I just ran the runtime test... but using the same sphere + luxmeter for another runtime test, so can't look at the data yet :D
The battery is almost finished charging, so I can measure lumen in my other sphere.
SMine didn't turn off after the runtime test, but was still at sub lumen output. The Voltage, however, was still 3.4V when I pulled it.
Ok, I got the following: highest measurement (about 2-3 seconds after turn on) was 162 lumens.
After 30 seconds: 156 lumens
Will do some throw measurements soon
Edit: just measured at 20 meters: 68,400 cd which is close to Djozz, but my lumen numbers are higher 
Oh wow, that’s worse than expected.
I think Lumintop has introduced the 14500 LEP category, but done it with a bit of a loser. Other than looks. They’re setting themselves up to get really leapfrogged by someone with actually 400 lumens and a 16mm lens.
It’s a shame disassembly of these LEPs is prohibitive, I’d otherwise love to see it.
At least you found a higher output than mine, let’s hope for a higher throw number too 
Edit: hmm, so not better throw.
updated the post.. almost same throw numbers as you.
Measured amps: 1.09A
Maybe Lumintop’s luxmeter likes blue light 
That's possible!
Thanks for doing this. I was debating on getting one, but I’m glad I got a Catapult Mini instead.
Mine came in also today, same serial # too!
I don’t really consider these LEP devices to be a flashlight.
They are rather a Laser pinpoint beam that has been Filtered and Blurred/Scattered into pretending to be a flashlight beam, That is why the lumens are so low and the tint is so high, it is a big compromise.
Where as a True Flashlight has a beam that has been focused into a fairly wide but sharp spot true beam if that makes sense and has real high lumen capability as well as various tints available.
If you like the LEP’s available now, that is great but they have not yet risen to the level of excitement to want to buy one, for me personally.
As always YMMV.
2 RCR2 cells might perform better if the driver can handle 8.4v.
That would be awesome but I doubt it can handle that that much.
I think that is true
the HDS luxmeter does that too, it reads a Novatac w P4 at 120 lumens
my meter reads it as 60 lumens
Hogo assures me he calibrated the Novatac himself, using a “special” conversion factor…
Lumintop probably adjusted their luxmeter (with a “special” conversion factor) for the spectrum of the LEP
Ouch!
Does your light still work properly with a different cell ChibiM?
Oofff poor Chibi!
You are already in the conservation, but for everybody else: Over at reddit Cheule got higher output numbers from his Ant Man. He is saying removing the clip and really tightening the light was helping a lot. @djozz maybe that’s something you could try with yours too?
Also let’s hope Chibi’s one is just a dud and that not every Ant Man is a battery killer!
I checked mine with an NiHh battery at 1.35V, and found no drain, not even a micro-amp.
In the meantime, with a H10 battery and a tail-spring bypass (of which I’m not sure if it helps at all), I measured 77 kcd at 30 seconds. Still short of the specs though.
Who here is going to be the first one to risk burning up their Ant Man by replacing the stock driver with a FET driver on Vapcell H10?
I am thinking of it. I am really disappointed with it's performance :(
Do it!