Quick question. This computer is getting pretty old now…MSI i9900k with a 2080Ti that has been fantastic for years. The last couple of months I’ve been experiencing some fairly predictable shutdowns when I’m taxing the computer hard. It believe it only happens when I’m in VR…and for a bit I thought it was only when I was playing VTOL VR…but tonight it happened while playing DCS.
I took the case off a couple weeks ago and used some compressed air to blow out all the dust…and thought I might have solved the problem…but it just reared its head again tonight. I even have a floor fan blowing on it. I can tell it is getting hot…but when I pull up Dragon Center after the shutdown…the CPU has probably already cooled out of the danger zone…and it doesn’t show GPU temperatures (that I can tell).
Anyway what I’m looking for is a data monitoring tool (hopefully free) that tracks the CPU/GPU temperatures and writes that data to a file that won’t corrupt when the inevitable shutdown occurs. The entire computer just vomits and shuts off…so there is no time to watch anything… Does such a program exist?
I posted on this in the…err some thread I forget the name of now…as I too had an older (it’s 10 years old now) rig: the OEM paste seemed to be the ticket. Took about 20 minutes to do. Cleaning all the dust-bunnies helped but not as much as the new paste job. Worth a try. Note you can’t pull it off and put it back on again (with the same paste).
Cool thanks. I just ran some web-based stress tests to see what would happen. When my computer shuts off…it kind of goes dormant for a bit and then resists being restarted…but I think I can hear that the CPU is actually back on…but the GPU is not sending a signal to my monitor. So I think the GPU is the one overheating perhaps…and then the computer must have a temperature monitor that basically says…uh…no…we aren’t going to enable that part because we are protecting you from damaging anything.
Any-who…I ran the stress tests and the computer did not shut down…but I have no idea if these temperatures are “normal” or not.
Yeah, a 100°C on the CPU and it will start throttling down and eventually shut down to protect itself.
Be careful with air duster cans as it is entirely possible to damage parts by directing high power airflow onto them.
Don’t come too close with the nozzle.
What kind of cooler do you have on that CPU?
These All In One liquid coolers, for instance, can run out of liquid due to evaporation.
In my experience (220VAC power) PSUs can have a mode as they get older where the voltage sag under heavy load causes problems - it doesn’t take much for the 3.3V rail that powers the RAM to go low and cause a crash, but it doesn’t seem like that’s your problem if it’s going to power off rather than a reboot.
While you’ve got it apart to change the paste it’s probably a good idea to reseat the CPU (after cleaning off the paste!) and the RAM… if this doesn’t work maybe see if you can borrow a PSU to test?
It is the stock cooler…I had to take it off within a week of buying the computer back in 2019 in order to upgrade the RAM or put in a SSD or something…
Like place names, it will vary with where you come from.
Heading back down the rabbit hole of US steam locomotives after many decades of not thinking about it… still coming to terms with “Lima” (a name I only ever saw written) being said “Lyme-uh” rather than “Lee-muh”…
Yeah, I get the whole tow-may-tow Vs. tow-mah-tow (I have even heard tow-mah-ta). I will even accept Lyme-uh if it a bean, otherwise it is Lee-muh (as in the capital of Peru)… But this I will not bend on.
Unless you are happy with me saying things like. “Have you seen that new game? It has awesome jaffics”
