ASRock N100M, no picture no boot, but semi-alive

[Exim]

Hi guys,
I managed to semi-kill an ASRock N100M board after updating Bios from 1.10 to 2.01. It's a zombie now, dead, but not dead.

ASRock N100M special features:
- CPU soldered to board, small Intel 12th gen
- iGPU
- four POST LEDs

Problem after updating Bios to 2.01 using the in-Bios update tool: no picture, no boot.
I am not sure anymore if it booted at least once successfully after updating the bios to 2.01, maybe not, I can't remember. I think it did not.

Symptoms:
- no picture/signal on any of the display connectors, no boot, no beep
- Powerup POST-LED sequence starts correctly (CPU and DRAM LEDs fully lit, after about 1 second CPU LED turns off) but then deviates from normal behaviour: CPU LED starts blinking/flashing once every ~2.5 secs and the initialization sequence seems to get stuck in some sort of loop. The flashes are very short (amount of time the CPU LED is lit) and dim, they are so short a 60fps video camera has problems capturing every single flash, so probably less than ~17ms. No progress in POST, the CPU LED flashes until I power off the board or cut the power.
(I took a short video showing the powerup LED behaviour, it's only 1,9MB in size but uploading .mp4 files is impossible. So there's a single video frame attached as picture showing a lit CPU LED, just imagine it flashing once for a very short amount of time every 2.5 secs )
- power button / power pins on the board work: when pressing the power button it powers up the ATX power supply, holding power button for 4 secs turns it off
- if powered up without RAM it beeps continuously and blinks the DRAM LED

What I've tried:
- battery out, clear CMOS
- externally flashed the Bios (CH341A + Linux flashrom tool) to 1.10, to 2.01 (ASRock currently only offers these two) and again to the chip content I dumped directly after the failure, the board behaves the same in all three cases

What I've tested and verified:
- power supply is ok
- RAM is ok
- DRAM LED behaviour (stays lit) is ok, it would be switched off in a later POST step, which the board does not currently reach
- CH341A reads/writes reliably
- poking around on the board with a multimeter without knowing what is where showed no obvious shorts, the voltages I found seem plausible

To me "beeps if no RAM" and a working power button sound as if the CPU works and at least runs some of the bios code. But I'm old school and from a time when mainboards didn't have SMCs and stuff.
Now that I think about it I would like to check what the board does when trying to power it up with a wiped bios chip.

I don't know if this is an Intel ME problem or if the board has an actual hardware defect. At the same time I am not aware of any external impacts, ESD or whatever, that may have caused a hardware defect.

Do you have any suggestions? *scratching_head*

[Exim]

Test, trying to turn it on with a wiped (all FF) bios chip: As expected, power button / power pins on the board do not work. When switching on the power supply manually (connecting PS_ON with COMMON on the ATX connector), two board LEDs light up, otherwise no POST activity.

So the board does read Bios code and executes it, at least parts of it. Question: You still need a working x86 CPU for this, because still today this initialization is performed by the x86 CPU itself and not by an integrated or external microcontroller, right? It still today needs the CPU to do the "no memory detected" beeping, right? If all is "yes" it would mean that the CPU is working, right?


Figured out how to attach videos, here is the short video mentioned above: POST LEDs during startup process, endless short flashes of the CPU LED. At the end I switched off manually.