Re: Music Roll Punch Advance
By Craig Smith
Wayne Stahnke's comment about punch advance was very informative; you learn something every day! He stated that the Ampico advance is 356 per foot rather than 360. That's only 0.030 inches per step or 30 per inch. Lets say the paper has expanded 1%. That would be 356 X 0.01 = 3.56 punch advances -- pretty close to the 4-punch error cited.
It seems to me that the original manufacturer would be much more interested in having the ability to divide the timing into the correct relative note timing than in having the right number of notes per inch. In fact, I'd be surprised if a person could set the timing lever to anywhere near 1% accuracy, so having the exactly correct number of holes per inch wouldn't really mean anything. The fact that the measurement is precise has little to do with whether it's accurate.
Craig Smith
[ Editor's Note:
[
[ "The fact that a measurement is precise has little to do with whether
[ it's accurate." Quite true, and until further independent data is
[ available we can only guess what the dimension accuracy is. Besides
[ answering the question, "What advance distance was intended," this
[ distance is needed to launch the phase-locked-loop process which tracks
[ the time-axis events during transcription to computer file.
[
[ The note-axis dimension is known accurately, because it is a published
[ standard dimension (9 holes per inch, or 72 holes per foot). It is
[ risky to assume that the old paper has expanded equally along both axes,
[ else we would know the punch advance distance along the time-axis with
[ equal accuracy.
[
[ The paper roll was punched by a synchronous perforator, and it's data
[ is a duplicate of the binary data of the master roll. The goal of
[ the transcription process is to recover the binary data with no error,
[ i.e., 100% data accuracy. The musical interpretation of the binary
[ data is a different, independent process. Fortunately, if the
[ transcribed data is 100% accurate, then the dimensional accuracy is
[ required only if a music roll of duplicate dimensions is to be punched.
[
[ Robbie Rhodes
|
(Message sent Tue 3 Dec 1996, 14:47:10 GMT, from time zone GMT-0800.) |
|
|
Mechanical
Music Digest™ Archives
