I'll go OT here too. Got my M31LL a week or two ago, and tried it out with Costco's Alkaline's using Valient's 2AA host, via Malkoff store. Got 7 hours 45 minutes of steady (bounce with lightmeter) output. Compared lightmeter numbers to my M61's, and I would guess 70-75 lumens. I am impressed, and will continue using the Alkaline's in that light setup, checking of course, for any leaking issues.
Bill
I just finished going over some rough runtime measurements I had done on several of my Malkoff drop-ins.
The M31LL gave 5 hours of runtime on 2xAA Eneloops (2000 mAh each).
Now, back to the main topic.
I had also been thinking, based on the specs for the M61SHO, that you might actually get similar runtimes when running it with 2xC123A (1500 mAh) and 2xRCR123A's (750 mAh). It turns out from Robert M's charts, this is actually true, when comparing regulated output. This is quite remarkable, actually.
One way to think about it, is that you can get 385 lumens (Malkoff estimate, Robert M gives a bit over 400) for more than half an hour using 2 small rechargeable LiIons. For comparison, we used to get only 20 minutes or so of runtime using old incan Surefires of roughly similar lumen output, which consumed several primaries in the process (maybe 6 primaries? if memory is correct using the M6 at 500 lumens as the benchmark).
Of course, you should roughly double the runtime using 2x18500's, and again double that if using 2x18650's.
Personally I think the best use of it might just be with the 2xRCR123A's, in a tactical use. For non-tactical uses, the M61SHO may not be the best drop-in.
One final thought, the 2xLiFePO4 format is definitely outmatched in this application by the 2xRCR123A's. It has no advantage and clearly lower runtimes.