I think you may have missed my point: many original sources of music are capable of emitting energy at frequencies that are not well handled by digital equipment. Correctly engineered, maintained and operated, analog recording systems can capture those frequencies and pass them along. Of course I cannot hear them, but that is not the point. Digital equipment behaves very badly when fed with data outside the frequency spectrum it is designed to receive, and the bad behaviour is heard as poor sound quality. To avoid this issue, you have to filter the unwanted frequencies beforehand. Over steep filtering causes phase shifts in the remaining signal, while gentle filtering avoids the phase issues but either fails to remove the unwanted frequencies, or removes part of the desired input signal. I work with seismic data in the oil and gas industry, and this is a very common dilemma that seismic data processors face every day. Some good digital music equipment handles the issue well, but most cheap kit does not, and the result is pretty poor sound quality.
The analog systems I refered to above are never inexpensive, but a lot of great music has been recorded with this sort of kit: crossed-pair mikes running into high speed half-inch reel to reel tapes; and cut to high quality heavy vinyl disks, printed in limited numbers. Those disks, played with very sensitive moving coil cartridges, into amplifiers powerful enough to deal with the very low amplitude output from the MC cartridge, make very good sounds indeed.
Can you do the same with a digital system? Of course, but are we really expected to believe that bcause of some mysterious property instrinsic to the digital world, proper engineering is no longer the difference between digital systems or recordings that sound like crap, and digital systems that sound great? I don't think so, and that proper engineering typically isn't cheap. pxd