Herbert Von Karajan asked to store his Beethoven 9th symphony in one CD only, that's why 16/44 was choosen.
While digital data generally are a no problem, always same data from source to receiver (checked and approved), when you consider digital audio which has "time" the battlefield changes a lot.
We're not even near to understand how digital audio data could be converted to analog in a perfect way, how it's possible to conduct an empirical blind test and conclude something?
A similar test could be done without changing anything with same souce, setup and methodology and still we'll have 50% of tested ears that will swear they've clearly listened a difference. What will be the conclusion? The glass is half full or half empty?
I can add that a 16/44 setup is easier to implement than a 24/192, a mathematical theory supports me, and for 24/192 systems we're just at the beginning of development.
If you do the test to have a secure conclusion you need a perfect setup, let's recap just a few things:
original recording studio, mic, recorder, converter, support, player, converter, preamp, amp, speaker and cables everywhere.
Even if state of the art or alien components are used it'll never be perfect as is mathematics.
Maybe my ears are limited to 35Hz-15Khz but I'm still able to pick sound quality differences, I don't care if mathematics tells me I can't, I like the sound which is nearer to my reference reality.
P.S. Frome scientists and mathematics I'm still waiting an answer about black holes and the beginning of the universe.
Edited by Thraex - 9/11/12 at 4:35pm