On Sun, Nov 28, 2004 at 06:49:15PM -0500, Chuck Swiger wrote:
> Richard Rauch wrote:
 [...]
> >(I can readily record up to 48Kbps, but if recorded
> >at that level, am I just retaining imperfections of the medium with
> >higher fidelity?
> 
> That's about right.

Do you have a number that you would suggest for a sampling rate
for "typical" tapes?


> >Or, if the tape is in good shape, is there always
> >room for better sampling, due to the analog nature of the tape?)
> 
> Analog signals don't have infinite resolution.  This is a popular 
> misconception that people have with digital versus analog recordings.  Most 
> people learn something about how CD's work, and they compare a digital 
> "stairstep" signal used in many examples versus the smooth curves of a sine 
> wave and (correctly) decide that the two aren't comparible.
> 
> The crucial factor missing in the popular understanding of digital audio is 
> that "stairstep" signal produced by the DAC's *must* be put through a 
> so-called brickwall or reconstruction filter at the Nyquist frequency, 
> which is half of the sampling rate.  In plain English, the "stairstep" 
> signal with square edges contains very high freqencies which go away when 
> you filter them properly.

Does this also mean that, given the human hearing range (something
like 20Hz to 20KHz in rough terms?), CD quality recording is nearly
as good as one could get---for raw audio playback---no matter how
good the tape is?

-- 
  "I probably don't know what I'm talking about."  http://www.olib.org/~rkr/