This is a copypaste of my post from reddit. Trying to find a solution to my problem.
Computer Type: Desktop
Motherboard: Asus PRIME X570 Pro
GPU: Sapphire Pulse 5700 XT
CPU: RYZEN 7 3700X 8 CORE 16 THREADSCPU Cooler:
Corsair Hydro Series H115i RGB Platinum
RAM: 2x 16 GB HyperX Predator 3200MHz
PSU: Corsair RM750X 750W 80+ GOLD FULLY MODULAR (2016 version)
Case: Kolink Phalanx RGB
Storage: SSD M.2: PNY CS3030 1TBSSD: Samsung 750 EVO 500GB
Operating System & Version: WINDOWS 10 PRO 19042
GPU Drivers: AMD GPU Driver 20.11.2
Chipset Drivers: AMD X570 CHIPSET DRIVERS VERSION 220.127.116.118
BIOS: PRIME X570-PRO BIOS 3202with Update AMD AM4 AGESA V2 PI 18.104.22.168.
Monitors: Primary: Acer XV430CK P - 34", 144Hz, 3440x1440p, connected via DPSecondary: Acer XF240H - 24", 144Hz, 1920x1080p, connected via DP
Background Applications: Steam/Battle.net, Discord
Description of Original Problem: System and/or GPU driver crashes randomly during gameplay. The game doesn't seem to matter too much, might be WoW, WoT, CP2077, etc. It might happen as soon as I launch a game or after for playing several hours. There are two consistent errors in event viewer, posted below. Thermals don't rise too high, with CPU being below 60°C and GPU/core temp staying below 70°C. GPU memory temp and hotspot temps stay below 90°C. MB and SSD/m.2 temps won't go over 40°C.
Errors in event viewer: Event 18, WHEA-LoggerEvent 41, Kernel-Power
Components purchased:PSU - December 2016CPU, Motherboard - September 2019Memory - December 2020GPU - February 2020SSD M.2 - November 2020SSD - October 2016Primary Monitor: August 2020Secondary Monitor: September 2016
Troubleshooting: Problems started some time after my previous GPU (R9 390) died early last year and I bought the new GPU. I have tried basically all GPU drivers since beginning of 2020, I have tried different bios versions, and updated windows and chipset drivers when available. I have tried different XMP profiles for the memory with current RAM and the previous ones (2x 8GB HyperX Fury 3200MHz). I have tried various bios settings with PBO enabled/disabled, PCI-E set to 3.0, 4.0 or automatic, etc. (I have disabled fast boot also). Windows is installed on SSD M.2 partion (100GB) and the drive has been on both slots available on the motherboard with no difference. I have tested memory, no errors. I have put the system through multiple stress tests without any crashes. I have updated all available DirectX and Visual-C Runtimes. I have tried different settings in GPU drivers (enabled/disabled freesync, etc.) I have run the system overclocked, undervolted, on factory settings, makes no difference. And the clock changes have been made by using calculation programs or automatically via drivers or bios. I have also set much higher fan speeds on GPU to keep the memory temps down.
That's all I remember for now, I will update when something comes to mind or anyone asks.Any insight to my problem will be appreciated.
Edit: Power settings in power options are set to AMD High Performance and I have currently GPU connected with two separate power cables. I have tried different PSU also and switched the memory and problem still persists.