That appears to be a well written piece though I am not capable of assessing its validity. Its worth considering that the well written piece carefully states its terms of reference: hearing, we can't hear tones under 10Hz but we can feel them (as I was personally reminded at a local orchestra last week). My neighbours wouldn't appreciate it but a closer approximation to that experience would include that thump to the chest, not just what we can hear.
After all basic theory (which was correct) _proved_ the limit of information transfer on a phone line to be some 100s of K. The ADSL bods simply used the wires at frequencies they happened to be capable of rather than what was required for the voice network and fed into the theory. Thank you ADLS bods.
Loss being inaudible depends on the psycho-acoustical theories being correct and complete and the algorithms well implemented, lossless compression is better understood and can be trivially demonstrated to be correct (or not) I'm happier with flac than mp3 for those reasons. I'd rather have 24/96 than 16/44 for similar reasons, if there's no difference then so what (space is cheap), if there is a difference I likely have the closer approximation, I can downsample myself for portable devices (until they have sufficient storage for higher resolution).
Biting the hand that feeds IT
