DN70 looks really nice
That’s 60W per cell, divided by about 4V equals 15A. Can be done theoretically, but most likely is not going to happen here. The included cells are 30Q according to Finnish site. I’d be happy if it pumped out 10000 lumens.
Yes, my preferred DN70 !!
The DT70 has 4 XHP-70 LEDs, not 3.
Yeah indeed, the picture was misleading for me due to the angle.
Dangit, there goes my calculation too. But, the efficacy of 4 emitters is higher than 3 emitters, so it’ less than 15A per cell you need…
Anyhow, I don’t think Imalent is even going to deliver 16000 led lumens…
Edit:
FYI, I asked Imalent: DT70 and DT35 have 4P cell configuration.
The difference is not monumental, I mean 243W input or 220W same thing.
But anyway there is not going to be a 4.0V available when the load is anywhere from 10A to15A, more like 3.85V for a minute.
Of course. Just ball park figures; easier to calculate. Just like 3 or 4 emitters, meh, it’s about the same. So the conclusion remains: verrrrry unlikely.
Maybe they really did it like 220W input for 30seconds, then drop to 4000lm (25%) and due to low voltage and rapid high discharge no more 16000lm at all after that (literally impossible).
Since with the new ANSI runtimes that would be totally normal.
It’s certainly possible, it has been achieved by 3x XHP-70 in a smaller host.
Anyway, 9k lm in the palm of your hand is already crazy, I’d love to see what 16k thrown at 120kcd looks like. 
Surely is possible, just for 30sec total possible runtime from 4*18650, after all a pull of 220W from max 40W available (lets not even mention the 2.5V cells)
Besides I am assuming all max efficiency and all, reality may be different.
Also notice on their website, it says:
ANS/NEMA FL1
… instead of ANSI…
4x LEDs and 4x 18650s, it pulls from each cell enough to make 4,000lm.
In theory if the emitter efficiency is 120lm/watt each cell provides 33W, maybe close to 38W after 15% driver losses. This gives ~10A at 3.7V.
At 100lm/watt it should draw ~12A per cell which is entirely possible with high drains.
Of course the above calculation assuming the 4,000lm per XHP-70 is LED lumens and not OTF.
The real problem here is how does the driver deal with 40A+ of input current if the cells are parallel.
Well, the Meteor is also 4P and pulls about 32A…
I think this configuration (4x XHP-70) kicks out ca. 12000 lumen OTF for about one minute and drops down dramatically!! …
It must be tested!! 
At least we know the flashlight can do 16000 ANSI lumen for 1hour and 30minutes, as the specifications say. Only when heated (50Celsius) the flashlight will drop to 1800lm output.
The second mode at 8000lm will work nicely for 2hours as it is rated.
This product sure is impressive, must output around 20Klm bulb lumens to be able to do 16Klm ANSI (which must measured at
least after 30seconds out the front). And it will not get hot in 30seconds to able able to get the 16Klm ANSI rating.
The 4 stages of a new Imalent light
1) introduction new Imalent lights
2) getting excited about it
3) finally a review with runtimes and teardown
4) buy another Meteor
Nice! Let’s get some coffee over there then.
Does anyone know if these lights will be available with neutral white? DN70 looks nice and has some good features i think!
I have not heard at all about neutral white versions…