GG: Tobasco, Noise, etc.

["John P. Hill" <jphill@frank.mtsu.edu>]

Hi, Tim!

On Sun, 19 Jul 1998, Tim Conway wrote:

> I understand your Tabasco analogy but I had thought that, with today's 
> computing power, it would be possible to go further than removing obvious 
> clicks and the like. I imagined that a really clever algorithm could 
> perform a comprehensive Fourier analysis, or similar, and with human help 
> separate the purely-GG noises (ie, piano strings and mechanism, humming, 
> fingers striking keyboard, chair creaking, etc) from non-GG noises (eg, 
> tape-drive mechanism hissing and whirring, doors closing, technicians 
> coughing, subway rumbling). Clearly I was wrong, but do you think that it 
> will ever be possible to electronically separate the many strands of 
> individual sound apparent to the human ear so that they can be recombined 
> in whatever combination suits the listener? Or is the Tabasco-removal 
> problem insurmountable?

It looks like some others on the list have already speculated, but it's
an interesting question.

I think that it *will* become possible to do what you've eluded to above,
but then the aesthetic question becomes *should* we?  Another list member
made the (quite valid) point that he would like to hear *all* of what was
on the original recording, tape noise, HVAC, door slams, etc.

I guess I come down somewhere in the middle on this one.  I can tell that
some of the Sony GG Edition CDs have incorporated some serious
post-production work in terms of noise-reduction, EQ, etc.  I generally
find these to be pretty convincing.  I *don't* like the sound of the CD
that has the Brahms Intermezzi on it, although I really like GG's
performance of those pieces. Part of what I don't like is the "ocean of
white noise" problem combined with a rather poor spectral balance and
extension.  I think a bit *more* work in post-production might have
improved this. It just ends up not being a very involving listening
experience for me, but then, I'm a serious audio guy/recording engineer.

On the other hand, I have old albums by Theloneous Monk and Bill Evans
that have pretty glaring sonic problems on them and I'm somehow able to
ignore these and give myself over to the music.

Miles Davis' KIND OF BLUE would be a good study on the relative merits of
digital post-production and remastering.  I know of at least four or five
different CD versions of this classic work and they all sound *very*
different.  Check 'em out.  I very much doubt that most people would
prefer a noisy 1:1 transfer or the rather poor-sounding stock Columbia
CD that was the only one available through most of the 80s.

This opens up some serious aesthetic questions that, to my knowledge, have
not been widely discussed or addressed in sound recording periodicals or
texts.

jh