A cautionary tale about assuming that the correct response to a failure is obvious and straightforward:
Once upon a time, I was involved with the addition of a FADEC (Full Authority Digital Engine Control) to the turbine engine on a single-engine helicopter. I forget why but it was single channel. In the unlikely event that it failed, the pilot was obligated to control rotor rpm with the throttle twist grip on the collective just as you did when flying a helicopter powered by a reciprocating engine back in the day.
Before deliveries began, I took it upon myself to fly and land one in manual mode and even though I hadn’t twisted a throttle in some years, there was no drama involved.
Unfortunately, as subsequent events proved, many pilots had never flown a helicopter with a reciprocatiing engine, having learned to fly ones powered by turbine engines with mechanical fuel controls that kept the rotor rpm constant, and training to fly this helicopter in manual mode not as prevalent or extensive as it should have been.
What was worse, the FADEC manufacturer had a circuit-board quality-control problem, which meant that pilots were being presented with the need to manually control rotor rpm after the FADEC had packed up fairly often and at least a few were not up to it, fatally in at least a couple of cases. The quality-control issue was subsequently resolved of course and after perhaps too many early incidents, more emphasis placed on training and occasional practice in manual mode. Moreover, my understanding is that the FADEC now has dual channels.