audio CD's in SCSI drives -- syntax?

[martinko]

Joerg Schilling wrote:
>> I think it might be worthwhile throwing out cdda2wav as the reference 
>> tool completely and instead rewrite the tutorial to feature cdparanoia, 
>> which is pretty much the state of the art ripping utilitiy these days 
>> (and has been even longer on Linux), works with device names rather than 
>> unit numbers for SCSI drives as well, does in fact not require the 
>> specification of any device by default as it will autoscan for devices 
>> and just use the first one it finds and finally is used as the ripping 
>> backend by nearly all modern GUI ripping utilities. 
> 
> Please try to inform yourself to avoid writing incorrect claims......
> 
> Cdparanoia started as a patch on cdda2wav in 1997. It did never improve it's 
> CD-DA code and for this reason, it's CD-DA properties are based on what 
> cdda2wav could do 10 years ago. The code is full of non-portable parts.
> 
> Cdparanoia died in 2001.
> 
> For this reason, I did pull out it's essentials and created a portable
> and bug-fixed version of the paranoia code as "libparanoia" in 2002.
> Then cdda2wav has been enhanced to use the libparanoia code.
> Since 2002, cdds2wav it definitely the best choice for CD audio extraction.
> 
> Monty did a great job when creating the code that is in libparanoia but he is 
> no longer interested in cdparanoia. 
> 
> Cdda2wav reads CDs that cdparanoia won't read at all and it usually even does 
> not need a dev= parameter at all! This is something Monty did try to implement
> a long time but never was able to devliver. Cdda2wav in attition writes a 
> summyry line for each track that tells you whether there are audible problems 
> in the extracted data.
> 
> Do you really like to give up maintained code in favor of a dead program?
> 
> http://cdrecord.berlios.de/
> 
> Jörg
> 

Joerg,

I can see the following in man cdda2wav :

        -paranoia
               use  the  paranoia  library  instead  of cdda2wav's 
routines for reading.

Now I'm not sure how to use that state of the art extraction -- with or 
without -paranoia parameter ?

And BTW why are there two methods of extraction, why not to have only 
the better one ?

Cheers,

Martin