Conference napalm::commusic_v1

	Ones and zeros.  What ones and zeros?
	Jesse Norman is singing.  At one instant she produces some
	a capella sound pressure level near her microphone.  The difference
	in air pressure around the diaphragm of the mic makes it move.
	It moves in an electric or magnetic field depending on the type,
	and a voltage or current is generated, respectively.  The current
	ultimately is measured as a voltage also.  The voltage of her voice
	in one instant of time goes into a mixer, which these days is
	probably digital, and is digitized before mixing.
	There are different approaches to conversion, but type an SAR
	(successive approximation register)
	collects bits that correspond to a binary number in one type of
	code or other (two's complement or unsigned binary or 2 or 3 others).
	This is written to the tape.  Or in the case of records it is encoded
	as little square depressions in the wax.  No only kidding.
	Anyway, the tape is used to make CD's and you buy the CD so why do
	you need a DAT?  
	When you play the day, the binary number, probably normalized
	two's copmlement I don't know it doesn't matter (nothing I don't
	know matters  ;-) typical engineer's arrogance),
	is read off the tape and sent in a digital-to-analogue convertor,
	which typically assigned different (2**n) weights to the bits in
	the word by allowing different amounts of current to flow for each
	bit into a summing network which is a varying current and
	is detected by an op amp and
	amplified into a varying voltage that is amplified even more by
	a power amp and drives the speaker da dit da dit da.
	Tom	

    Uh, could you repeat the question...
    

  Your conception on how the data is converted back to what we hear (analog)
  is wrong.  It is simply the reverse of how the sound was recorded.
  Recording is called Analog to Digital and is done thru ADCs (Analog to Digital
  Converters), playback is just Digital to Analog using DACs (you guessed it).

  It's basically a stream of digital numbers that represent the amplitude of
  the sound wave at specific intervals of time, with some interpolation
  between the intervals.  The less time between intervals, and hence the
  less time the machine has to "make up", the better, which is why faster
  sampling rates yield more realistic reproductions of what was recorded.
  The standard sampling rate for CDs is 44.1 Khz, DAT I believe is higher,
  like 48 Khz.

  Of course there's different methods and degrees of quality/accuracy in how
  different CDs and DATs interpolate between intervals, do error correction
  of incoming data, and convert to analog (ie some manufactures boast 18 bit
  DACs, etc)  At best they're subtle difference in my opinion.  They all sound
  great, even the cheapest ones.

  If anything, you'll find quality difference in how well they do error
  correction, which is why you'll find a CD that plays fine in one player
  but skips on another.

  This procedure is nothing more that "sampling", and you'll find many
  notes in here on sampling/samplers as well as DATs/CDs.  Try a DIR/TITLE,
  or keywords.  (One note that sticks in my mind because of the title is:
  DADADADADADADADADADADADADADADDADAD)  Also check the AUDIO notesfile.

-TR

A question for the engineers --

	O.K. so what's the latest '1 bit' technology? 

Richard.

Well, I could tell you about my drill, but that would be boring. It's a pretty
sharp topic, though, if you get the point- that's the key issue.

Well, that was bracing; back to work...

    Thanks for the quick replys, I have a much better understanding of 
    how these thing now work. The discussion on voltage levels shed a lot
    of light....
    
    
    Now.....about those $200 DAT machines....
    
    Thanks again,
    Gary