Re: Which Should Sound Better? / Common to screw-up Dithering?

[Joe Hartley <jh@brainiac.com>]

Before I jump into the question, some background:  the three letter designation
on CDs refer to recording/mixing/mastering and is called the SPARS code.
( https://en.wikipedia.org/wiki/SPARS_code )  On a CD, the last letter will
always be 'D'.

That's a little misleading, though - what's "the master tape" in this 
instance?  Is it the tape that has the multitrack recordings on it,
or is it the tape that the multi-track got mixed to and sent to the mastering
engineer(ME), or is it the tape that came back from the ME and got sent to the
vinyl pressing plant?

It's unlikely that they had the multi-track, as they certainly would have 
mentioned a re-mix.  Down that road lies trouble, though.  Frank Zappa decided
to re-mix some of his albums and re-release them at one point, and the results
ticked off a number of older fans who expected to hear Hot Rats sound one way,
and got something rather different.  (Very good, but different, and not everyone
appreciated the changes.)  If these had been re-mixed, I'm sure Tom would have 
been involved.

If it's the tape that came back from the ME, which is my best guess, then
what we're really looking at is an "AAAD" recording, in which the analog
master is re-mastered to digital.

On Fri, 11 Nov 2016 18:57:54 -0500
"Leo Casey" <LeoCasey@comcast.net> wrote:
> Q1: Which of 2 hypothetical cds, in theory, should sound better: 
> 
> 1) ADD: Using original 1979 master tape of 1st album, TOM VERLAINE, that
> first converts the analogue signal to 96k Hz, 24 bits, and then does a
> second conversion to 44.1 k Hz, 16 bits from which the commercial cd is
> created and packaged. (****Assume, for sake of argument, the dithering is
> done correctly.)
> 
> OR 2) ADD: Using original 1979 master tape of 1st album, TOM VERLAINE, that
> directly converts the analogue signal to "just" 44.1k, 16 bits, from which
> the commercial cd is created and packaged?

It used to be Number 1, by a wide margin.  Remember when CDs were first 
coming out, and albums we all knew sounded a little more brilliant, if not
downright harsh?  That's because they were produced by method #2, when 
44.1/16 was the best that the hardware could handle at the time, and dithering
was not widely used in the process.

> Q2: Is the dithering process to convert from 96k Hz, 24 bits to 44.1 k Hz,
> 16 bits commonly screwed up by record labels (like BGO Records or other
> labels)? [Albeit cds are probably going the way of 8-track tapes : > ) what
> with mp3s, cloud storage, etc. Except for audiophile nuts like me.]

These days, no.  Dithering's widely understood and it's easy to apply in
moder Digital Audio Workstations (DAWs), which is a fancy name for a computer
and an audio program like ProTools, Reaper, or MixBus.

So why bother going to 96/24 since the result is going to be a CD anyway?