My Specs are below AMD Ryzen 9 3900X, RTX 3080 10GB, 32GB DDR4, 1TB Gen4 SSD, 360mm AIO, 850W Power supply. When i play more intense games my PC will just turn off and wont turn back on for a few minutes unless i turn the power supply off them back on. I downloaded MSI Afterburning to log the history of Temps and Usages. Im wondering if anything stands out that can point to my problem. Im not too tech savvy so dont understand the charts and what to look for. Anything helps, thanks guys.
Just to be sure, check if the Windows system log reports anything, though I kind of doubt it. If you stress test the CPU, RAM, and GPU separately, it might reveal individual weaknesses, helping you to pinpoint the exact area of the problem. If CPU stress test crashes the PC, then it's not likely the graphics card, although a broken PSU could cause any sort of weird problems. If I had to guess, it's either a faulty PSU or related to thermal expansion in any of the crucial components, like mobo or graphics card. I suppose you don't have another PSU you could test? At least you should check all the connections (cables) within the PC, remove the graphics card and reseat it, maybe consider different rails from the PSU to the graphics card. Also removing the CPU and reseating could be worth it, even if a bother, as well as the RAM modules (easier). Naturally remove any dust everywhere.
There was a Critical alert that states : "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly." in the windows Event Viewer. Also note i reformatted and reset BIOS to default settings. While i was playing i watched the temps on the GPU and CPU and both looked normal the entire time. Ill try to use a different PSU.
Yeah, that alert is generated after the fact, so it's of not much help. If you have a cracked CPU socket, or another similar problem, it could only manifest after the component gets hot enough. That would be the thermal expansion flaw. Naturally that would also be the worst case scenario, especially if your warranty has expired. This doesn't sound like a problem caused by firmware or software in my opinion. Though it's just my opinion. The first thing you should do is to run the various stress tests. You should also try a memory test to check the memory integrity, unless you already have.
Gigabyte had some problems with power connectors... https://www.reddit.com/r/gigabytegaming/comments/jkur4r/3080_3090_gigabyte_eagle_gaming_oc_power/
Make sure you have individual 6+2 pin cables going into the card. If possible, try it out with a different PSU. That being said, it might very well be the GPU, I had 2x3080Tis die on me, and both acted in a similar way except the fans would spin to +100% after monitor lost signal and eventually wouldn't be recognized anymore, even after cold booting. When that happens, RMA will be the only way.