[paranoia] DATparanoia?? What do you think?

[Monty]

> So I've been looking into reading DAT's with DDS-3 drives which potentially
> can read up to 7X. The problem is these drives don't have audio firmware.
> I have discovered though that linux 'dd' and 'mt' command can be used
> to read DAT tapes as if they were data. Of course this introduces new problems.
> One is that all the subcode information is read in as well as the audio data.
> A 10-15 line C program can strip these. The second problem is the tough one.
> When DAT's are read as data any digital errors *on the tape* are not corrected.
> This is different that errors during reads. Interleaved correction is used
> when reading the data but is not applied to the audio data so errors that
> are corrected when doing digital playback are propagated.
> 
> Enter cdparanoia?
> 
> Does anyone think that the cdparanoia library might be used to implement a 
> DATparanoia?

Unlikely; cdparanoia is meant for a very specific purpose, and that is
to deal with limitations in the design of reading RedBook audio, and
vendor-specific inconsistencies from drive to drive.  Little/none of
the code in cdparanoia would be useful for reading DAT this way.

This is *not* to say that DATparanoia would not be a useful tool; just
that the current cdparanoia code is not really useful to you.

> I'd like to do some more research and would like to understand some of the
> error correction techniques used in cdparanoia. Where do I start?
> Any references?

None other than the code.  The problem with reading CDs has more to do
with overlapping and loss of streaming errors than it is errors to be
filtered by the ECC layers.  I assume that your streaming is fine, and
you simply need somethign to apply the Reed-Solomon-like error
correction, hopefully present in the data, to your rips.

Monty

--- >8 ----
List archives:  http://www.xiph.org/archives/
Paranoia homepage: http://www.xiph.org/paranoia/
To unsubscribe from this list, send a message to 'paranoia-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.