Question GPU Driver (probably) crashing for random reasons. (No BSOD)

Oct 18, 2024
5
0
10
Hello guys, It's been a long time i've been having an issue with my computer. Mostly when gaming, my screen goes to one colour and either the heat goes insane or the gpu driver restarts alone. (Note that the heat goes insane AFTER the gpu crashes)This pc is 5 years old now so i guess the graphic card is getting old. Also i had huge heat issues for a while before i changed thermal paste. Now its not overheating anymore.I was told that the Legion Y540 had a manufacturer defect.

Its a Legion Y540-15IRH laptop
NVIDIA Geforce RTX 2060
Intel Core i5-9300HF CPU @ 2.40GHZ
Windows 10 Pro 64-bit

I can replicate the crash 100% of the time when playing some games or when i try to use FurMark.
 

Lutfij

Titan
Moderator
Welcome to the forums, newcomer!

First try and see if your laptop is pending any BIOS updates. Following that, see if your OS is pending an update. I see you're on Windows 10, which version(not edition) of the OS are you on?

Following your OS update, run DDU to remove all GPU drivers(Intel, AMD and Nvidia) in Safe Mode, then manually reinstall the latest GPU driver sourced from Lenovo's support site(bespoke to your laptop) in an elevated command, i.e, Right click installer>Run as Administrator.

Mostly when gaming, my screen goes to one colour and either the heat goes insane or the gpu driver restarts alone.
+
I can replicate the crash 100% of the time when playing some games or when i try to use FurMark.
What sort of temps do you see before the issue crops up? Have you tried replacing the thermal paste and pads for a higher quality thermal interface material? Have you performed any undervolt on your CPU and GPU?
 
Oct 18, 2024
5
0
10
Welcome to the forums, newcomer!

First try and see if your laptop is pending any BIOS updates. Following that, see if your OS is pending an update. I see you're on Windows 10, which version(not edition) of the OS are you on?

Following your OS update, run DDU to remove all GPU drivers(Intel, AMD and Nvidia) in Safe Mode, then manually reinstall the latest GPU driver sourced from Lenovo's support site(bespoke to your laptop) in an elevated command, i.e, Right click installer>Run as Administrator.

Mostly when gaming, my screen goes to one colour and either the heat goes insane or the gpu driver restarts alone.
+
I can replicate the crash 100% of the time when playing some games or when i try to use FurMark.
What sort of temps do you see before the issue crops up? Have you tried replacing the thermal paste and pads for a higher quality thermal interface material? Have you performed any undervolt on your CPU and GPU?
Not sure if that's you want to see but this is i think the bios version :

BIOS Version, Date : LENOVO BHCN45WW, 24/05/2022
SMBIOS Version : 3.0
OS Version 10.0.19045 Build 19045


I'm gonna try to do a fresh GPU install.

I see around 71°C on temp and around 80°C on hotspot right before crashing, i dont think that the temp is making the gpu crash.

Also I tried to replicate it once more to see what's really happening when crashing : The GPU crashes then restart after few seconds, and the computer does 2 small bips.

Thanks for the quick answer
 
Oct 18, 2024
5
0
10
Yeah, you're on the last BIOS version released for that laptop. You didn't get back to my other questions.
I did replace the thermal paste for a better one, not sure what's undervolt but the laptop battery got swollen when i had high temperature issues so i removed it and i'm now only using the computer on sector, i never performed any overlocking or "undervolting"
 
Oct 18, 2024
5
0
10
Ok I tried to see what's happening in the event logs : i have these logs when crashing :
From NVIDIA OpenGL Driver :
Unable to recover from a kernel exception. The application must close.


Error code: 3 (subcode 2)
(pid=11328 tid=1232 furmark_gui.exe 64bit)

Visit http://www.nvidia.com/page/support.html for more information.



From nvlddmkm :
\Device\000000bc
Restarting TDR occurred on GPUID:100