I'm hoping Starfield didn't kill my 3080 Ti. Now I come back to Starfield and it also gives me that error message during weird times, like alt tabbing back in after alt tabbing out, or during loading, or just randomly. Afterwards I played Divinity Original Sin 2 with my friend for several hours (as we had been doing for about the entire past half week), and all of a sudden I got that crash message. Then I think at some point I did get a crash in it. I just started playing Starfield yesterday. The problem is that this error has never popped up on my 3080 Ti before in the past the entire year and a half or more that I have had it. I know it's a GPU driver crash, but until I can rule out this motherboard and RAM, since they're both open box, I'm putting a bit more scrutiny on them. I haven't had that weird wattage spike again and all of the CPU temps look fine, though. I tried doing DDU and doing a fresh install of current Nvidia drivers, no luck. I'm not sure if that's the actual name of the. I can show them in a bit, but I'm running a memtest86 to rule out memory errors. I'm having some "Nvlddmkm" (or similar) crashed and recovered errors in event viewer. I'm just posting in this topic because it's still open and it MIGHT be related. The problem is that I can't tell what could possibly be causing them. It will look like some ancient aliens scrolls, but there might be a human-readable description of a hardware error it noticed. Maybe launch some "Live" (non-installing) version of Linux and leave it running overnight, after which open the terminal and type "dmesg -T". I can't speak to the overvolting issues on some mobos, or the specific voltages your CPU should run at - I haven't seen one IRL yet. I also assume some of these numbers (voltage readings, motherboard electric stuff) are provided sometimes by some chip on the mobo other than the CPU, so there's that also. ("Windows Hardware Event Architecture") or just a short "mce" somewhere.īut - that might have made its way to the logs, but might have not, too. These can be benign, some ECC corrective action happened, etc. Is there anything I should test?Ĭlick to expand.Machine Check Exceptions and other hardware weirdness can occur without Windows needing to stop the machine. Stress tests only push it to ~92W, as expected. I didn't notice any burning smell, nor any bulges or anything odd. So I don't think it actually pulled 4.8kW, but question is what actually happened, then.? Has anyone seen this weirdness with AM5 and OpenHardwareMonitor themselves? I smelled around in my case after opening the side panel, and then looked at the back of the PCB behind the CPU (or what I could see near the backplate window). I did also notice that I had a max value of CPU package temperature around 86C I'm not sure if it happened at the same time, but 86C is still within the expected thermal envelope for this chip anyway. Obviously I have heard about the AM5 boards randomly burning CPUs, but as I understand it this was mainly an issue on ASUS boards. I haven't had weird crashes in games or stress testing. I keep my computer on 24/7 and it has been running stable since I built it about 4 days ago. Obviously it didn't actually happen or I think my entire case and desk would have melted and/or burned down. Some of the individual cores had wattages ~180 as well. What happened was I randomly looked over at it and looked at the max wattage of the CPU Package in OpenHardwareMonitor and noticed that it was around 4800 watts.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |