BSOD : VIDEO DXGKRNL FATAL ERROR (linked to NVIDIA)ACER PREDATOR PH317-53

ThomasJohnson
ThomasJohnson Member Posts: 2 New User
edited June 2023 in Predator Laptops

Hello,

For quite some time, my ACER PREDATOR PH317-53 has been experiencing BSOD with this code, when I'm in a video game. I have a RTX 2060.

Tuneup: https://1drv.ms/u/s!AsVxnDhQIr3Sg6tXyISKi6bGZl2-gA?e=BAqa4oMemory dmp: https://1drv.ms/u/s!AsVxnDhQIr3Sg6tWQe6AcaU0pMKJbg?e=ZU4UdrThe game crashes randomly after 10 to 100 minutes of play, and then I get this BSOD. This has happened with various games using the GPU intensively (my PC gets very hot most of the time, without exceeding 90°). It seems the NVIDIA card just turns off when the game crashes, because when I have other programs running with the GPU, the NVIDIA widget shows me that no program is active just after the crash.

I've already tried everything, including completely reinstalling Windows and deleting all data. No change at all. The Windows help utilities (hardware etc..) find no problem. Changing power settings also. I performed a DDU in safe mode with connection disabled and reinstalled GPU driver only, no change. Sometimes, the same games run throughout my gaming session, no parameters having been changed.

I'll go ahead and share you some files you could use to detect the issue :

Tuneup : https://1drv.ms/u/s%21AsVxnDhQIr3Sg6tXyISKi6bGZl2-gA?e=BAqa4o

Memory DMP : https://1drv.ms/u/s%21AsVxnDhQIr3Sg6tWQe6AcaU0pMKJbg?e=ZU4Udr

3DMark Benchmark : https://www.3dmark.com/3dm/95938896?

FIRESTRIKE folder, where you can find HWinfo details and logs during the Firestrike benchmark : https://drive.google.com/drive/folders/1uBrHsAhHJlpWQurhuuhk5MU6RfFDEQa5?usp=sharing

HWinfo logs, whose monitoring ended right after the game crashed and just before BSOD, you may find something happening near the end of the monitoring period : https://drive.google.com/file/d/1hOz_iV0iPJp3G5871BR80VYPBT8SpX2x/view?usp=sharing

I would love to have the help of this community!

Thanks in advance

[Edited the thread to add model name to the title]

Answers

  • JackE
    JackE ACE Posts: 45,085 Trailblazer

    without exceeding 90°

    Discrete GPU temps should not exceed ~86°C or it will shut off. Only the integrated CPU/GPU is designed to operate up to about 92°C. So the games are crashing during the process of the nvidia card shutting down. Repeatedly exceeding 86°C will eventually damage the nvidia chip. Heatsinks should be re-pasted to the processors about once a year for machines that are heavily gamed. This should be considered normal maintenance.

    Jack E/NJ