[net.audio] Digital Remastering

wjm@lcuxc.UUCP (B. Mitchell) (05/22/85)

Digital Remastering is, at best, a necessary evil, and at worse another case
of major record company hype.
It is necessary to produce a CD version of an older (classic) recording
that was not recorded digitally.  However, now one even sees "Digitally
Remastered" re-releases of classic analog performances on LP!  Somehow,
I fail to see how taking an analog master tape, digitizing it, and then
re-converting the digitized version to analog to cut an LP master will
produce higher quality results than a careful transfer to LP of the
original analog master tape (like those done by Mobile Fidelity) - and
they then have the gall to charge digital prices for the results!
(I can see a case for digitizing if one is going to do some signal
processing, but that should be done with extreme care, if at all!)
What digital remastering does, is take the analog master tape, digitize
it, perhaps re-mix it, and produce a digital version.  This is then either
copied onto CD's or converted back into analog form to produce LPs and/or
analog tapes (cassettes or open-reel).
Regards,
Bill Mitchell ({ihnp4!}lcuxc!wjm)

mwilliams@mahler.DEC (Mike Williams 229-6258 LTN1-2/B17) (05/23/85)

[]

I don't claim to know anything about the commercial recording process.
I would, however, guesstimate that digital remastering is worthwhile for
the following reason -- the integrity of signals recovered from an analog
tape decays as the tape is used over and over (e.g. record production runs),
although I can't say at what rate this occurs. As long as the 1's and 0's 
are distinguishible from the digital tape, the integrity of the resulting
analog signal remains constant.

Mike Williams
Digital Equipment Corp.

thomas@utah-gr.UUCP (Spencer W. Thomas) (05/26/85)

One advantage of "digital remastering" comes in if you are going to make
MANY copies of the new master.  Each time you copy (play) an analog
master, you lose a little fidelity.  Each time you copy (play) a digital
master you lose nothing.  

-- 
=Spencer   ({ihnp4,decvax}!utah-cs!thomas, thomas@utah-cs.ARPA)
	"There are only the pursued, the pursuing, the busy, and the tired."
	- F. Scott Fitzgerald

greg@olivee.UUCP (Greg Paley) (05/28/85)

> I fail to see how taking an analog master tape, digitizing it, and then
> re-converting the digitized version to analog to cut an LP master will
> produce higher quality results than a careful transfer to LP of the
> original analog master tape (like those done by Mobile Fidelity) - and
> they then have the gall to charge digital prices for the results!
...
> Bill Mitchell ({ihnp4!}lcuxc!wjm)

This wouldn't make much sense if that were all that's involved.  If,
in fact, an LP master were actually cut directly from the "master tape",
it would, as you said, be adding an extra, unnecessary link into the
reproduction chain.

Since only two channels can be cut on a standard stereo LP (forgetting
about the now-obsolete CD-4 format), you need to have a two-channel
medium as input to the master.  This, as I undertstand, was even the
case with the so-called "4-channel matrix" (SQ and QS) discs.  However,
there are very few stereo master tapes that are actually only two-channel.
Even the early stereo Mercury and RCA recordings from the 50's were done
on 3 tracks, and currently it's not extraordinary to have a master tape with
24 or more separate tracks.  This means that at least one mixing/copying
step is required to produce the tape that will, in turn, be used for the
disk master.

Since at least one copy must be made prior to disk mastering, and since
the digital process supposedly results in a more perfect copy (I'm not
saying it does or doesn't - that's a whole different topic), it makes
some sense to use digital equipment to produce that copy.  

This does beg the question of why we are also seeing "digital remastering"
of mono recordings whose master tapes are single channel and wouldn't 
thoeretically require the mixing/copying process.  Again,
my understanding (and I'd be happy if someone who knows otherwise would
say) is that in these cases it is also standard to use a copy, rather
than the master tape itself, to produce the disk master.  The reason
for this can be questionable - it's often a chance to "enhance" the
sound by means of various EQ tricks, producing fake "stereo", etc.
which would be probably better left undone.

Charging "digital prices" is yet another matter.  I feel it was a sneaky
way of raising standard LP prices to begin with and certainly don't find
it justified for reissues.  At least the EMI/Angel reissues I've seen that
involved digital remastering and the DMM disk mastering process have been
at a lower price than their "standard" series.


	- Greg Paley

peters@cubsvax.UUCP (Peter S. Shenkin) (05/30/85)

In article <> thomas@utah-gr.UUCP (Spencer W. Thomas) writes:
>
>One advantage of "digital remastering" comes in if you are going to make
>MANY copies of the new master.  Each time you copy (play) an analog
>master, you lose a little fidelity.  Each time you copy (play) a digital
>master you lose nothing.  
>
One thing I've always wondered, and even though it's not quite within the
provenance of net.audio, now seems a good time to ask it.  When you copy
(play) a digital master, isn't there a finite proability of mis-copying a
bit?  I mean, even if parity bits are used, there's a finite probability
of mis-copying an *even* number of bits, and no matter how many double-
checks one cares to make, there's always a *finite*, though small, prob-
ability of a miscopy getting through all of them, it would seem.  Thus, isn't
it incorrect to state that "you lose nothing"?  I mean, think of how many
bits there must be on a CD.  (OK, how many?  I'm sure many of you know!)
Even with a very low error rate there must be a few that have been mis-
copied... and if you know what the error rate is, and also a ballpark figure
for what percentage of bits would have to be wrong for an audible effect,
one could calculate how many generations of copies would be necessary to
audibly degrade the signal -- I agree it's going to be a big number!
(Assume random distribution of wrong bits....)

Thus, the *nature* of signal degradation through repeated copying would 
seem to be the same with analog and digital encoding... it's just that 
digital *technology* allows one to keep the error rate down much more
*cheaply*.  Is this a correct understanding?  (Obviously, I'm not an ee....)

Peter S. Shenkin	philabs!cubsvax!peters		peters@cubsvax.UUCP

ron@brl-tgr.ARPA (Ron Natalie <ron>) (05/31/85)

>  Somehow,
> I fail to see how taking an analog master tape, digitizing it, and then
> re-converting the digitized version to analog to cut an LP master will
> produce higher quality results than a careful transfer to LP of the
> original analog master tape
A master is the last step before the record.  Typically, multitrack recorings
are done on 24 track machines.  It is then mixed down to two.  Every pass
through the analog/magnetic chain eats away at the S/N and frequency response.
Changing this to digital whenever possible will improve things

> they then have the gall to charge digital prices for the results!

Here you have a point.  Good digital equipment is cheaper than good
analog equipment.  Unfortunately, what a recording is worth is how
much someone will pay for it, not how much it costs to do.

> What digital remastering does, is take the analog master tape, digitize
> it, perhaps re-mix it, and produce a digital version.  This is then either

You can't remix a master tape except to mix up the two channels.  Masters
have not more channels than the final product.

-Ron

sjc@angband.UUCP (Steve Correll) (05/31/85)

> When you copy
> (play) a digital master, isn't there a finite proability of mis-copying a
> bit?  I mean, even if parity bits are used, there's a finite probability
> of mis-copying an *even* number of bits, and no matter how many double-
> checks one cares to make, there's always a *finite*, though small, prob-
> ability of a miscopy getting through all of them, it would seem.  Thus, isn't
> it incorrect to state that "you lose nothing"?  I mean, think of how many
> bits there must be on a CD.  (OK, how many?  I'm sure many of you know!)
> Even with a very low error rate there must be a few that have been mis-
> copied... and if you know what the error rate is, and also a ballpark figure
> for what percentage of bits would have to be wrong for an audible effect,
> one could calculate how many generations of copies would be necessary to
> audibly degrade the signal -- I agree it's going to be a big number!
> (Assume random distribution of wrong bits....)

You're right that there's a finite probability of degradation whenever you
copy a digital signal.

The degradation mechanisms for digital and analog are qualitatively
different, however.  Copy an analog tape through a channel which
increases the amplitude of the signal by 5%, and after 100 copies, your
signal will be 1.05 ** 100 times too big. But if you know that the
information on the tape is a digital bitstream, you can in theory copy
the signal without error: whenever the signal is less than 50% of the
maximum, you reduce it to the "0" level, and otherwise you raise it to
the "1" level so the 5% corruption vanishes on each copy. (This digital
capability is called "noise margin".)

Of course, other degradations affect analog and digital equally.  A
tape dropout can destroy a bit completely. But compared with analog
techniques, digital techniques make it much easier to spend bandwidth
to buy error-resistance. The more redundant information you supply when
encoding a signal, the lower the probability of an uncorrectable
error.  (One reason for the enormous number of bits on a CD is the
presence of vastly more redundancy than a simple parity scheme would
provide.) But, as you have observed, the probability is still
non-zero.
-- 
                                                           --Steve Correll
sjc@s1-b.ARPA, ...!decvax!decwrl!mordor!sjc, or ...!ucbvax!dual!mordor!sjc

steve@amdimage.UUCP ( system admin) (06/02/85)

There seems to be a lot of talk about digital recording, and 
quite a few questions too. First, digital recordings and analog
recordings are both done on magnetic tape (yuck), at least 
until a CD-ROM appears in the recording studio. This means that 
the recording has a very limited life, due to tape stretching,
nearby magnetic fields, etc. The one thing that digital recordings
do to improve this is introduce a highly sophisticated error
correction technique known as the Cross Interleaved Reed-Solomon
Code. Without going into the details of what the code does, 
suffice it to say that the raw recording bit-error rate is
in the neighborhood of 10E(-4) to 10E(-5), and by using the
Reed-Solomon code you can reduce your bit-error rate to about
10E(-15). This is 10 orders of magnitude improvement.

As for parity calculations, they only yield a 50% improvement
in error *detection* (not the same as correction).  Even protocols
like XMODEM only yield about a 10E4 improvement in bit-error
rate, and this depends on the packet size transmitted.

Off onto another subject, thid discussion of CDs vs. LPs.
The one thing that most of the discussions on the net have 
failed to point out is the dynamic range of an analog recording
can *never* be as good as that of digital recording. This is because
of the compression involved to be able to keep your turntable's
cartridge from mistracking when very loud passages are played
back (1812 Overture cannons). The digital recording is still 
a matter of bits being interprets and your CDs laser can track
any bit pattern you throw at it. 

As an editorial in the June issue of Digital Audio observed,
the reason audiophiles aren't attracted to CDs is they like to
fiddle. They're not happy unless they've just rebalanced their
tonearm to perfection or replaced every capacitor in their preamp.
CDs don't have much of anything to fiddle with except the programming
features.

Anyhow, if you're really interested in the technical points of
CDs, I recommend the article "Communications Aspects of the
Compact Disc Digital Audio System" by J.B.H. Peek in the
Feb. 1985 issue of the IEEE Communications Magazine. Dr. Peek
works for Philips Research Labs.

Enjoy, 

steve

amdcad!amdimage!steve