A few comments regarding inconsistencies...
One source of inconsistency is the normal variation of the electrical components. To deal with this I do a calibration on each unit, but that calibration isn't perfect so there is some variation. For fw4 I have come up with a new way to do the calibration that should theoretically give much better results and just finished building a programming/calibration/test interface unit that will enable this much improved calibration. Initial results are promising so i think this will make a significant difference.
The switch to 18650s and higher power should also help. The fw4 heater is more capable of keeping up with faster draw speeds, so you won't get users on one hand who draw slowly and get great consistent results and users on the other hand that like to draw faster and get less consistent results (maybe the faster technique works on a full charge but not low for example).
btw, the higher power is also seeming to improve draw restriction on my development unit. I think the main source of draw restriction (probably for any vape) is when some of the vapors condense right away on the downstream herbs and screen area. You might also get some oils migrating to that area without ever vaporizing. The result is an oily clumpy area right behind the screen that reduces airflow. Higher power can help reduce this because you can't make the mistake of drawing too fast, which causes the temp to drop too much, making the problem worse.
I am sold! Where do I sign up for the waiting list?
This update in the calibration process alone justifies the wait for the FW4. I officially declare my search for my Arizer Air replacement to be completed. If the FW4 is going to live up to all of the promises while being consistent from unit to unit, I see it being the most sought-after vape this year once it has been released. Every single point of criticism mentioned in this thread would no longer exist. I would really appreciate if you kept us informed about the progress you are making.