I was going to address this before, but I let it slide. That graph shows a constant resetting to turbo instead of letting the programming kick the light down a mode and stay regulated at the lower output. The direct drive pattern you're seeing in that graph is the limitation of the cell, not the light. No single cell setup is going to be able to sustain that high draw for any extended period in a regulated manor and no 2 cell setup is going to be able to sustain an MT-G2 on turbo for any extended period of time either. Even Thrunite and Eagletac lights drop modes with their 3 cell setup because it's too high of a thermal demand, even disregarding the power requirements, for a light that size.
Sorry to say that any light from any manufacturer is going to be limited by the same rules. Look at the severe lumen drops on Thrunite or Nitecore. That's why I don't like lights that push for the highest output they sacrifice everything and play word games with run times just to achieve 1000+ lumens or just to achieve 100 or so lumens over their competition. That's not even going to be noticed at those light levels.
