[SOLVED] Restart crash during gaming/high usage

Feb 14, 2021
11
0
10
So recently my computer started crashing during games and heavy loads.
(system crashes during gaming just shutsdown and restarts)
(also gpu has high memory temps (102c-104c))

I presumed it might be the psu (because its less than recommended) Tho that seems to not be the issue (after checking power draw of the computer never exeeds 500 watts. (checked with external watt meter)
than i thought maybe driver issues and unninstalled/reinstalled it, also reinstalled all programs with issues. no luck still crashing.
next thought was maybe a ram issue. so i took out a stick, interesting now its not crashing anymore, however my gpu is only using 60% of its performance now.
so ? is it the ram the gpu or the psu.
I already purchased a 1000w power supply that i am waiting for but with corona its taking a while.
do i get a new set of ram?

UPDATE:
I think i found the issue, i had xmp II set in the bios i think it was making the system unstable, have both ram sticks in. pc running warzone for 2 hours with a cpu and memory stress test running. no crash yet so i think its solved. leaving the thread open until i am 100% sure.

UPDATE:
Still crashing.
Its really inconsistent, only been able to get it to crash on warzone. Sometimes 2 hours no problem sometimes crashing very match.



SYSTEM:
cpu: intel i9 9900k
gpu nvidia rtx 3080 fe
power supply: bequiet 650watts
ram: hyperx fury ddr4 16gb x2
mb: assus rog strix (not sure what the board is called exaclty)
 
Last edited:
Solution
nope no mining (i tested with it a couple days but the temps where too high and i need this gpu to game not to mine so i left that, i was thinking a couple bucks a month is nice, but not worth destroying my gpu over)
now im considering puting in new thermal pads, its not that hard but still the card is pricy haha
Putting in new thermal pads is probably your best option for now. Let us know how that works out.
Hey there,

What exact model BeQuiet is it? As you've said, the recommended is a 750w power supply. Even though your system may not draw more than 500-500w with a 3080, it's the fast switching frequencies for the RTX3xxx series that cause issues, for what would ordinarily be capable PSU's. If you have PSU that is more than a couple of years, it's voltage regulators may not be up to the task of coping with fast load changes.

Hopefully you've gotten a quality PSU as a replacement, and it will make the difference. Which one did you buy?

In terms of your edit. Perhaps that will help. Hopefully running the ram at different XMP is the solution. Ram crashes are typically BSODs or freeze. Restarts typically point to PSU.
 
Feb 14, 2021
11
0
10
Those GPU memory temps are very high. I'm not sure if those temps will cause your system to crash or maybe to throttle down. Are you doing any crypto currency mining?
nope no mining (i tested with it a couple days but the temps where too high and i need this gpu to game not to mine so i left that, i was thinking a couple bucks a month is nice, but not worth destroying my gpu over)
now im considering puting in new thermal pads, its not that hard but still the card is pricy haha
 
Feb 14, 2021
11
0
10
Hey there,

What exact model BeQuiet is it? As you've said, the recommended is a 750w power supply. Even though your system may not draw more than 500-500w with a 3080, it's the fast switching frequencies for the RTX3xxx series that cause issues, for what would ordinarily be capable PSU's. If you have PSU that is more than a couple of years, it's voltage regulators may not be up to the task of coping with fast load changes.

Hopefully you've gotten a quality PSU as a replacement, and it will make the difference. Which one did you buy?

In terms of your edit. Perhaps that will help. Hopefully running the ram at different XMP is the solution. Ram crashes are typically BSODs or freeze. Restarts typically point to PSU.
right now i have a
bequiet straight power 11 650w
i ordered the same but with 1000w
they are kind of expensive but i trust the German engineering,

so about the edit: i have been running cpu gpu and ram all at 100% for 2 hours now and am overclocking the gpu a bit too see what it does.
but its all stable for now.
 
Feb 14, 2021
11
0
10
  • The GPU temps are far too high, get the fan curves tuned up.
  • i9 has to be undervolted for stability.
  • Run series of stress tests (with OCCT for example) on individual components to see what is causing the crashing.
you sure the gpu temps are too high? i googled it and they say its a normal temp for the memory of the 3080.
the temp of the gpu never goes over 80c tho
 
nope no mining (i tested with it a couple days but the temps where too high and i need this gpu to game not to mine so i left that, i was thinking a couple bucks a month is nice, but not worth destroying my gpu over)
now im considering puting in new thermal pads, its not that hard but still the card is pricy haha
Putting in new thermal pads is probably your best option for now. Let us know how that works out.
 
Solution