Hi fellows!
Just stumbled upon this: A123 ANR26650M1-B 2400mAh 3.3V 26650 Bench Test Results...an extraordinary battery, with issues @ E-Cigarette Forum
All of those LiFePO4 discharge curves I've seen seem to start below 3,6V full no-load SoC. What gives? 
Cheers ^:)
They are the safest Lithium cylindrical batteries and have significantly better performance than any Li-Ion battery.
The bad thing is that they have a nominal voltage 3.3V at full load, arriving at 3.6V.
But with discharge at 30A have somewhere to 0.8 V voltage drop, while Li-Ion reach even 0.7 V with a discharge of 20A.
But the main thing is that this battery will not descend to a lower level of performance.
Will drop voltage, but will keep these things Volts by the end of its duration.
Unlike the Li-Ion, will then loses in performance as discharged.
Here we see the battery discharge to a measurement of the NKON.
A123 cells’ qualities (low IR, flat discharge curve) make it good for some applications, but the low voltage and low energy density limit its applications.
150 Watts per cell… nice…
Been interchanging some more data with Mr Mooch.
Let's briefly say that, for LiFePO4s, the initial voltage drop is steeper, as can be observed on the pulse discharge graphs. Example:
However, it is clearly seen how the slope flattens then.
These batteries can indeed do more than 50A continuous without breaking the 80° mark. 
Cheers ^:)
Sounds like pocket rocket material. Triple XHP70, single cell, 10000+ lumens easily. Glorious 10 seconds… Yeah, heat is still limiting factor…
Speaking about direct drive, hIKARInoob? 
Mmm, cross-referencing this and that, this is how I see such a setup:
And some heavy duty sunglasses while testing, just in case. 
Cheers ^:)
I have been thinking about it briefly. I do wonder whether the extra power is actually useful. Ok, we want HUGE amount of light (always more). The major problem seems to be the heat generated from the leds that are decisive in the design. Look at the Olight X7 or the Noctigon M43, or even small lights like the Manker E14. The battery is able to handle the power, but you got to reduce output due to the heat. Looking at the X7 or the M43, you can have these lights powered by a single Li-ion cell I think. We’re talking about 30A total, and the light is only able to handle the heat for about a minute or two. This means you don’t have to look at what an IMR cell can handle continuously, but only what the higher pulse current is. I believe some cells can easily realise this.
This automatically means that with a 4x18650 high drain set-up, you can get a lot more than the current offerings of around 10000 lumens. But this is simply not done, primarily due to heat. Therefore, higher power LiFePO4 cells are not going to change that game I think…
We need new leds with much higher efficacy if we want to see radical changes in light output…
So what do you think Mr. Barkuti?
Cheers ^:)