Re: Do CD's Degrade?, and DCC Question

[larry@cs.com (Larry Spence)]

In article <neilg.717356145@sfu.ca> neilg@sfu.ca (Neil K. Guy) writes:
>mujad@uxa.ecn.bgu.edu (James A. Drenter) writes:
>
>>In simple terms, yes it is currently believed that CDs begin to oxidize
>>after as few as 10 years. [...]

It is currently believed (by some) that marking the edges with a green
pen magically improves the sound.  A lot of things are currently believed,
especially by companies who make gold CDs, green "CD marking pens," etc. %)

> If you want to see this right now, hold an older CD up to a bright
>light. You'll probably see a number of bright points of light on the
>disc. 

You can see these on many brand new CDs, too.  

>Those pinholes are apparently spots where the aluminium has
>oxidised and is thus transparent. 

Nope.  Pinholes are spots where the aluminum didn't coat the CD when it
was manufactured.  I have never, ever heard of pinholes developing spon-
taneously over time.  Pinholes, unless they're really bad, are easily
handled by the CD's error correction.  I bought a used CD the other day
that was pressed by one of those awful Canadian plants -- countless small
pinholes, hundreds certainly -- and it plays perfectly.  

Aluminum oxidizes to alumina (Al2O3?), which is extremely hard and very
shiny.  Corrosion from atmospheric pollution could eventually be a problem.
One person wrote me in email that they had observed a tiny bit of corro-
sion at the extreme outermost edge of some CDs, but it hadn't progressed
in far enough to get to the data.  One person in rec.audio (where this has
all been beat to death before) has kept a CD outside on a peg for many
years, in the direct sun.  Once a year, he wipes the bird droppings off
the disc and verifies that it still plays perfectly.

>In fact, if you look at one of the
>early CDs, which were made when oxygen contamination was more of a
>problem than now, you'll see quite a peppering of little bright
>pinholes. (for some reason these pinholes always give me the
>creeps...) So one day all our CDs may well be totally clear and thus
>unplayable.

Not unless you scrape all the aluminum off them. %)

> I hear also that a number of very early CDs used poorly formulated
>paint for the lettering on the front, which ate away at the
>polycarbonate. (the silkscreened side of the CD is the thin side,
>closest to the actual aluminium coating.) Supposedly this is no longer
>a the case.

True.  But only a tiny number of CDs pressed by MPO in France were
affected.  This was a long time ago.  Current MPO discs are fine.

> This is usually not a major problem, especially with higher quality
>players. Good CD players with a decent tracking mechanism can
>accurately track the laser and compensate for data loss by filling in
>the gaps - extrapolating your favourite Kate number. So when you
>listen to a CD a small percentage of the sound you hear was never made
>by Ms. Bush at all, but is a kind of computer best guess. Sort of.

Not even close.  A CD has to be pretty badly messed up before interpo-
lation occurs.  Almost all errors are _perfectly_ recovered by the 
error correction code.  The bits are identical in all but extreme cases.
There's no "guessing" involved in error correction; your player as a
matter of routine corrects hundreds of raw errors per second (even on
a gold disc).

>However, cheaper CD players often skip and otherwise choke on badly
>pinholed CDs. You can try this by putting thin strips of black tape on
>the shiny side of a CD, from centre to edge (radially). Lousy players
>can't play a CD with a strip of tape on it, good ones can sometimes
>compensate for a few strips. Of course, this can be risky thing to do
>as you can completely wreck a player if the tape comes loose, so
>forget I ever said that.

Just take a worthless disc and write on it with an opaque black marker.
Try it and you won't worry about a few pinholes any more.

> If you're *really* worried about this sort of thing, you can always
>buy gold CDs which don't oxidize. The life of these discs is limited
>by the lifespan of the polycarbonate plastic itself.

The whole oxidation thing is a marketing hype.  CDs have been around 
for 10 years, and confirmed stories about "rotting" CDs are very, very 
rare.  Statistically, if CDs tended to "rot" in, say, 15 years, we'd
be seeing a decent number of them starting to die now.  It's sort of
like gerbil stories; many people have "heard about it" but no one has 
much convincing evidence.  Not to say that there haven't been bad runs 
of CDs (e.g., Talking Heads _Naked_ apparently had a lot of problems), 
but CDs that have been fine for 5-10 years are not just up and dying.

> (disclaimer. I'm no expert on this stuff. I've picked up this through
>reading the odd audio mag and books on the subject of CDs and the
>like. Corrections appreciated should I be wildly wrong on anything)

Which audio mag/book said that pinholes are caused by oxidation?!?
Sounds more like something you'd read on usenet... %)

-- 
Larry Spence
larry@cs.com
uunet!csccat!larry