🚀 go-pugleaf

In article <3678CAFC.985389AC@interstroom.nl>,
Chris Cotellini  <ccotellini@interstroom.nl> wrote:

>> The pops are almost always due to data loss. This usually happens when the CD
>> is read.
>
>Then why isn't the "pop" sound there when you play it on the pc cd-writer? If it
>is a read error and not a write error you should hear it when you play the
>original disc. If you don't hear the faults when you play the original cd (on the
>burner) it can't be a read error, can it? Or does the copy process somehow make
>these faults more noticable?

The "reading" side of the copying process makes the faults more
audible, because (on many drives) it bypasses the important "error
concealment" process.

What happens fairly often is this:

-  There's a scratch, dust blob, or manufacturing defect on the
original CD.

-  The problem causes an error burst which is too severe for the
two-level Red Book C1/C2 error-correction algorithm to deal with.
After error corection is performed, the data in the sample buffer
still has one or more bit errors in it. The C1/C2 error corrector has
flagged each sample with errors as "bad".

-  If you're _playing_ the CD (on a CD player or via the CD-ROM
drive), the player/drive's audio circuitry "sees" the bad-data flag on
the bad samples.  It activates its "error concealment" circuit, which
either interpolates between good samples on either side of the error,
or (in really severe cases) mutes the analog output for a few
milliseconds.  In either case, your ears don't notice the error.

-  If you're _reading_ the CD ("ripping" the data to a .WAV file), the
drive delivers the samples from its buffer, even though one or more
samples are marked "bad".  It doesn't interpolate or mute the bad
sample before giving you the data.  You end up with a very sharp
(one-sample-wide) "spike" in the data, which sounds like a sharp
"tick" or "pop".  When you burn a CD-R (or make an MP-3) the incorrect
data is burned accurately - when you play back the CD-R, the "bad
data" flag isn't set, and so your CD player reproduces the glitch
quite faithfully.
  
A lot of CD-ROM drives (and in particular the less-expensive ones)
have this sort of problem - their firmware doesn't implement audio
error concealment when reading digital-audio tracks.  To "rip"
reliabily using drives of this sort, you'll probably need a software
ripper which implements some form of scratch detection and
concealment... doing in your PC what the drive's own firmware isn't
doing.  A few years ago, when doing the CD-ROM drive for the 3DO
videogame system, I had to insist on a drive firmware revision from
the CD-ROM vendor - they weren't doing error concealment, and we
didn't have the CPU horsepower to do a decent job of scratch removal
via software methods.

Other CD-ROM drives do an excellent job of error concealment, or
automatically re-read audio data to try to do a better job of C1/C2
error correction.  Plextor SCSI CD-ROM drives have a good reputation
in this respect.


--
Dave Platt                                        dplatt@feghoot.ml.org
Visit the Jade Warrior home page:   http://feghoot.ml.org/jade-warrior/
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!