Hard reset into a bios "loop"

Posted by clothing_throwaway@reddit | buildapc | View on Reddit | 2 comments

 Mobo: ASRock B550 Phantom Gaming-ITX
 Case: Dan A4-H2O (mini ITX)
 GPU: RTX3080 (ASUS TUF)
 CPU: Ryzen 7 5800X3D 
 Cooling: 240mm AIO w/2x Noctua (I forget the size) 
 RAM: 32gb Corsair Vengeance LPX DDR4
 PSU: Corsair SF750
 OS: Windows 11 Pro

My Problem

I've been having an issue (lately, only when playing Final Fantasy 16) where my PC will freeze (audio continues to play for a little bit), screen eventually goes black (no blue screen, messages, error codes, etc.), and it eventually spits me out into the BIOS. I then get stuck in a sort of "loop" where if I choose to exit the BIOS (whether or not I make any changes), it just throws me back into the bios.

The only thing I can do from there (as far as I know) is to force the whole PC to shutdown by holding the power button, wait about 10 minutes or so, and then turn it on again, in which case the PC boots into Windows and works like normal.

This also happened to me when I played Indiana Jones and the Great Circle (so back in late 2024/early 2025). I struggled to figure out what was going on back then, but the problem did eventually go away. It's never happened to me again until now with Final Fantasy 16.

Overheating?

I was wondering if it was an overheating issue, so I've been monitoring GPU and CPU temps. In general, they seem fine. GPU hovers in the 65C-75C range, CPU in the 70C-80C range (highest I tend to see is maybe 81C or 82C and usually only for a little bit when transitioning between scenes or going in and out of menus).

One time, I caught this right as it froze. CPU at 90C is definitely high, but it only got that high for a fraction of a second and that was right as the game froze and the PC shut down. Also, the fact that my "fix" is to force a shutdown and let the PC sit for a while makes me wonder if it's really an overheating issue. The problem is that this freeze/hard reset problem also sometimes happens when temps seem normal (like 70C-75C or so max for both GPU and CPU).

PSU?

Some people seem to think it's a PSU issue because the whole PC shuts off unexpectedly without any errors, blue screen, or any other indication of what could have gone wrong. This makes sense to me, but I don't have any spare PSUs to test with.

So...before I go buying a new PSU just to test, is there anything else I should consider? Anything else I maybe haven't thought of?

[-]

Creative_Bedroom_448@reddit

The fact it dumps you into BIOS and then refuses to see the boot drive until a full power-off makes me suspicious of the SSD more than the PSU honestly. If the drive temporarily disappears/crashes under load, the motherboard suddenly has nothing to boot from and drops straight into BIOS.

clothing_throwaway@reddit (OP)

This also makes sense to me. I checked CrystalDiskInfo: https://i.imgur.com/hse2CIy.png

Seems mostly good as far as I can tell, but I'm not sure how to interpret all those IDs or their raw values. Regardless, temps are obviously being flagged at 65C and that's only on idle. Could my problem be because it's getting even hotter under load/while gaming?

My mobo has two m.2 slots: front and back

My 2TB is my "main" drive with OS and I have that on the front slot mostly just because it actually has a heatsink. I have an older 1TB (which is just for "overflow"), but I stuck that on the back slot. Maybe it would have been better to do the opposite? The front slot has a heatsink, but it's directly below the CPU and in this system the CPU definitely gets the hottest.