Re: minidisc generations

["Ken M" <wgold@mecasw.com>]

>someone said (god knows who) ...

'twas me

>I just wanted to clarify the posting that this was a response to (not my
>post, but I have a nice example).  In addition to any loss you get in 
>going
>to hardware, you'll get the compression error.  Here's a way to look at 
>it.
>

True, my argument didn't take the MD's compression ratio into account,
it was a more broad stroke answer to what seemed a rather broad
question.

 I interpreted the question as "Is it worth it to do digital copies when I
don't
seem to notice any tangible degredation using analog lines to copy."

And as a broad answer to a broad question, Digital -> Digital copies are
many
orders of magnitude better than going Digital ->Analog / Analog ->Digital.

Of course when talking compression, there's always lossage, BUT... Since
the data is compressed at record time onto the MD, then the D->D transfer
from one deck to another should copy the pre-compressed data so it doesn't
have to go through another compression (I would HOPE this is how it works,
please correct me if I'm wrong).  BUT, if you go from the MD's digital
information
to analog (Uncompress the information, send it through the D:A converter)
and into another MD recorder (Into another A:D converter, and now run
the compressor on the "New" information) You are going to end up with
serious
generation loss after a while.

It's like (Going back to your JPG reference) converting a JPG (Lossy
Compression)
to GIF (Lossless Compression), and then back to JPG.. The image will not
handle
many time doing this without a noticable blurring on the edges of the data.

My last post on this off-topic topic... Glad to debate in e-mail though.

Ken M.
wgold@gnu.ai.mit.edu