It doesn't matter how you drain a battery, voltage under load is voltage under load.
I'll believe tests over your anecdotal "puff" counts any day of the week.
If you would look at the tests you would see that tests under different loads are different, that's why they test different loads. If you ever used a battery for anything in your life and were aware while doing so you would know that resting them between loads allows them to "recharge". Given our short bursts of load those resting periods do add up to make a difference.
So no, a constant discharge until dead or frequent discharges until dead is not equal to ecig use.
I will take anecdotal real life results over faulty testing methods when real life contradicts the tests any day. But you are free to stick to faulty testing methods if you want.
And it's not really anecdotal, it was a fairly controlled set up with an exact puff count (that were rounded off for here). It is a small sample because it takes a long time to drain batteries in the 3000mah range in real use situations. The one thing that throws it off is the eVic has a small drain even while not firing so while testing I was trying to focus on that one mod.
The usage perceptions in the mech with the NCR-A are purely anecdotal. I don't offer those as proof beyond that they seem to follow along with what I saw in the VV/VW. Which goes against what the metered tests say.
If you want a good test for ecigs it has to be designed around typical ecig use not constant flashlight/laser drain.
What was the puff drawing (what was the device attached)? Personally I look at stats down to 3.2v under load. A constant discharge test is not light on the battery by any means. However at a low discharge current, the non-high drain batteries' higher energy density does provide more discharged energy, the B moreso than the A, so your results seem to be in line with that.
I am not familiar with how the evic puff tester works, can you explain?
eVic is a PWM VV/VW mod with all the bells and whistles (although a low amp limit). The puff counter is just a puff counter that I reset with every battery change.
2.2Ω. Wattage went like this, 10w for 1 second, 9w to 1.3 seconds, 8w/1.6 seconds, 7w/2 seconds, 6.5w up to 10 seconds. You seem to know how those translate to drain in a PWM VV/VW over the voltage range with a boost better than I.
With auto puff turned on so most puffs were 5 seconds long, although auto puff can be disabled by holding the firing button and that happens sometimes
It also has a battery meter that shows percentage so you can see how many puffs you get per percentage point, which I was paying attention to while testing. And it shuts down at 3.2v.
I did run some low and sub-ohm set ups (1Ω and .9Ω duals) though the mech the other day with the NCR-A's and between setting up and vaping two sets of duals the NCR-A did seem to drop to 3.9v quicker but that is to be expected with any battery at the higher drain (the set up before was 1.5Ω dual). Now that I know the NCRs are the same chemistry and I trust the NCR-A's I'll do the math and see what other's I can try in it to compare. Although I have no way (or desire) to provide exact counts with a mech.
BTW, I think you like lighter wire for sub-ohm? I could not get those working right in my IGO-L clone. I liked heavier wire. I think the legs are too long in the RDA. The lighter wire might work better in my RBAs.