Xref: utzoo misc.consumers:4431 rec.music.misc:10923 sci.electronics:2667
Path: utzoo!utgpu!water!watmath!clyde!att-cb!att-ih!pacbell!ames!oliveb!pyramid!decwrl!decvax!dartvax!eleazar!tga
From: tga@eleazar.Dartmouth.EDU (Greg Ames)
Newsgroups: misc.consumers,rec.music.misc,sci.electronics
Subject: Re: Experience with Compact Disc Club?
Keywords: CD clubs, masters, CBS
Message-ID: <8520@eleazar.Dartmouth.EDU>
Date: 2 Apr 88 04:35:27 GMT
References: <253@antares.aero.org> <1510010@hpfcbpr.HP.COM> <328@mergvax.UUCP> <23843@clyde.ATT.COM> <336@mergvax.UUCP>
Reply-To: tga@eleazar.Dartmouth.EDU (Greg Ames)
Organization: Dartmouth College, Hanover, NH
Lines: 72

In article <336@mergvax.UUCP> ted@mergvax.UUCP (William Klein) writes:

>[stuff deleted]

>A master tape (analog or digital, the final mix tape) is delivered to the
>mastering shop. There it is corrected (?) and eventually converted to
>CD-compatible digital and 'burned' onto a nickel master.

To the best of my knowledge (from talking to friends in the music field,
and reading various articles on the subject), the processes of correction
(noise reduction, re-mastering to digital, etc.) and conversion "CD-compatible
digital" is generally done by the studio, not the pressing plant.  All
most plants do is just that: press CD's, although some do provide preliminary
processing serrvices, for a fee, of course (it's not part of the "standard
deal").  The pressing plant then takes this "CD-compatible" master, which
includes more than just music, including a "table of contents," start and stop
marks for tracks and indices, and SMPTE (pronounced "simp-tee") time codes,
which provide the timing info to your CD player (ie, length of disc, length
of current track, time remaining, play time, etc.).  This digital tape is
then transfered directly to CD, w/out changes.  This digital master is used
to make the metal plates to press the glass dies.

>This master is then used to 'stamp' glass masters. It is apparently good for
>not that many glass masters, I do not remember the number but is less than
>100.  The nickel master goes into the vault. The glass masters are then used
>to 'stamp' the plastic CD's. They wear out after some number of thousand
>pressings and are discarded. The size of a CD production run is generally
>controlled by the number of glass masters made * number of output copies
>per glass master. The reason the glass masters are 'used up' is because
>they have a shelf life of about 1-2 months (breakdown mechanism? I don't
>know). The nickel master has a shelf life (in the vault) of no more than
>6 months.

The main reason that glass masters have a "shelf life" of 1-2 months is
partially because the glass points "dull" from use, but, more importantly,
glass is a *very* viscous liquid, which actually flows.  When you heat glass,
it doesn't actually melt, it just gets runnier and runnier (ie, no definate
melting point).  This is why very old windows (~100 years) distort your view
of the outside world: the glass "flows" to the bottom of the pane, bulging it.

I used to work in an electron microscopy lab.  When slicing thin sections of
specimens for EM slides, you use glass knives formed by breaking 1/4 inch
glass into little right tringles, 1/4" thick, and 1" square on each side.
Glass knives are used because they are sharper and make cleaner slices when
working at such a tiny level.  However, the knives only last about 24 hours.
After that time, even if they aren't used, the glass has flowed enough so that
the edge is now to dull to use.  This is the main reason why the glass dies
go bad; the glass flows and becomes inaccurate.  I don't know why the nickel
plates go bad, but I suppose it is because high levels of oxides building up
on the surface start to plug up the holes (they ARE damn small!).

>When someone else is 'licensed' to press CD's, they start from either the
>mix tape (often in analog) or maybe (rarely?) the final digital repro
>tape. This is why there is definite differences in quality (mainly noise
>floor) between different pressings of the same music (Pink Floyd's DSOTM 
>is a prime example). They are using different masters.
>I hope this is somewhat enlightening, and not too incorrect (as it is
>coming from volatile organic memory).

Yes, but ANYONE making CD's is going to have to start somewhere, and the only
starting points available are the mix tapes, or copies of it.  What is the
difference between, say, Warner Brothers letting Columbia House make a digital
master from their old mix tapes, or WB making the digital master themselves
and giving that to Columbia House?  I can't see how this would make any
difference in quality.

                              -Greg

[-----------------------------------------------------------------------------]
[   Greg Ames                             US Mail - HB 1362, Dartmouth College]
[   E Mail - tga@eleazar.Dartmouth.EDU              Hanover, NH 03755         ]
[-----------------------------------------------------------------------------]