There seems to be a surprising focus on the failure to make a successful auto following double engine failure. Perhaps this is due to the report which states that the reason for unsuccessful autorotation is unknown.
If I was putting someone through their paces in a simulator, would we fail one engine, and then while they were sorting out the first failure, give them a second failure at 600ft, low speed, at night, no landing light, over a city?? Of course not. You'd be setting them up to fail. It might be a demonstration of why not to get yourself into that position in the first place, but I can't imagine that many would turn that into a successful auto.
You're low level
The first engine fails
You delay for a couple of seconds with shock, then carry out immediate actions
You're scanning the instruments and switches trying to work out why you have this confusion
You secure the engine
Suddenly the second engine goes
Multiple lights illuminate
The outside lights go dark
The NR has already dropped below safe
You have no option but to bottom the lever
Dropping through 300ft trying to recover Nr
Knowing you are descending way too fast and you have no accurate height figure
You know you have to raise the lever or certain crash
The difference between a Flaring too high, successful auto or crashing to the ground, probably about 2-3 seconds either way. Good luck perfecting that when you're also trying to look for somewhere safe to put the aircraft down and also trying to understand what has gone wrong.
The fact is the aircraft was upright and level when it hit the roof, just a few seconds after the second engine failure. Looking at the wreckage, it almost looked survivable and it possibly nearly was. A few seconds later before instigating the final 'pull', and it might have just had enough energy to arrest the descent into a heavy but survivable landing. Little point in debating this final element which the report acknowledges was in exceptionally demanding conditions.
As to the FDR/CVR. Its absence made no difference to this accident, just our understanding subsequently. As warnings are recorded by the onboard systems, this has yielded useful information. Just a shame that there isn't a time stamp on those captions as that would be an easy fix and would have provided more useful info.
Ergonomics, switch positions and 'what if' scenarios are being debated in the report and on this forum and within them is quite likely the answer to what went wrong. It probably doesn't matter which one of them is correct, as EC135 pilots are identifying possible scenarios that all need to be eliminated to help prevent similar accidents happening again.