I think I'll want to try this, but I'm not sure what battery you're using, unless you're still testing which one is best for that led.
I started with an AW IMR18650 1600Mah because I already had one, 4.6A peak, then quickly declined.
Next I tried 3 D cell 10,000Mah NimH that were ok but didn't give
the amperage I thought they would, 4A. Didn't do a long term test.
Then I bought an IMR26650 4000Mah from Batteryspace.com. 4.25A peak and could maintain close to that for 6 min. test. This or a KD D cell Lithium-Ion are probably the best for a one cell setup right now.
AW has a prototype F cell (a long D) 7000Mah Lithium-Ion in an SST-90 light. Don't know when that may come out.
Right now I'm trying a 3D body with 4 C cell NimH, 2500Mah Energizers (not very good, but all I had) for 6A. Higher capacity cells should work better. No indications of overvoltage. The only problem is the D2Flex 10A multilevel driver has a whining noise on the middle levels with this setup that wasn't there before. I don't have any more money for this project and won't buying any other batteries, plus I like smaller lights, so I'm going to cut my 2D down to 1D size and use the IMR16650 for now. When higher bin emitters are available, I'll upgrade this one.
My forecast is the best output will be with a high amperage/normal voltage buck driver (nothing with the needed specs exists now that I know of) and multiple IMR cells, or to continue raising the voltage using direct drive until we find the actual limits.
When high bin SST-50's are available I think this might be the best choice. The smaller single die should give some nice throw.