> The problem is, sampling aliasing is sampling aliasing... how are 
 > the digital filters to separate it from the signal?  And on the D/A 
 > output, there's still significant filtering necessary to remove 
 > artifacts.  Use of oversampling, though, was exactly my point.
        
        The thing is that many so-called "analog" advocates seem to think 
this problem is a "digital" one.  *ANY* bandwidth limited system, analog OR 
digital is going to show some phase shift and harmonic (e.g. square waves) 
problems near the upper boundary.  If you put a square wave through a top of 
the line cassette deck at 15kHz, you'll get out a noisy sine wave.  A CD will 
also put out a sine wave, albeit without the noise.  Don't even bother with 
the LP.  The problem is not inherent to digital, but to bandwidth limited 
systems.  True representation of a square wave would technically require an 
infinite bandwidth, which is highly improbably, and quite possibly impossible 
to realize in the real world.  I dare say phase shift is also much less with 
the CD than the LP or a cassette and even reel-to-reel tape.  The very BEST 
analog recording systems can do better in some areas, but digital can be 
improved and has been improved on the professional level.  The consumer 
market only needs to catch up.  The question really is, is it even 
neccessary?  There are much better ways to improve sound reproduction than 
adding more bits and higher sampling rates.  Multi-channel recordings (say 
around 8-12 channels) to more accurately reproduce the original soundfield is 
one such example.  Properly done with the new DVD medium, the additional 
channels could be combined back down to 5 channel, 3 channel, or even mono, 
if neccessary, in the digital domain.  One could simply upgrade at will for 
better performance.
--- FLAME v1.1
---------------