i think the straighter the path (less processing) the better. granted you need processing of some kind always...

if the system is setup up correctly, sizing and delays, etc. you wont get delayed sub frequencies.

i would think that the signal would be processed (mixed up or down, etc.) before the delays are considered. delay is more of an effect on the output, so its natural that it would occur later than manipulating the actual sound.

its good that man. are doing these auto-calibration systems, because it eliminates a lot of the guessing of whats best. it electronically measures things, like distance, etc. taking into account resistance in speaker wire, etc.