kraken you might be using CAM but I would use HWinfo64 and just its sensor modules on the startup splash screen then see your: pump rpm, cpu temps (cores), disk temps. Just leave it running in the background for a bit.
Cinebench r23 on loop would also saturate your AIO and let you know if its gonna overheat due to old paste or pumps issues.
For ram I personally use Memtest86+ and it needs to make it 9 passes or more not the first pass.
If it is bluescreening when rebooting then get nirsoft bluescreen view to easily see what it thinks is triggering it.
start there and list your temps: ambient, idle, load, and your specs to.
https://www.memtest.org/
https://www.memtest.org/download/v6.20/mt86plus_6.20_USB_Installer.exe or grab the latest iso and use RUFUS
https://www.hwinfo.com/download/
https://www.hwinfo.com/files/hwi_760.exe
https://www.nirsoft.net/utils/blue_screen_view.html
https://www.nirsoft.net/utils/bluescreenview_setup.exe
https://www.maxon.net/en/downloads/cinebench-r23-downloads
EDIT: as said event viewer is good to look at to though can be overwhelming, hit the WIN key and type reliability and open the "view reliability history" "reliability monitor" to help narrow that down too.
5740432
No blue screens, just instant restart during games. Last night I thought it might be because my graphics card was rendering more frames than it needed to (4k120 when monitor is only 4k60) and was getting up to 87-88C.
5740437
normally that wont cause a crash just a slow down of clock speed. But it kicking that much heat out could help overheat another component that is on its thermal edge sometimes. If anything it normally just causes a lot of coil whine. edit: also since its been 4 years might want to blow out your pc and rad to make sure its not getting choked.
5740439
Looked at the Event Viewer. There have been six "Critical" levels, each with Event ID 41, Task Category 63, which says "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
It has occurred roughly every five days for the past 30 days.
July 5th, 10th, 14th, 20th, 25th, and August 1st.
Downloaded HWInfo64. A lot of information, not sure what to do with it all. Also opened the case, checked the cooler, it's seated rock solidly, not the slightest bit of wiggle.
5740452
Look for the temps in the hwinfo64 it's near the middle. I can post screen shots later if you need but I gotta rest atm or I'd help more. If you have down time you should run the memtest 86+ if you can since it takes a long time to do 9 passes. At least rule that out.
5740459
No worries, go get some rest. Memtest86+ sounds like something I'm not prepared to do immediately, I don't have a spare USB drive.
5740437
I've run my laptop in mid 90's before and all it did was take a couple years off its lifespan.
5740495
That's why I'm likely going to start setting aside for a new build. This rig has seen heavy use over four years, and isn't get any newer.
But if custom water cooling really preserves components longer, I see why enthusiasts do it. I'm not brave enough for that.
5740499
No, custom watercooling doesn't do that. It is just pretty, not more efficient. After four year, the thermal past might be too dry, that is maybe something to look into. I myself had overheating problem earlier this year with a several years old NZXT Kraken Cooler, so it might be that too.
The airflow is important as well, make sure there is enough fan and a clear pathway from one side of the case to the other.
A simpler problem might be your power supply. If it is failing, you will never see any info on it. And if you had a few power surge in the last month (flickering light) that might be your problem.