Archive of the former Yahoo!Groups mailing list: MOTM

previous by date index next by date
previous in topic topic list next in topic

Subject: Re: [motm] Re: digital source and its medium puzzle- the answer

From: ixqy@...
Date: 2002-11-06

In a message dated 11/5/02 10:59:47 PM Central Standard Time,
ringmod45@... writes:

> The problem is caused by the wobble, thereby creating an error. Since
> you have lost the bit in question, you can not recover it


It's been 20 years since I've read about CD encoding technology, so can
someone correct me if I'm wrong here?

I seem to remember reading that the data on a CD is duplicated once or
possibly twice on every disc. That way, is there is a loss of any data, the
error correction tells the laser to go to one of the extra tracks to recover
it. The chances of the same bits of data on both tracks being bad is very
slim, so this takes care of most of the errors.

Also, I seem to remember that the laser pickup is on a transport that moves
up and down. There is a sensor and circuitry that determines how far away the
disc is from the laser at any given point in time. It then physically moves
the laser up and down to follow any movement of disc surface. This keeps the
laser in focus on the disc surface to avoid loss of data.

When data cannot be recovered, error correction then kicks in to the phase
that Paul mentioned somewhere, and interpolates any lost data.

For what it's worth, I used to work at a high end audio shop years ago. One
of the guys from McIntosh (I think it was) came in to talk to us about the
product line. To demonstate the integrity of the error correction on their CD
players, he attached thin strips of masking tape (drafter's tape?) to the
discs. He put them on one by one, radially across the surface of the CD. He
got up to something like 6 or 8 strips until it caused noticeable problems
with the music! This is pretty wild considering that each one of those strips
of tape was about 1/8" thick. Comparing that to the size of each bit on the
disc surface (forgot how many microns that is), you get an idea of how well
the error correction works. This was 15+ years ago, so I'd assume there have
even been more improvements since then.

Andrew