πŸš€ go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: rec.audio.tech
Article #1107

Re: Audio quality of CDR, any theory ????

#1107
From: Dave Platt
Date: Fri, 18 Dec 1998 00:00
41 lines
3382 bytes
In article <75e4j7$bon$2@vixen.cso.uiuc.edu>,
Stefan Kozak <skozak@vixen.cso.uiuc.edu> wrote:

>Arny KrοΏ½ger (arnyk@flash.net) wrote:
>: There are two kinds of errors on audio cd's. "Correctable" and "Concealable".
>
>: The correctable errors are precisely numerically corrected using redundant
>: data off the disc.
>
>slightly incorrect.. i believe the errors are determined and corrected by 
>error correction/detection parity codes such as hamming codes. up to 2 possibly three bits can be corrected for in any 8 bit data read. (actually 12 or 13 bits (with parity))
>
>in addition, hamming codes can be used on any word bit length.

Well, both ways of saying it are correct... the error-detection codes
can be considered to be "redundant data".

The actual coding is a two-level cross-interleaved Reed-Solomon
algorithm.  It's capable of completely correcting error bursts of up
to 3874 bits, and can handle bursts of up to 13282 bits by
interpolation (i.e. bursts of up to this length will result in an
uncorrectable error in one sample, surrounded by samples which are not
in error, and thus a linear or other interpolation between the valid
samples can conceal the error pretty effectively).

The interleaving is an important part of the error correction process.
By interleaving the data from multiple frames before it's recorded on
the disc (and de-interleaving it when it's read back), the algorithm
"spreads out" the effect of an error burst.  Instead of having all of
the data bits from one sample wiped out by a divot, you end up having
one or two bits damaged in each of many samples, spread across quite a
few blocks.  Since the number of scrambled bits in each block is
small, the redundency inherent in the Reed-Solomon error correction
codes can be used calculate the original values of these bits,
resultin in accurate correction of the error.

-- 
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!
Technical Details
Message-ID: <75eevh$eah$1@remarQ.com>
Path: arc1-ca.pugleaf.net!unknown!not-for-mail
References: <753ar1$p8u$1@justice.csc.cuhk.edu.hk> <1C5AE5CA0F2F9ADC.A0C451A2A5364D00.E599AFEBD487605B@library-proxy.airnews.net> <zbue2.2011$IO1.1954@news.flash.net> <75e4j7$bon$2@vixen.cso.uiuc.edu>

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII
X-Google-Thread: 109f11,48d41a4dac9741b4
X-Google-Attributes: gid109f11,public
X-Google-Thread: 10d85e,48d41a4dac9741b4
X-Google-Attributes: gid10d85e,public
X-Google-Thread: f6728,48d41a4dac9741b4
X-Google-Attributes: gidf6728,public
X-Google-Thread: 10ff83,48d41a4dac9741b4
X-Google-Attributes: gid10ff83,public
X-Google-Thread: fb1f8,48d41a4dac9741b4
X-Google-Attributes: gidfb1f8,public
X-Google-Thread: 105a81,48d41a4dac9741b4
X-Google-Attributes: gid105a81,public
From: dplatt@feghoot.ml.org (Dave Platt)
Subject: Re: Audio quality of CDR, any theory ????
Date: 1998/12/18
Message-ID: <75eevh$eah$1@remarQ.com>
X-Deja-AN: 423726462
References: <753ar1$p8u$1@justice.csc.cuhk.edu.hk> <1C5AE5CA0F2F9ADC.A0C451A2A5364D00.E599AFEBD487605B@library-proxy.airnews.net> <zbue2.2011$IO1.1954@news.flash.net> <75e4j7$bon$2@vixen.cso.uiuc.edu>
X-Complaints-To: newsabuse@remarQ.com
X-Trace: 914014001 PJ0EG9MGB2146AA01C usenet53.supernews.com
Organization: Posted via RemarQ, http://www.remarQ.com - Discussions start here!
Newsgroups: rec.audio.tech,rec.audio.opinion,rec.audio.misc,hk.rec.audio-visual,uk.rec.audio,tw.bbs.rec.audiophile