Purchased a fresh monitor, but GPU fails within moments of launching the game.
Purchased a fresh monitor, but GPU fails within moments of launching the game.
My initial message, sorry for the lack of polish. Here’s a revised version:
I made an attempt to repaste, but now my PC is stuck in a boot loop with purple glitches. See the images in the replies below. I recently got a new monitor to finally enjoy 1080p at 60Hz and fully utilize my GPU, but when launching certain games the temperature rises to 81-84°C and then it crashes at full fan speed, requiring a reboot to fix. So far, I’ve only managed to play a few games: Minecraft (with shaders) – no problems; Stellaris (midgame) – mostly fine except for occasional lag; Anno 1800 – fan ramping in the main menu and crashing when entering a game at quality settings; Fallout 4 (Atomic Warfare modpack) – GPU spikes to 80°C and crashes right away.
I’ve kept my usage around 30% GPU and about 40-50% CPU across all cores. I also tried reducing the monitor’s frame rate to 120Hz from 170Hz, but the problems remain. I changed Anno’s quality to medium and set the game resolution to 1969x1108. I almost considered upgrading my motherboard and CPU, but most of these issues stopped after Fallout 4.
My setup: Ryzen 5 3600 CPU, Asrock B450M Steel Legend, BIOS 8.01 (or 4.60), Corsair Vengeance LPX 3200 4x8GB RAM, Nvidia RTX3070 8GB GPU, ASUS TUF GAMING case, Fractal Meshify, Adata SU630 storage drive, Samsung 860Evo and WD Blue SN570 drives. Power supply: EVGA B5 550W. OS: Windows 10 Home 64-bit.
I’m open to any advice, especially since I’m new here. Thanks!
Edited October 20, 2023 by sheo
Added specs
Adjusting your monitor refresh rate to 60 still happens. Test with Furmark or another benchmark tool to confirm performance. Review Event Viewer after a crash, focusing on system and program logs for helpful details.
The GPU was bought new or came pre-owned? If it was used, it had been refurbished and cleaned before installation. Based on your case and components, you shouldn’t be experiencing such high temperatures. It seems there might be an issue with the GPU’s cooling system.
Not always altered unless you’re familiar with the individual; likely it was used in mining and might require a new copy. The safest approach is to either use a professional or handle it yourself if you understand what you're doing—re-paste the card carefully. It seems like a basic overheating problem.
Did you attempt to re-paste it? The pink lines are artifacting and would indicate the GPU is failing. If it's been overheating for while and it was used in mining it's most likely bad.
I tried to reattach it carefully, avoiding any contact except for the processor and using Isopropyl. It might be a problem from me or possibly the GPU gave up. Fortunately, the seller I bought it from had the original receipt, so I’ve started the return process with Asus.