One time ET6 reported the bus multiplier had changed from 11 to 4. The Prime95 stats showed that 3 of the 4 cores had slowed down, only core 4 was running at full speed. So I updated the BIOS because I know that, in theory, bugs get fixed and that might have been a BIOS bug.
So, anyone got anything constructive to add? With BIOS F12a and ET6 I see glitches in the CPU1 fan speed measurement, that I didn't see with F11. With F12a the multiplier has been at 11 and hasn't changed.