A simple Leabarjan roll punching and editing machine should not
cost too much.  Not sure how available these are.  I started to make
something similar last year.

At this time of year, I usually work on getting new music prepared
for my 20-note busker organ, so am in the process of cutting rolls.
Surprisingly, a simple X-acto knife and a sheet of glass is all that
is needed.

A decade or so ago,  I was going to make a perforator.  The main issue
was, and remains, getting an accurately ground die set.  The other part
that seems to get overlooked is space.  If using a punch then it needs
to either be really sharp, or at least three or more sheets of paper
are needed to get some density to the drift.

All this paper must go somewhere.  Loops or tables are needed to keep
it from twisting under the punch die.  This means dedicating a whole
room to the perforating.  All the setups I have visited indicate one
would need at least an 8 foot by 10 foot area to locate the equipment.

Once punched, the paper will need to be spooled, leaders attached then
boxed -- so more space for storage is needed.

What will be used for the masters?  At the foot of my bed sits 700 book
stencils: a stack of boxes 2 feet wide, 5 feet deep and about 3-1/2
feet tall.  In the garage is a metal shelving unit with my APP rolls
taking up a lot of the space.  Ironically the whole collection of MIDI
files fits onto a memory stick.

Over the last few weeks I have been working on some new Macintosh
programs to help with all this.  A simple viewer for the compressed
scan files which are called CIS files.   Using some of the newer Apple
code writing tools, I am working on a dedicated database, which will
let me sync the files between different computers and memory sticks.

The most interesting thing is that I was able to get a short segment
of the printed grid paper used to create the Wurlitzer master stencils.
These are scaled 3 to 1.  I was able to work out that the spacing of
the dots is metric.

In effect I have written a program, which acts as a virtual perforator.
Since there are a fixed number of dots per quarter note, when a MIDI
file is plotted out against this grid one can see the errors introduced
by the capturing process.  Most MIDI files are scaled either 120 or 480
ticks per quarter note.  When printed this is about 4 centimeters on
the grid, which makes viewing the scans real easy.

Robbie, the MMD editor, has also kindly supplied me with an arrangement
of the "Wallis and Grommet Theme" for my Caliola, so I am able to use
this as a test set for my virtual perforator program.

Much of what I have working on, can also be found in the Midiboek program.
The main difference is that my programs run on Apple equipment.  I am
also able to handle the lock and cancel registration.

In theory I should be able to run these new apps on an iPhone or iPad.
I also run some of my programs directly on the printer.  This way
I have a new overlay stencil which to cut or punch through.

Taking a lesson from Mike Kitner, I find that one can staple the master
stencil onto the roll paper.  There is just enough give in the staples
that one can keep the edge of the master aligned.  I have yet to do
this with the APP spacing but it does work well with the 20-note busker
organ stuff.

Mostly I find that the cost of perforating is time and space.  It takes
one hour per foot to cut a roll by hand.  Interesting is that a 20-note
busker or 44-note APP roll, has about the same number of notes and
chords per measure.

Julie Porter