Music Trade Review -- © mbsi.org, arcade-museum.com -- digitized with support from namm.org
THE MUSIC TRADE REVIEW
11
^IffWfflgfflffffl^l^lfPTiljra^lf^lg^^
Herewith Is Given Part of a Record of Experiments Which Have Been Con.
ducted During the Past Year to Determine Means of Improving the Expression
Control of Player-Pianos—Some Suggested Improvements in the Construction
The writer has from time to time described
the process of development, and tHe general re-
sults obtained, in various experiments under-
taken, for the purpose of improving the expres-
sion controls of the player-piano without
bringing in either the scheme of operating such
devices through the roll or the employment of a
record made by some pianist. In a word his aim
has been to see what can be done to improve,
within the limits of the ordinary pneumatic
action and of the ordinary straight-cut roll with-
out special side perforations, the personal con-
trol of dynamic and metrical expression.
During the year 1917 a number of experi-
ments have been made on a player-piano chosen
for the purpose, with the result that certain
definite results seem to have been established
firmly. It appears probable that the sum total
of these is sufficiently interesting to warrant
some description in this Player Section.
Whether it is entirely wise to devote oneself
to the development of personal expression, when
the general tide of technical development seems
to be running in quite another direction, must
be left for time to decide. But it will not be
wrong, perhaps, to say briefly why this course
was undertaken.
Why?
The proposition comes down to this. As Mr.
Keeley most properly said in the Holiday num-
ber of this paper three weeks ago, "The most
important thing is to concentrate on the correct
selling of player-pianos by emphasizing t'he
'participation pleasure' they afford. . . . The
average player-piano prospect is thoroughly hu-
man; and he wants to see, above all else, what
he personally is going to be able to do with
a player-piano. He wants to feel confident that
he can play wellwith it."
Now Mr. Keeley has for years been making
a player action built strictly on this personal
participation plan. His associate, Mr. LeCato,
has made nearly 150,000 player actions built
on the same plan; actions without a single au-
tomatic expression feature. That means that
some of the livest manufacturers regard the par-
ticipation pleasure as foremost among the at-
tractions of the player-piano.
Nearly all manufacturers of player-pianos are
now making both the personal and the repro-
ducing type of instrument. It is not necessary
to argue about relative superiority. The two
are wholly different. You cannot match the
musical feats of the reproducing piano, try how
you- will. P)iit, in order to keep up interest in
the reproducing piano, we must also have the
personal-participation player; just as in order to
keep up the interest in concerts we must have
pianos and violins that amateurs may try to
play on as best they may. Moreover, the amount
of high-grade technical work being done on
the reproducing piano is so large that no need
exists, or is likely to exist, for outside advice.
On the other hand, the personal expression
player-piano is, at the moment, the more numer-
ous and, in spite of its musical inferiorities, is
likely to remain the stock commercial player.
Therefore, since these things are so, there seems
to be ample room for experiments tending to
improve it in its musical efficiency.
Conditions
In what follows, it is essential, first of all, to
remember that the idea has been throughout to
take an ordinary player action and, by modify-
ing and changing around its expression mechan-
ism, to improve its musical efficiency. No at-
tempt was made to change the rolls, to put in
side perforations, to work out tempo guides or
to do anything at all that might interfere with
existing inventions or might require extensive
and expensive changes. The governing thought
has been to make use of the material at hand
and show that it can be improved as to results
merely by changing the design of certain parts
and slightly altering the manner of assembling
them.
It should also be observed that nothing is fur-
ther from the writer's thought than to criticize
existing player actions. On the contrary, the
writer has tried to do only what would give him,
in playing, greater personal satisfaction. In
experimenting he has had no notion that his
ideas would suit everybody; but describes them
here simply because the facts disclosed by them
are technically valuable to practical player
builders. The interest is purely technical and
should be so regarded.
The Bellows-Work
When one attempts to analyze the conditions
upon which are based the expressive mechanisms
of a player-piano one finds it easiest to divide
these into two parts, the bellows-work and the
roll driving. All others are subsidiary to these.
The bellows-work includes all that gives speech
to the tones, and the roll driving all that has
to do with the succession and duration of tones.
Thus, within the term "bellows-work" is in-
cluded any and all mechanisms for obtaining ac-
centuation, etc.; for this is merely a matter of
cutting off power from the action at some point,.
and suddenly restoring it again. In all cases,
tie foot-driven player-piano is one which works
vp from zero in power to whatever maximum
may be possible to it. Therefore, accenting al-
ways means one of two things, either kicking a
little harder at the right moment and thus rais-
ing the tension level throughout the whole scale,
or else restoring the ordinary power level that
would be given at any rate of foot work, by
cutting out an artificial pressure reducing valve
that had been artificially kept on the action. In
either case, it amounts to the same thing. There-
fore, all accent devices may rightly be con-
sidered ancillary to the bellows; for all of them
ultimately derive their efficiency from the foot
work.
Now it is plain to all who have played music
with the player-piano that the problem of divid-
ing the right hand from the left hand part of
the piano score is extremely complex. This is
so from a mechanical point of view even when
the roll contains only a simple melody with a
simple accompaniment. But when we come to
think of the possibility of composing for the
player-piano, as Dr. Schaaf has done, in eight
and ten parts, with the composer's score written
on three or even four staves, the problem be-
comes immense. If we do the obvious thing by
dividing the hammer rail or the pneumatic chest
into two parts separately controllable, we take a
step towards that solution; but only a step. For
it is evident that there is no way of causing
a fixed division of the action to coincide with
the melody line of a piece of music, save by mere
chance. No matter how many divisions of the
action or hammer rail might be made, it is evi-
dent that the solution would still be only partial,
for the chance of the upper end of the left-hand
part and the lower end of the right-hand part
(speaking from a piano music viewpoint) touch-
ing in the middle of any such division is very
great, and in fact this is perpetually happening.
A complete remedy is, of course, only to be
found in a complete double action for every
pneumatic; that is to say, by having means
whereby each pneumatic may separately be
played at either one or the other of two separate
power levels. This, however, means, if it be
literally attempted, much trouble and complex-
ity and an entire rebuilding of the action in an
extremely expensive manner. Another and par-
tial solution must suffice.
Again, even if this were possible, it would
still mean a flexible high level for the melody,
subject to the foot control, against a fixed level
for the accompaniment; and music is not played
in this way. Again, even if, by great care and
skill, the accompaniment could be manipulated
up and down from governed to ungoverned pres-
sure, it would only be a change from a fixed
foundation upwards and back again, not to men-
tion the chance of running foul of the melody
and smothering it every now and then.
Can we then, without changing the ordinary
system of expression control, get an accompani-
ment subordinated to the melody and under
instantaneous control, while at the same time it
is flexible and capable of being raised or lowered
rapidly if not instantaneously? Can we, at the
same time, find means, without making elaborate
changes, of obtaining a high level of power
under pedal control for melodic and accented
work, which will at least be tolerable and will
avoid the very great majority of the cases of
interference between melody and accompani-
ment which disfigure all playing by ordinary
means? If we can do this much, we can virtu-
ally, and to all intents and purposes, solve the
expression problem of the foot-driven player-
piano, for the present at least, and until the con-
struction of said player-piano has been much
modified in the direction of greater expressive-
ness.
In point of fact we can do all this, and do it
very simply without making any change of great
moment or introducing any expensive novelties.
Changes
In the first place it is necessary to do away
with reliance on the hammer-rail control. This
had better be reduced to the simple plan of a
pneumatic adapted to raise the entire hammer
rail, just as the soft pedal does. This may be
controlled by a button, in order that the hands,
which will be otherwise busy, may easily touch
it when necessary.
Then there should be a divided pneumatic
chest, with an expression governor for each half
thereof. This governor will contain a pouch
and valve, the rising of which adjusts the passage
ways so that the air must pass through the re-
ducing valve of the governor, and thus be, main-
tained at a pressure level nearer the atmosphere
than will be the case on normal pumping when
the air passes direct from action to bellows.
When the pouch falls, the air will pass through
direct.
Controllable Springs
This much any ordinary player may have. But
at this point the difference begins to be seen.
The worst feature of the ordinary expression
governor is found in its rigidity. The level of
power—that is to say, the working pressure on
the action—which can be had when the governor
is in operation is determined by the tension of
the spring. If this spring be set at a given ten-
sion—as is the common practice—then the gov-
erned or soft pressure is always the same. This
gives a very unmusical result, for reasons ex-
plained above. But it is very easy indeed to
arrange that the tension on the springs can be
(Continued on page 12)