@mijostyn
The .3 dB limit of human hearing is right on. I spend hours adjusting frequency response in 0.1 dB steps and in direct comparison 0.3 dB is the limit of my hearing's ability to detect changes and this is switching back and forth between different curves.
If you really want to improve the performance of your system and learn what you are doing in the process get one of these and get down with it.
https://www.parts-express.com/Dayton-Audio-OmniMic-V2-Acoustic-Measurement-System-390-792
You will not find anything useful with the microphone, because when you reverse the wire, there is no change in the electrical signal at the speaker terminals.
As for linearity, its value is greatly exaggerated. There is absolutely no need to make the linearity of the system with an accuracy of 0.3 db, much less 0.1 db. If it was so important for perception, we would not be able to enjoy listening to cassette recorders, receivers, and so on. By our nature we automatically adapt to changes in the frequency response within a wide range, this is how our perception works. Adaptation to changes takes a few tens of seconds, after that we get used to the new reality and the new frequency response becomes "transparent" for us, it does not interfere and does not help us to perceive music, it just becomes a kind of reference point, relative to which we perceive sounds..
This statement is easy to verify, if you have timbre controls in the system, turn the controls a bit, spend 10 minutes listening calmly and analyzing impressions, make a conclusion.
And do not forget that the linearity of our hearing changes depending on the volume - https://en.wikipedia.org/wiki/Loudness. Why do you set up a system with an accuracy of 0.1 db if with a small enough change in system volume, the linearity of your hearing changes by several DB? Isn't it stupid?