I feel compelled to reply to past posts which contrasted the Cremona "M"
roll with the Operator's "O" roll.  In this the comment was made that "M"
rolls have no advantages over the "O" rolls.

He made the comment that the biggest difference between "M" rolls and
"O" rolls are that "M" rolls could be used with a tune selecting device.
This is not the case, as I will explain.  Let me first say that I am a
fan of both systems, and have heard some wonderful "O" arrangements over
the years.  I enjoy the snappy sound of an "O" roll arrangement as much
as the next guy -- but we are talking true comparison here, and there's a
lot to be said.

First off, when I say that the arrangements on an "M" roll are better
than the "O" roll, it is not only a personal opinion, but is based on the
theory of the two scales when viewed side by side comparing their actual
physical limitations.  Forget about multiplexing rolls by cutting in
actual perforations to multiplex and add further arrangements.  It is
time consuming, and no one is going to do that in any quantity to make a
difference.  For that matter, I could transform a home player piano roll
into an orchestrion roll given a razor blade and a few hours time.

Editing of the music roll does not constitute a fair comparison.   A true
comparison is made only with two unmolested original scales.  Multiplexing
an instrument mechanically to enable a machine to play more instruments
than the roll calls for is acceptable (new construction only, please!)
and I would encourage it.  Multiplexing is a different matter entirely,
and has no place in the discussion of original roll scale comparisons.

The "M" roll does have the ability to play one extra melody instrument
than the "O" roll.  That's really great, but the details go even further.
"O" roll machines built in the 1920's with bellows type pumps had a
physical limitation.  The pumps were very small, so there was not a lot
of reserve pressure for blowing pipes.  In order not to run out of wind,
"O" roll arrangers had to keep their solo arrangements weak and thin,
rarely blowing more than one or two pipes at a time.

You never hear a full chord played on an "O" roll.  Some call this
resulting type of sound more "sophisticated", and in a way, I guess it
is.  But if you like fuller, jazzier, melody arrangements full of chords
and the resulting overtones and harmonics produced, such as the "G" and
"H" rolls can produce, the "O" roll will not and can not give it to you.

The "M" machines built by Cremona, namely the styles Orchestral K and J,
had no such pressure limitations, so therefore were free to do whatever
the arrangers felt inclined to punch into the roll.  The result is a
sound much like the Seeburg "G" roll, however with much more punch, due
to the fact that the melody instruments play solo against the piano in
the same fashion as the "O" roll does ("G" roll instruments play whatever
the top few octaves of the piano does, with no ability to solo).

On top of this, the melody arrangements of the "M" roll are 32 note, as
compared to the "O" roll having 24.  This gives an added eight notes with
which to arrange the solo melody.  An additional "accompaniment" set of
pipes or xylophone extending down into the piano section as was proposed
earlier to "fatten up" the sound is obviously not necessary or desirable
due to the fullness of the "M" roll arrangements.

One final word about the differences in the solo section.  On an "M"
roll, when the xylophone is called for, the arrangement on the roll seems
more like that of an "O" roll.  There are usually about two or three
notes playing at one time, and they tend to "jump around" in a fashion
similar to that of how a human would play the instrument.  When the pipes
are called for, either set or both sets, the arrangement gets more
chord-like structure, perhaps with two or three notes in more of an
elongated pattern, imitating more like a violinist or flutist style.

I think this type of forethought in the solo melodies shows the great
skill and care that went into the arranging of the "M" rolls.  The "O"
roll does not make a distinction in the arrangement style for xylophone
or pipes in any that I have heard.

The "M" roll also has the ability to actually turn the piano treble off
in the solo area, or leave it on so the piano can play full arrangements
with the solo melody instrument.  The "O" roll does not have this abil-
ity, as the piano is automatically "muted" with a felt strip (not really
turned off) in the solo area whenever one or the other solo instrument is
called for.  This feature is certainly not a waste of tracker bar space,
in my opinion.

Some Cremona "M" roll instruments did not have three solo instruments,
but instead only two.  In this instance, Cremona simply tubed the missing
xylophones passage to the piano treble, giving a nice solo piano passage.

The piano scale is very similar in actual practice to the "O" roll,
contrary to what was said.  In order for the "O" roll scale to have a
fuller bass in the piano, a trick was used in which the lower octave of
the piano is coupled to the octave below it.  As a result, anything
played in the first of octave of the piano will be duplicated.  If you
do not couple the lower octave of the "O" roll and "play it for real",
you will see that the two piano scales are nearly the same, with the
exception of the "M" roll having only two fewer piano notes, 64 instead
of the "O" rolls' 66.

The "O" roll does not actually have 78 separate playing notes as was
stated.  It was the octave coupling that brought it up to that number.
I disagree with the suggestion that a clever arranger can do things with
a coupled bass.  There's not much you can do there except octaves.
Anything else this low on the piano sounds horrible.  If you don't
believe me, play it yourself.  Also, refer to page 715 of the Encyclo-
pedia of Automatic Musical Instruments, in which this very topic is
discussed at length with regard to the same technique, which was used
with "G" rolls.

I guess you can couple the bass on an "M" roll too, if you want to get
it richer, (assuming the arrangements don't get too fancy down there) and
bring your playing notes up to 76.  Assuming you tube your piano A to C
as Cremona did, you will play the first note on the piano keyboard,
and with a big resonant upright, this is pretty nice.  I have heard a
Peerless machine that seemed to really enjoy playing that low, and it
was a good effect.

Now a word about the tune selector system.  Cremona pianos used a tune
selecting device on some of their later instruments.  All of these pianos
used the style "M" roll because it was specially scored to work with it.
However, contrary to what was stated, these special perforations were in
_addition_ to the standard 88 holes dedicated to the instruments' musical
and functional demands.  They did not compromise the instruments'
capabilities in order to add function holes for the tune selector system.

Both the "O" roll and the "M' roll use 88 note trackerbars, with the
exception that the "O" roll has two extra holes- one at each end- usually
referred to as A and B, for re-roll and play.  The "M" roll has eight
extra holes- four at each end- used for the tune selector device with the
re-roll and play holes being inclusive in these.  If you are clever
enough, you can build a working tune selector yourself, which you can use
on a new "M" roll orchestrion of your own design.

The statement made that the "O" roll has more capability than any other
9-to-the-inch, 11-1/4-inch-wide orchestrion roll is not true in my
opinion.  Even when compared to a Seeburg "H" roll, the "M" roll will
hold its own ground, although we are talking much more equivalent here in
terms of capability.  The only benefits I see an "O" roll having over the
"M" roll would be a few added percussion effects, if you call that a
benefit.

Most "O" roll machines use way too much percussion.  Many novice
collectors like this additional effect, but in my opinion again, I prefer
the sound of _subtle_ rhythm percussions _behind_ the music, not what I
would classify as a dance band trying to play louder than their
percussionist with an ego problem!

The percussion effects on an "O" roll are identical to the "M" roll, with
the exception that the "O" roll has a crash cymbal, a wood block, and a
separate beater for single stroke snare drum.  These benefits are
minimal, although the crash cymbal is nice.

It would be hard to add that to an "M" roll machine utilizing multiplex-
ing due to the fact that no other instrument plays in nearly the same
musical fashion.  It is used as an accent, and it would be difficult to
find a suitable tracker bar hole to tube it to in an "M" roll machine.
It is really no big loss, however, as it was punched only in the very
latest "O" rolls, and hardly gets used at all.

The exact same applies for the tambourine, and the "high vacuum" ex-
pression in the "O" roll scale.  These too were unfortunately very late
features in the "O" roll, and as a result are not used except in the late
rolls.  The "M" roll does not have the "high vacuum" expression device,
but it does have a nicely used tambourine, a great percussion effect.

As I said above, the "O" roll has two beaters for the snare drum -- one
for reiterating, or "snare roll", and the other for single tap.  The "M"
roll provides only a reiterating beater as in the Seeburg style.  This
difference is really negligible in a well regulated machine, and although
detectable, there is very little difference in actual performance.

If the extra tap percussion effects are important to you, it would be a
simple matter to add them by multiplexing the instrument to accommodate
them.  There are no particularly difficult problems with that.  An "M"
roll can be multiplexed just as easily as an "O" roll, and both machines
can be built up into monsters if you want.  You don't need to have extra
unused holes to multiplex with when you do all of your multiplexing in
the instrument itself.

Multiplexing is better achieved by using conditions set in the given
instrument at a particular time and taking advantage of them, rather than
EVER modifying a roll with a punch.  Like I said above, you can make a
roll do anything with a razor blade.  Make the machine itself do what you
want with the perforations you've got on your roll already.

If you absolutely wanted to, there are three blank holes in the "M" roll
scale that you could punch if you just can't control yourself.  Refer to
the scale for examination.  Two of these are in the holes reserved for
the tune selecting device, and the third is tracker bar hole no. 2,
evidently unused.

Again, I would only do this to a tune or roll which I composed myself, as
I think it is unscrupulous to alter an original or copy of an original
roll.  If you do, please label it as such, so that in the event it is
ever copied, these new perforations can be noted and deleted if not
desired.

I hope this helps to clarify some of the points of the "M" roll, and to
make clearer to some that popular does not necessarily mean better.
There are a lot of machines around, original but more so newer, that play
"O" rolls.  A project that re-cuts the wonderful "M" roll needs to be
embraced wholeheartedly.  True, there are a lot of "O" rolls available
out there, but that is *because they are being cut!*

The exact same thing could be true of any roll of similar quality, I
personally am a bit weary of every new machine playing the same sound.
The variation in the musical arrangements are what keep this interesting
to me.  This is one of the biggest reasons I have always liked the "M"
style.  Superb arrangements, tasteful percussion, as well as a nice
selection of early to late tunes make this an extremely fine choice.

I am looking forward to modern arrangers doing new rolls of modern as
well as traditional coin piano music, as well as seeing newly constructed
machines utilizing the "M" format.  I want to repeat that these are my
own opinions and observations, combined with fact by comparison of data.
I invite anyone to comment on anything I may have reported in error, or
discuss these issues in greater depth.

John D. Rutoskey
Automatic Music Machines
Baltimore, Maryland