Audio CDs have very good error correction, as demonstrated by Tomorrow's World when they were first made in the early 1980s. They can cope with many, many radial scratches across the track, although long ones parallel with or tangential to the data track will cause problems, and can even play a CD with 3mm diameter holes drilled in the data area of the disc, as long as there aren't too many of them. If you want to know how good the error correction actually is, then get a drive that can read the raw file format, and use an error checking program. On a CD-R burnt as an audio CD with printing on it, you can actually see the printed pattern in the errors on read.
There is a good explanation of the error correction system on audio CDs here:-
I've used "dd" to make copes of cd-roms. AFAIK it only works for single-session CD-ROMS - anything that can be represented as a single 'ISO' image file.
The other method of imaging CDs (CUE and RAW files) can, I think, image multi-session, audio, and mixed audio/data CDs.
Yes. However, bit perfect reads are not a requirement for consumer audio while they are fairly important for data. Hence the extra error correction on data CDs.
Yes, that is what I meant to say, but wasn't concentrating on what I was posting properly. My bad :(
Lets see if I get this right (from memory again).
CDs use data interleave with bits spread out around the disc to minimise the effect of damage.
On audio discs losing 1 bit of each of a hundred or so bytes on a track can cause a bit of noise, but it is not disastrous to the average user. Losing the entirety of a handful of bytes, on the other hand, will cause at best an annoying "pop", at worst make a track unlistenable.
On a data disc, losing anything is not acceptable, hence the addition error correct as Anssi stated.
Yes, the data correction is spread out along the spiral track, to protect not just a single block but several blocks. The spreading out of the correction code provides extra redundnecy to protect against scratches (but not extended scratch damage that results from tangential scratch marks - which is why we're advised to wipe the surface with a cleaning cloth in a radial direction and to avoid circular motion - scratch damage is a risk due to the possibility of a grit particle or two becoming embedded in the cleaning cloth).
That's not even close to how the error correction works. In the light of familiarity with the 'Lossy Compression' techniques employed in most audio storage formats (and picture file storage, both still and moving), a simpler way of describing the error correction and masking technique employed by the audio CD format is to regard it as a "Lossy Correction" method.
For mild errors, the correction is lossless, for more aggressive errors, the correction starts to make use of reconstruction techniques (aka 'faking it'), initially, by simply repeating the earlier samples to fill the gap, perhaps interpolating into the following error free samples to hide what could otherwise be a very nasty glitch eventually resorting to simply muting the playback when the damaged section has wiped out more than a second or so's worth, ulimately giving up on streaming audio playback altogether.
The standard Audio CD playback data rate is 176.4 KB/s[1] whilst that of a single speed CDROM is 150KB/s. The extra 26.4 KB/s is sacrificed to the more robust ECC required to satisfy the requirements of lossless data storage (that's on top of the audio ECC from which the
176.4 KB/s audio data stream is sourced from).
[1] The raw data stream rate from which the 176.4 KB/s audio stream is extracted is a little bit higher again than this. ICBA to google an answer to this so I'll leave that as a question for an expert to answer. ;-)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.