It is really an impressive light from the reviews. I hope Amanda can help us to reflect our opinions about the dielectric layer to their engineers. We want direct-copper MCPCB.
LOL, when I'm doing bold and brainless statements, I'm not even checking my own results 
Pulled the LED - confirmed not direct thermal path - no continuity, but if you see the tracing on top of the star, it's pretty apparent there is a layer over the surface of the star that continues under the pad. K - will attempt to sand/grind it out...
Let.us know how it turns out Tom, got one inc also. Perhaps I should stop reading your reviews haha.
Ok - finished the mod to the MCPCB. Sanded then scraped out most of the dieelectric layer, fillin the void with solder paste, piled up, then reflowed the LED - looked good. My concerns:
- no extra solder flow out of the center pad, maybe didn't use enough?
- wear/tear on the LED from the reflow (used hot air from below, but took long to melt the solder)
- I did a couple of tests, so maybe didn't allow enough cooling off time
Before: on a fresh 26700 (KK) 5000 mAh cell, 6.42A: 1,462 - 1,377 lumens, throw: 120 kcd @5m
Now: on a fresh 26700 (KK) 5000 mAh cell, 4.22v, 6.10A: 1,612 - 1,455 lumens, throw: 130 kcd @5m
This is an improvement, but not as much as I thought. Also not sure why I'm only getting 6.1A now when before I got 6.42A. Maybe unreliable amps measurements - little tricky to work around the fat 26700, or something else is going on. Also, the 150 lumens drop in 30 secs is fairly high - probably attributed to the poor bonding of the star to the pill top (uneven pill top). I added more thermal grease and still didn't see any squeezed out after screwing it down.. Hhhmmm....
Thanks for all the updates and testing Tom!
Improving emitter cooling (like going from standard aluminium to Noctigon) increases Vf. Probably seem thing with the mod you did to the mcpcb. Increased Vf resulted in lower amps but at the same time higher output.
No wonder all my lights modded with Noctigon have lower amp reading.
New driver setup. Stock driver is stripped and used as a contact board. BLF17DD with 6 modes (hi-med-lo-moonlight-strobe-battery check/beacon):
Yes - that's the general pattern. Agree - could be right there with the better thermal path up'ing the Vf. Thought my XinTD X3 got higher lumens (domed LED) even at 5.X amps - gotta check #'s. I know the XinTD X3 had outstanding OTF. Of course best is maybe my TN31, measured 6.2A at the LED, and over 1,800 OTF lumens (domed stock LED) - but you got a rock solid big copper direct thermal path MCPCB, and nice buck driver with bridged resistors, and much bigger reflector, good qual AR lens, etc...
That is what the medium and low modes are for. 
True, should be in the 3.2 to 4A range, depending on the cell. With the stock 5000 cell they offer, it's pretty good. It's really in the range of a modded light. Two cell lights, like the M24 would be better if you want more runtime - classic tradeoff of runtime vs. size/weight.
” Two cell lights, like the M24 would be better if you want more runtime - classic tradeoff of runtime vs. size/weight.”
True - be interesting to know exact runtime on continuous high of the stock M12. Can someone do a test with their new 26700 batt?
For comparison, the modded M12 vs. my modded XinTD X3 (mod thread here):
Modded M12: fresh KK 26700 5000 ICR 6.10A, lumens: 1,612 @start - 1,455 @30 secs, throw: 130 kcd @5m
Modded XinTD X3: fresh KK 25550 4200 ICR 4.56A, lumens: 1,469 @start - 1,448 @30 secs, throw: 76 kcd @5m
The drop is higher on the M12, but there's more to lose with a DD at much higher amps, while the X3 is using a stacked Nanjg. The X3 has advantages with sheer mass, big chunky pill top, but of course it weighs a ton too, but the smaller width reflector results in the lower throw, but respectable for it's size.
Note: all measurements done on my home brew PVC lightbox and $35 meter, so no guarantees 
Ok, you’ve all convinced me: Where can I get the M-12 and perhaps also their Gladiator!
Aliexpress has them for $67.50. A desert rat can appreciate a light like the M12. I use mine out in the Colorado Desert just south of you all the time. You should also get the 26700 battery from Richard at Mountain Eletronics. That batt will give you 5,200+mAh of use with low internal resistance to boot. Quite a pair.
but Richard at mtmelectronics.com says he's going to be carrying/selling these, and his prices are always pretty good.
That Richard is a Genius!
Do they make extension tubes for this—to make it a 2 cell, like the M24?
And the driver probably couldn’t handle 8.4v anyways.
Maybe:
From the OP:
“”If you look at the MaxToch M24, the head is identical in looks and dimensions to the M12, therefore, I’m thinking they are using the same driver. if so, the driver could possibly take 2 cells (8.4v).“”