I've been doing my temp testing and have some findings on the SXK
I started out testing Ni200 but was then rather surprised to find my Ni200 wire does not have the expected TCR of 0.006 / 0.0062! Rather it's right down at 0.0045. So while the results were useful for this wire, it's not generally applicable. I have ordered some more NI200 from a different supplier so that I can hopefully re-test with more generally applicable wire.
What I did learn, as you will also see below, is that the NP offset required for "low resistance" is far smaller than anticipated. It's still there, but it's in the region of 5 - 10 rather than the 10, 20, 30 type offset that my calculator was suggesting.
Below is a copy of a post I just made to the Beyond Ni200
thread, where I first verified the TCR of my Germany Titanium wire (0.0035 as expected) and then tested this with varying NP settings on my SXK.
More details, including a graph of the same coil on the Dicodes 2380 (somewhat more accurate, as might have been predicted!) in the Beyond Ni200
thread
from this post onwards.
SXK NP setting for a 0.41Ω coil read as 0.35Ω using Zivipf Titanium Grade 1 (TCR 0.0035): recommended NP40
NP 35 is the 'right' setting according to the TCR; NP 40 is the most accurate setting. So the resistance drop appears to require a slight increase, but nowhere near what my calculator indicated.
Putting these numbers into the calc gives a recommendation of NP 51! Clearly that is way, way off.
As to why, I cannot yet say. Most likely conclusion is that it's using a different resistance figure for the TC calcs than the one it shows on screen. Maybe the "low resistance" is really mostly a display problem? Hard to know as yet. I will try and do some testing on the output volts of the device to see if I can conclude anything further from that.
What we can see immediately is that SXK's TC is rather less accurate overall - jumps a lot both above and below, so the target is more a 'trend' than a limitation. Though this is still a dry coil situation, a test where it's reasonable that it doesn't do super well (even if its more expensive competitors do better.) A wet coil temp test will be more interesting.
I believe the resistanec displayed on screen is not the resistance used for the chip's TC calculations. It's even possible that it's purely a display issue - we had some thoughts on that when the volts didn't seem to match, although I thought we did also see volts that were too low for the real coil resistance.
Actually all I want to do is vape without trying to remember all my thermal dynamics classes verbatim. But oh no I find myself Googling things like Thermal Coefficient, Bernoulli equations and Venturi calculations in an attempt to find out what to set my TP at. 
