[ Brendan Adamson wrote in 040318 MMDigest:

> My problem is with Piet Paardekam's roll template program, MIDIBOEK,
> obtained from http://huizen.daxis.nl/~Ppaardekam/midi1e.htm  ...
> ...  For some undiscovered reason, whenever I print larger
> files of 20 or 30 pages, it chokes and says "problem with printer:
> error code 55".  I cannot split up the file into two parts without
> having his measure and page numbers be incorrect, and the numbers
> cannot be set that I know of.  I would like to have a nicely printed
> copy.

While the main question regarding MidiBoek is out of my experience
(I did download it and play with it a few years back), I have worked
a lot in the past with Postscript and printers.

I do not have a reference at hand to the cryptic error codes from the
printer driver.  Most take effect when the printer runs out of what is
called "Virtual Memory", or VM.  When this happens then the system is
no longer stable for generating more detailed errors.

There is a program called Ghostscript, which is supported as part of
the GNU project.  A Google search will find it.  This tool will allow
you to get a stack dump at the error and allow for bumping up the VM
allocated.

Acrobat Distiller is the Postscript core that resides normally in a
printer.  This will give a log of the error conditions, however they
are hard to interpret without understanding the file generated by the
driver.  Most drivers are set to image the output to a rectangle the
size of a page.  This means that what ever is used to number the pages
should allow for page splitting.

Instead of splitting the file, try using the "Page range" option in
the driver.  This will allow only say pages 1 through 5 to be rendered.
Then one can render, say, 5 through 10, and so on.  This might help if
it is a memory resource issue.

About six years ago I wrote some pure Postscript routines to image MIDI
files as Wurlitzer 165 perforation masters.  I made this code available
to the MMD.  In case it is not easy to find I attach it again.

The file clara1f.ps was written as a pure Postscript program.  It can
be opened with Wordpad.  Since it will generate a few hundred sheets
of paper, I have set it to only print the first 3500 ticks.

I am also including the (rather large) PDF of the output from a
distillation of this program.

If you have a Macintosh, you might take a look at a program I have
on my web site for roll scanning and transcription:
  http://www.delectra.com/rolltransport

This program, called "Maddalena" after an Opera character, is still
at the alpha development stage; eventually I intend to rework the
Postscript notation into it along with a number of other features.

This program will eventually allow for click and drag of the
perforations.  At present it only works with converting a special type
of file called an SCN file which was defined by a group called the
"International Association of Mechanical Music preservationists",
  http://www.IAMMP.org/

Other members have done a lot of work on the windows side, to allow for
statistical recovery of the original master of piano rolls.

Another program which is useful for simple tunes is "ABC" MIDI, which
could easily be extended to do what you want with the generation of
roll masters for the understanding of how they are put together,
although it mostly works with simple folk melodies.  This is at
  http://www.gre.ac.uk/~c.walshaw/abc/

Let me know if there is more that I can help with.  Most of this I am
working on as I have a large collection (of over 700) books from 35
instruments which I am currently scanning indexing and archiving as
time permits.

Julie Porter

( A reply to Mr. Adamson the poster was sent with attachments;
these should be in the MMD Archives somewhere.)