Robbie commented in 030415 MMDigest:
>[ * The data quality isn't related to the data storage format.
>[ The "Standard MIDI File" (SMF) data format is routinely used to
>[ store precise data for replicating music rolls, as well as for
>[ storing imprecise transcriptions. Don't blame MIDI -- it's simply
>[ "garbage in garbage out!"
In fact, a Standard MIDI File could store all the chaining data, with
a sustained note broken up into the bridges and repeated holes exactly
as scanned from the roll. The exact start and end time of each
perforation would be preserved for duplicating real rolls or figuring
out what the original punch advance spacing was, etc. The only problem
I see is in defining where a perforation starts and ends; is it from
the leading edge of the hole, or later by the tracker bar hole's
radius, or what?
Of course you wouldn't play this file, since an electronic "piano" is
much faster than a player piano action and will re-strike every one of
those holes. A simple (?) program could process the exact file to
bridge the gaps into sustained notes, simulating the slow player
action, to yield a playable MIDI file.
Mike Knudsen
[ The music roll perforator accepts distance data that defines the
[ position, in inches or centimeters, of the center of the hole to be
[ punched. Time is a not a perforator parameter. The state of the
[ note channel data, 'on' or 'off', is checked at each revolution of
[ the perforator crankshaft.
[
[ The perforator is nothing more than a simple computer-controlled
[ automatic machine tool, such as a vertical mill. The modern
[ computer that controls it is a direct descendent of the sprocketed
[ master piano roll and the 8-level punched paper tape control systems
[ of earlier years.
[
[ Perforator control data may be stored as MIDI data if a relationship
[ is defined that relates the MIDI "time" event to distance along the
[ paper. The relationship is straightforward if the music roll is
[ played while pulled by a capstan, as in a crank-organ. If the music
[ roll is read while pulled by a take-up spool turning at constant RPM,
[ then the relationship of time and distance is non-linear.
[
[ -- Robbie
|