Re: Lionheart CD: The Distorion Chronicles

[dbk@tove.cs.umd.edu (Dan Kozak)]

In article <22190@hydra.gatech.EDU> gt4586c@prism.gatech.EDU (WILLETT,THOMAS CARTER) writes:
>In the interest of trying to guess the source of the nastiness on my copy
>of the Lionheart CD, i'll make a few comments as to why my system is
>blameless.  First of all, when i use my cd player i set my PS Audio 4.5 
>preamp on STRAIGHTWIRE, so there is essentially a direct connection between
>the amp and the CD player, and therefore I cannot be overloading my preamp.

Ok, quick test: set everything up as you have described and then turn
off power to the preamp.  Do you still get sound?  If so, then your
preamp's STRAIGHTWIRE setting _is_ a passive connection and your
conclusion is valid . . . if turning the preamp off causes sound to
cease, then their is some active component in between (buffer amp or
whatever) that could potentially be distorting.  Next check, hook the
CD to the amp directly (ain't component systems grand? :-)

>Second, my amp is not causing the distortion because 1) my Adcom GFA-555 has
>distortion lights which go blinky-blink when the amp starts to clip, and
>these lights never came on, and 

I can't make any definitive comments on this other than to wonder what
these lights are supposed to show, i.e. output stage distortion, input
stage distortion, driver stage, or what?  I know that Adcom makes high
quality gear, so I'm sure that these indicators indicate something
important, the question is what. (see next item)

>2) the distortion can be heard at all volume
>levels, from very quiet to loud.  

Important info.  Where is the volume control in this circuit?  On the
power amp, or on the preamp?  What is being indicated here is that the
distortion has occured/is occuring before the volume control.
Depending on the placement of the volume control, this could include
the input stage of the power amp, the input stage of the preamp, the
buffer stages of the CD player itself, or of course, the source
material.

>That leaves the only possible system
>culprit as the CD player, and I think we both agree that it is 
>impossible to overload a properly designed CD player.

Mmmmm . . . depends on how you define "properly" :-).  Since there do
seem to be audible differences between CD players, I have to assume
that there is room for variation between them -- headroom in the
filtering and buffer amp stages may be one variable parameter.  When I
said that digital doesn't distort, I was referring to record, not
playback (see next item) and in any case the buffering and (possibly)
filtering stages of a CD player are analog.

>I think, therefore, that the distortion was recorded onto the CD.  The next
>question is where in the recording chain did it appear.  If the RIAA
>equalization is applied to the cutting lathe after the master tape, then I
>guess that somebody got really lazy with the track level settings when
>mastering the master CD and let them overload.

But that would produce glitching, not distortion.  A digital audio
signal models an analog waveform as a series of numbers that represent
the amplitude (level) of the waveform at discrete intervals.  The
highest amplitude signal would be represented by a binary number where
all bits are 1.  As I understand it, when a signal louder than this
comes along, the sampling circuitry gets confused and produces a
binary number that is, for the purposes of this discussion, random
(it's not actually, but in audio terms, it might as well be).  So
overmodulating a digital sampling stage produces "glitching" which
sounds something like dropping in very short samples of Einsteuzende
Neubatten or a metal foundry or a full tilt aerial assault over top of
everything.  It _does_not_ sound like the analog distortion (or tape
saturation, or whatever) that we all know and hate.  Of course it's
completely possible (tho' unlikely, mastering engineers are well paid
for a reason) that the distortion occured in some analog stage on the
path between the master tape and the first digital stage.  It's not
likely that the master tape itself is the culprit, since I've never
heard that _Lionheart_ was remixed for CD, so the master tape is
presumably the same one used for LP and the other CD, etc.

If your really interested in this, you could try A/Bing a few
components to see if a different CD player makes a difference, etc.
(you could also try the CD on other complete systems) I do know of at
least one recording (_What's_New_ by Linda Ronstadt) where the
mastering (in this case LP) was so hot that the peaks would distort
almost any phono preamp out there.  I know this because my father uses
it to show off _his_ design (which it doesn't distort), but I wonder
about everyone else out there . . . On the other hand, I don't
remember _Lionheart_ being very peaky, rather the opposite.  As Alice
once said, curiouser and curiouser . . . 

#dan

Clever:         dbk@cs.umd.edu    | "Softly her tower crumbled in the 
Not-so-clever:  uunet!mimsy!dbk   |  sweet silent sun." - Nabokov