Hey everyone, I recently picked up a used Zotac 3060 Ti from Facebook Marketplace for £180 to upgrade from my old GTX 970. I didn't have a chance to test it before buying, and while the seller seemed legit and the card looks fine physically, I'm starting to suspect I might have bought a dud. I've been dealing with crashes linked to nvlddmkm.sys event ID 153, mainly when playing games like Control and Borderlands 3 after about 30 minutes. I tested it in my partner's PC too, and the same issues occurred, which suggests it's not a problem with my own system. I've tried various drivers and made adjustments in BIOS and Windows, but nothing seems to work. The only thing I haven't done yet is to repaste the GPU, but I'm skeptical if that would help since temperatures seem okay. I'm feeling pretty stuck here—anyone have advice on how to fix this or should I just accept that I have a broken GPU?
5 Answers
Have you checked the VRAM? Sometimes mining cards can mess that up, and replacing the VRAM could be a repair option. It's worth testing it out because that could resolve your issues.
Before you go repasting, I would suggest uninstalling Destiny 2 and then using DDU to cleanly remove the graphics driver. After that, reinstall the graphics driver and the game. The crashes when loading into the Astral Plane might be tied to an issue with shaders from your old GTX card.
One thing you might want to try is undervolting the GPU, as some folks who've had similar issues found it helped. Also, make sure your PSU can handle the 3060 Ti; it's definitely more power-hungry than the GTX 970. What PSU are you using in both setups?
Yeah, that sounds like a solid setup, but keep in mind that the 3060 Ti does draw more power than the 970.
Repasting could be a shot in the dark, but it's free if you have some paste lying around. The card is getting older, so it might be worth a try regardless. If that doesn’t work, checking if the card's BIOS needs an update might help too.
Honestly, sometimes all it needs is a good repasting. If the thermal paste is dried out or not fully covering the chip, it can cause issues without showing obvious signs. I had a similar problem with an older card that crashed frequently until I replaced the paste. Just a thought!
I already tried undervolting but it didn't help. For the PSU, I've got a NZXT SFX-L 650W in one PC and an older Corsair SF600 Platinum in the other.