I measured the parasitic drain of my V4A when the light off and not locked out at 0.77 mA. So it will drain my 2000mAh Eneloops in a bit over 3 months. Not too bad if you realise that NiMh cells do not become dangerous when fully drained, just annoying to have to remember either to lock-out the light or to charge the batteries in time before use.
(for who wonders how I measured it without frying the DMM: separated head from battery part, connected the head and tail shell (batt-) with a test lead, connected the batt-plus spring to the plus-pad on driver with two connected test leads, clamped the DMM on micro-amp setting from batt-plus spring to driver-plus pad too, parallel to the two connected test leads. So now head and battery part are fully connected (and thus this weird light is switched on) while the DMM is bypassed. Click the switch untill the light is off, disconnect the two test leads so that the current is diverted through the DMM)
Edit: measured the current of the lowest (20 lumen) and highest (850 lumen) setting too, resp. 60mA and 3.6A. These measurements were done with thick short test leads connecting head and battery compartment, and a clamp meter.
So the runtimes for the highest output will be a bit over half and hour (but longer because the current and output will certainly not be flat with this simple double-FET driver), for the lowest output you will have 33 hours.