OFF: CD Rot

[Paul Mather]

On Wed, 2004-10-27 at 05:54, Carl Edlund Anderson wrote:
> On 27/10/2004 10:11, Jon Jarrett wrote:
> >         Weirdest thing I've seen in this line, or rather heard, was last
> > time Sherman and I were working on Larry Boyd's Disarray stuff. The `first
> > album' we have now put together was generating test discs (burnt through
> > Nero using files created in Cool Edit) which were full of unlistenable
> > crackly fizz. On one machine only. We burnt a second one and the effect
> > had got worse, as if delay had been applied to it. But the blaster in the
> > kitchen played them absolutely clean, while two computers' CD drives and
> > soundcards picked up a tiny fragment of crackle but no more. We couldn't
> > figure out what on earth had happened to the files that made it so
> > disastrous for just this one player...
>
> It could be that errors introduced by a somewhat wonky burner are
> smoothed out by the hefty error correction that's built in (especially
> these days) to consumer audio CD players but that computer CD drives
> (not equipped with such error correction features) cough and spit a bit
> more when they run into a bit of digital gristle.
>
> In my admittedly non-pro experience, computer CD drives are much more
> picky about what they will or won't read than standard audio CD players
> (which you have to feed a pretty suspect slab of CD before they start to
> run into trouble).

Both consumer audio and computer CD players have the same
error-correction abilities (unless you're speaking about "anti-skip"
buffering used in portable CD players, which is a different kettle of
fish).  The major difference between consumer audio and computer CD
players is the quality of the transport: it's usually worse in the
computer unit, due to less well-engineered and constructed components.

A poorer transport will increase the raw error rate coming off the disc
(e.g., because it doesn't track the data spiral as well due to an
inferior quality stepper motor), leading to an increased likelihood of
uncorrectable block errors, which will manifest itself in audible
defects.

Also, the primary function of a computer CD-ROM drive is to read data
CDs, which do have extra redundancy to help out.  (That's why data CDs
have 2048 bytes of data per sector vs. 2352 bytes of data per audio CD
sector.)

Cheers,

Paul.
--
Somewhere in Texas, a village is missing its idiot...