As an engineer, it's in my nature to try and understand the math behind everything. It's a bit of a curse. I come home from work and play with Solidworks (3d modeling) and Octave (computational analysis) for fun.
It took me a few attempts at reading that to figure out what you were saying, but I think the gist is that you wouldn't bother accounting for the affect of ambient on the subject and would be ok with overexposing the subject by 50%. Either that, or decreasing the aperture 50% and letting the background be slightly underexposed. Either way, I think we are reaching similar conclusions.
The only thing I question with your response is needing a second GN1100 flash. My reading suggests that the total guide number of a series of flashes is the sum of the squares of the guide numbers. That is:
Guide_Number_total^2 = Guide_Number_1^2 + Guide_Number_2^2
That would mean I'd need 4 GN1100 flashes. Talk about some serious $$$$.
You got it! 'Overexposure' by 50% is well within the normal range of variation when you meter with your reflective light meter, rather than using an incident meter to measure the light falling upon the scene!
Ignore Guide Number math and simply think of this principle: If your ONE flash is -1EV, you need to DOUBLE the light to make the supplemental light 0EV...after all, every whole 'f/stop' is double with light. So instead of one GNx flash you need TWO GNx flashes, to up exposure by 1EV; you need FOUR GNx flashes to up it by 2EV, EIGHT GNx flashes to up it by 3EV!
So if GNx = 500 w-s, two is 1000 w-s, four is 2000 w-s
So if GNx = 1000 w-s, two is 2000 w-s, four is 4000 w-s
So if GNx = 1600 w-s, two is 3200 w-s, four is 6400 w-s
when the lights overlap each other...if they come from different directions with NO overlap, the additive nature of light is not involved.
