===================================================
= NB: Over 50% of this message is QUOTED, please =
= be more selective when quoting text =
===================================================
"David W. Tamkin" <[EMAIL PROTECTED]> writes:
> CDs and MDs both store sound sampled at 44.1 kHz, but a CD has 588 samples
> per frame (75 frames per second) and an MD 512 (about 86 frames of stereo or
> 43 of mono per second -- Sony decks with single-frame resolution offer 86
> places per second to divide a mono track but half of them are not valid; you
> actually get a division one position earlier or later). Track marks can come
> only on frame boundaries, not in the middles of frames, true?
>
> So, in digital transfers from MD to CD or from CD to MD, would that mean that
> when the track number changes in the source, the destination unit is likely
> to be in the middle of a frame and cannot mark a new track until the begin-
> ning of the next frame? Should the destination unit pad the current frame
> with silence and continue recording audio in the next frame, or might it con-
> tinue recording samples that belong in the next track but mark the track
> start slightly late?
>
> I've occasionally found, in copying from MD to CD with no silence between
> tracks, that the CD track mark can be slightly late, but only at the ends of
> lower-numbered tracks on the CD. Starting every MD track with eleven frames
> of silence (six in mono) has been enough to prevent that, but if the reason
> is the framing difference, then two frames of stereo silence (or one of mono)
> at the start of each MD track -- enough to make sure a CD frame boundary can
> get into the silence -- ought to be enough. That wouldn't explain why it's
> never happened beyond the beginning of track 6, though. Maybe it's just be-
> cause after it happened twice I've always left silence at the start of every
> MD track before copying it to CD to avoid making more coasters, and on the
> two where it happened, coincidentally later tracks framed similarly on the
> CD to the MD?
David, you're attention to detail would make you a great scientist.
My sense is that most recorders' track mark mechanisms do not run in
perfect synchrony with the input samples (as you have surmised), and
if the track change doesn't coincide with a frame boundary, it will
get pushed out so that it appears to happen in the next frame. I can
imagine it getting pushed out even farther, there's probably a lot of
processing invoked on a track change. (Remember those occasional
CD->MD digital recording glitches that have been reported, wherein a
*wide* discrepency between CD and MD tracks occurs?) Your "safety
padding" sounds like a reasonable solution.
Rick
-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]
