Re: What is the audible effect of 180 polarity change between HF and mids?
While I don't know about any asymmetry in level sensitivity to polarity, apparently there is some difference because we can perceive polarity with some specialized (asymmetrical) signals. This is subtle and I am not aware of any evolutionary advantage this offers.
The brain evolved during our formative cave man days to reward the ability to discern first arrivals, to determine true direction of sound sources. This can make a life and death difference when those sounds are coming from predators. Dead cave men did not reproduce, so those who hear better have more offspring and that trait is reinforced. This is a strong perception trait, and the basis for all delay based source localization enhancements (delayed speakers, etc).
I suspect in multiway systems, the leading edge of sounds first emerging from different drivers, (especially when opposite polarity), will sum and play together differently than the more persistent pitched tones following after that initial transient. This initial sum is the lesser flaw, than the latter sum being more wrong more of the time.
Ivan noted that the tone bursts don't sound as loud as constant tones... This is simple math, if the burst is on x% and off y% of the time, the average energy is that fraction of the continuous. Loudness is more of an average than peak SPL phenomenon (thus so much compression). By this same metric the leading edge summation is less of the signal than the following, most of the time so transition region summing is more audible.
Loudspeakers are always full of compromises, learn to manage the sundry evils in proper proportion to their audibility.
JR
