Reaction Time
It appears to me that everyone is in agreement that if the engines stop, the collective lever is lowered and appropriate movements of the cyclic control need to be made immediately. The exact order that this is done and the extent of the control movements necessary depend on the type of helicopter and the flight regime at the time. The key to a successful autorotation seems to be that the control inputs are made within only a couple of seconds depending how close the helicopter is being operated to the avoidance curves. Given a couple of hours flying at night, base in sight, perhaps a bit of chat going on with the observers, you're chilled and relaxed and suddenly both engines stop. How many helicopter pilots could honestly say that they would have mantained control and performed a successful autorotation to the ground? Given the apparent reliability of twin engined helicopters generally, it surely would take at least a couple of seconds to suspend disbelief at what had just happened before the brain engaged and corrective action was initiated, by which time it is too late. That doesn't address the question of what caused the engines to stop suddenly, but given the circumstances known so far, I'm fairly convinced that very few pilots would have managed a survivable landing. For the purposes of this statement, just assume that getting the helicopter anywhere on or in the vicinity of the ground with the cabin intact would do. The links to the helicopter flying manual and other documents have been very informative, thanks to everyone who has posted those.