On Mon, March 24, 2008 5:26 pm, David Brown wrote: > On Mon, Mar 24, 2008 at 02:12:48PM -0700, Lan Barnes wrote: > >>The fact that cdparanoia has paranoia makes me wonder how necessary it >> is. >>While the time waiting for one's tracks to rip is slow, the time ripping >>bad tracks quickly is wasted. and one wouldn't know without trying them >>all. > > When cdrom drives first came out (think 1X), they basically were audio CD > players with some extra hardware processing the output. Each audio frame > contains an additional layer of ECC and includes the sector address, when > read as data. But, to read the direct audio, the addressing information > doesn't come through. Issuing the command to read raw audio would return > audio data starting approximately where requested, but it wasn't precise. > Programs like cparanoia were needed to recombine the audio data (basically > doing overlapping reads and figuring out what the overlap was). > > It's been completely unnecessary for about 10 years now. Any drive that > reads at a higher speed isn't just a modified audio player, and will have > accurate addressing. Most now use that addressing information to read > audio data precisely. > > If you rip with cdparanoia -Z it will either work fine, or you'll get > massive amounts of skipping. It is unfortunately still necessary, because > there apparently are still plenty of cdrom drives with broken firmware. > > The cdparanoia faq gives some useful information > <http://xiph.org/paranoia/faq.html>, especially why getting the accurate > positioning is hard for the drive to do. > > David > >
Thank you. Very clear. Knlg == Pwr -- Lan Barnes SCM Analyst Linux Guy Tcl/Tk Enthusiast Biodiesel Brewer -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
