Re: MD: Future MDLP with variable compression?

[Eric Woudenberg, Minidisc.org Editor]

Hi Le!

Le MD Freak <[EMAIL PROTECTED]> writes:

> I'm an old MD lover and I've seen the introduction of
> the MDLP format with great interest.
> 
> Now, I don't know much about how the LP is working,
> but do you - better knowing persons - know if it might
> be possible to introduce a LP with a VARIABLE
> compression? It would be very interesting! If you
> can't get all of your recording into a SP recording
> you just turn the knob and adjust the recording time
> from e.g. 74 min to 90 min - from a SP to aprox. a
> LP1.2 recording so to say. It will be more compressed
> than the SP be not as much as a LP2 recording. You
> will use all the space you need for your recording,
> and it will be the least compressed possible.
> 
> But maybe it's all a dream. Maybe the algoritms in the
> ATRAC won't permit something like this. Or what do you
> say?

It's a good dream, it might even be possible. (I assume by the way
that you are not talking about dynamically variable bit rates, as with
MP3 VBR encoding, rather a fixed (but user selectable) bitrate for an
entire track.)

The principle problem is that currently the MD's ATRAC/ATRAC3 encoding
sizes are set to fit neatly into a soundgroup (212 bytes). If the MD
designers are willing to be flexible about that relationship (and it
would be a serious programming headache for the MD micro-controller
programmers, to be sure), then it should be possible. I think this
would also complicate the user interface for DIVIDE, since you would
have virtual (i.e. encoded) soundgroups that spanned sectors. You
might have some weird restrictions on edit points (i.e. ending up with
irregular edit intervals).

Which brings up a question: What's the edit point accuracy of an MDLP
machine in LP4 mode? Is it 11.6*4 == 46.4ms, or 11.6*2 == 23.2ms? I
would guess the latter since the smallest addressable unit of the MD
TOC's 3 byte Cluster/Sector/Group disc address is one SP mode
soundgroup (see http://www.minidisc.org/md_toc.html#sec0) which is 212
bytes, and two stereo LP4 frames fit in there.

And the SP mono editing interval is 11.6 ms also, right?

Rick


-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]