Regulator Noise

[Don Tillman]

   Date: Thu, 18 Sep 1997 10:58:42 +1000 (EST)
   From: Paul Perry <pfperry at melbpc.org.au>

   well looking at these figs I have to say that in a practical system it isn't
   the intrinsic noise from the reg outputs that will be the limiting factor.

Not necessarily.  I learned about cap bypassing on variable regulator
chips the stoopid way; I breadboarded a circuit, listened to it, and
said, "What the hell is with all that noise?".

   Unless the layout (PCB and hookup routing) is truly amazing then other
   effects will far swamp anything from the PSU. It is like the high end
   audio people, with .00003% thd amps driving speakers with distortion
   ten thousand times greater.....

The "golden ear audiophiles" have an important point to make.  They
just have a habit of talking gibberish when they try to describe it.
I doubt it's intentional, but it's a lot like a Zen Koan in a funny
way.

For example, in the .00003% amplifier distortion case, the enlightened
student will see a source of great wisdom:

  "Is some given amount of distortion going to annoy the listener to the
  same degree regardless of the process that causes the distortion?  Of
  course not.  .01% distortion from one process might be screetchy as
  all hell, while 10% distortion from another process might not be too
  bad.  Heck, it might even compensate for some other shortcomings in
  the system.  That's what you get for taking a many-dimensioned
  attribute and try to squeeze it into a single number.  Take a listen
  while I crank 'Radar Love'." 

  -- Don