RE: [AV Media Matters] CD's Robust nature-cigarette burns

[jnorman@maxoptical.com]

Richard,

You are correct,  there are several levels of Audio error
correction,  E11, E22, E32.  Each of these levels are increasingly
worse with the E32 being uncorrectable.  If E32 errors are
encountered the tone will either be averaged,  held or completely
dropped out.   This is what can cause the music to sound differently
and if the Drop Outs are present you will hear the all to familiar
pop and clicking sounds.

There is also another specification of Disk testing know as Block
Error Rate (BLER)  If the block error rate gets to high we have seen
a difference in the sound compared to the same material on a CD with
low BLER rate.  Not necessarily bad sounding but different from the
original source.

John Norman
Max Optical
www.maxoptical.com

-----Original Message-----
From: lists@richardhess.com [mailto:lists@richardhess.com]
Sent: Monday, June 25, 2001 4:48 PM
To: AV Media Matters
Subject: RE: [AV Media Matters] CD's Robust nature-cigarette burns

At 01:31 PM 06/25/2001 -0400, jnorman@maxoptical.com wrote:
>Sally,
>Audio CD Players are designed to compensate for defects in the CDs but
>what you must understand is there is error correction being applied to
>your music.  If the error correction is heavily applied the sound or tone
>of the music can be changed.  So even though you could here the music from
>a CD with the cigarette burn it most likely did not sound the same as the
>original source.
>John Norman
>Max Optical
>www.maxoptical.com

John,

Please help me out--I'm confused here. I thought there were actually two
flavors of things happening:

Error correction: Isn't this the built-in redundancy --doesn't this have to
do with the Reed-Solomon code and all those things that I gloss over (the
preferred choice to glazing over)--that actually can fully correct errors
so that the output bitstream is the same as the input bitstream?

Error concealment: Doesn't this take over when the error correction's
capability to correct is exceeded? This may introduce audible artifacts
into the program.

I've read (but don't feel that I know this for a 100% fact) that when one
rips an audio CD track via a CD-ROM drive, at least the level of error
concealment and perhaps one level of error correction is bypassed. Can
someone confirm this with a serious URL?

Thanks!

Richard