I don't mind an animated technical discussion in which two rebuilders
chat about their different viewpoints of a particular system.  It's
good and healthy.  The problem comes in when one or both are not
reading what the other says, but launches off on a tangent, refuting
nothing and tilting windmills.  I think we should watch that tendency
a little more closely.  The other thing I believe is important is to
get all the facts before accusing the other of not knowing what he is
talking about.  I am going to quote Julian's central theme statement,
and we're going to then see whose story holds up -- mine, or his.

> First, Craig states that it's necessary to have a significant suction
> difference between Theme and Accompaniment levels of the same coding
> number to make the theme stand out in rolls.  That's obviously not
> true: the Theme uses higher dynamic coding to make it stand out!

Julian, if you would only read your Aeolian manual first and let them
answer you, I wouldn't have to.  But you have challenged me publicly to
put up or shut up as it were, so I'm going to now put up.  And I would
also remind you that there are more genteel ways of stating one's case,
and not getting it back in your teeth so hard.  You might practice
that, instead of having everybody who reads this, smiling sweetly as
they probably sense what is about to happen.

Aeolian: "As was stated above, the theme zero setting is one degree
louder than the accompaniment."

What does this simple sentence portend, to those pianos whose
accompaniment and theme springs are identical? (meaning equal rate
springs).

Identical springs make the theme and accompaniment characteristics
identical.  So by raising the theme curve the prerequisite 1/4"
to 1/2", you _transpose_ the theme line, which is identical to the
accompaniment response, about 1/2" above the accompaniment.
(Transposition is different than rotation referring to graphs, agreed?)

Now we know already that a Duo-Art grand is designed to reach  about
40" of vacuum dynamically when playing a loud passage, right?  (Yes,
I know the pump will produce more with a non-slip chain drive.  But
let's be sensible.  40" of vacuum is the maximum vacuum needed to play
parlor grands.)  So I would ask you then to get out your grid paper and
jot down the 15 steps plus zero level on the horizontal bottom line,
and 0 to 40 going up the vertical left edge of the paper.  I want you
to see how unbelievably impossible it will be to achieve the capacity
you are claiming, with any line or curve one could conjure.

Now yes -- we can all connect the dots in some fashion, but that's not
the point.  Unfortunately, you will have to remain within the capacity
of the design limits of the box, not of your imagination.  And I will
tell you right now -- such a curve in an American box would be
impossible.

And then once you have invented it, I want you to sit down and justify
it to all of us, and then I will shut up when I see that you obviously
have something.  But first, you must be able to explain to those
interested how an inherently linear box (given a few wiggles and
aberrations, granted) can clearly produce the curve the British require,
and also why such a curve is superior to the American Duo-Art.

We all know that, for all reproducers and player pianos alike which
use a rotary pump, that between 18-20" vacuum produces the threshold of
single forte' and is called "nominal" pressure, meaning un-amplified,
spilled, or otherwise used as reroll pressure.  All Duo-Arts, in order
to produce their first forte, require the knife spill to be closed.
This is the power they call 10.  That means, everything from zero
through 9 plays between the vacuum pressures of 5" (zero) and 18"-20".
Surely, we agree to that much! I do not speak of the variations between
pianos here, only some general figures for the purpose of explaining
this very simple concept.

So on your graph paper make a dot at step zero and 5" pressure.  At
the 10th step across, go up and make another dot at, say, 20" of vacuum
pressure.  Draw a straight line between them and extend that line on up
to the edge of the paper and the 15th step.  What pressure did it land
on? Maybe 27"? Fine.

Now then, go back to step zero and 1/2" of vacuum higher, make a dot.
At the 10th step and exactly 1/2" higher, make another dot.  Draw a
line between then and extend the line to the far right edge of your
paper.  What pressure did you reach at the 15th step, Julian? What's
that?  27-1/2"? That's correct.  And if that is how British Duo-Arts
play more powerfully than American models, then undoubtedly I take my
hat off to you.  You are a genius.  Somehow, you are achieving 40" with
a system capable of only 27-1/2."

Now, there is only one way that you are able to get around the impasse'
you have made for yourself.  You will have to show that actually, the
British Duo-Art theme response curve slews upward smoothly from 5-1/4"
through the threshold of forte' (which is about 18-20", take you
choice), and then quickly upward to 40."

I can do that in general, and show about what you or anybody else would
have to see with a geometric construction, given the least amount of
curve at the low end, that's possible in order to still make it to 40"
while bisecting 18" - 20" at step 10 -- a requirement of all Duo-Art
music.  One likely set of normal values are shown for comparison in
brackets at the right of each of your numbers:

  0 = 5-1/4"    (5-1/4")
  1 = 5-3/4"    (6-5/8")
  2 = 6-1/4"    (8-/14")
  3 = 7-1/4"    (10")
  4 = 8-1/2"    (11-1/2")
  5 = 9-5/8"    (-13")
  6 = 11-1/3"    (14-1/4")

As you may guess, other curves are hypothetically possible, but they
won't vary off of this one more than 10-15% in this short a distance,
and 15% is not enough difference to create the curve you claim  reaches
the full power of a rotary pump--i.e., a British piano.  Mathematically,
you are stuck.  So either you admit to a basic linearity of design
within the box, or you admit to low intensity step changes that cannot
be heard by normal people, and cannot even be resolved by normal
Duo-Art boxes.

Besides rescaling the accordions, redesigning the spill, and
overstretching the springs, I don't know how else you are able to
generate the chart you claim.  The box is designed by the factory to
be reasonably linear -- within practical limits.  Were its response to
curve rapidly upward as you and Paddy both claim, no D/A arranger would
be able to add an intensity step to large chords beyond step 10 and
get only a single degree louder to compensate.  Below step 10, he would
have to add 2 and perhaps 3 or 4 additional steps to compensate for
many notes because of the insufficient changes being made
proportionally, according to a curved response generated by these 3
fixed theme points -- 5-1/4", 18," and 40".

Here's what Aeolian said about  shortening (hence, overstretching) the
springs -- which will create the curve you are claiming here to a
degree (but then it tops out and your last 3 steps are about the same):

Aeolian:  "It is also necessary that there be just enough tension...
Likewise the tension on the regulators should not be drawn down...
but just enough that they do not rattle (in other words, their entire
length is designed to be used to get that linear response).  Pulling
down the tension on the regulator springs is frequently done because
of lack of knowledge.  When this is done it upsets the zero tension of
the regulators... carried to the extreme, it will ruin the springs for
the fine setting of the accompaniment or the theme and it will be
impossible to bring down the tone to that softness so much desired by
real music lovers.  If the springs have been ruined (overstretched),
the only remedy is new springs."

As if this were not enough (and I do not get into adjectivizing
performance when facts fail me) most Aeolian Duo-Arts made in the
United States came with a theme and accompaniment spring installed.
The rate of one was double the rate of the other.  My argument
therefore is not hypothetical, but chiseled in stone and backed up
completely by Aeolian.  It is not a private opinion.

Were you to chart the effect of two different springs like these,
you conclude that Aeolian already knew that merely translating the
theme line upward on the chart, parallel to the accompaniment line and
separated by only a fraction of an inch, was not sufficient to create
a realistic theme, as their true theme intensities required a different
slope.  This fact, of and by itself, dissolves Julian's argument
entirely.

Contrary to Julian, it happens all the time in rolls that the same
theme step number is called for to displace the same number step in
the accompaniment.  That fact, too, is "in stone," so to speak.  In
order for Julian's piano to pick out the theme pressure _at the same
step number_ at, say 20" when only a 1/2" difference exists between
them, the expression box would have to resolve in an instant a 2.5%
change in vacuum.  That is impossible.  The box cannot do it, no one
can hear it.  I do not stand corrected on this point either.
Physically, it cannot resolve such tiny increments.

In America, when a customer complained that his piano was playing
too loud, the store he bought it from would come out and install
another accompaniment spring and toss the theme spring.  It was
standard practice (some old players still have the theme spring tied
to a hose with twill tape).  Those customers wanted their piano as a
background instrument to play through dinner or something.  They didn't
really want a concert performance, and it was still very nice.  But, it
was no longer a fully dynamic Duo-Art.  Get the picture?  They wanted
background music -- sensitively ravishing background music.

In a properly set-up Aeolian Duo-Art, the reason you actually do see
a theme equal to the accompaniment many many times, is because at step
10, where we expect to see about 20" of accompaniment pressure, the
same theme pressure at step 10 is about 29." That means when we are
playing along at accompaniment pressure of 20" and the theme accents
a note at step 7, an American Duo-Art would raise the pressure of the
accent to about 21-3/4" (a bolster, or reinforcement instead of an
actual accent), while the British Duo-Art would have dropped to around
13."  That's almost half-way (dynamically speaking), using the only
possible other scenario curve that you can claim! Our only conclusion
then is, if Julian is right, that British Duo-Arts not only have
superior expression boxes, but will reinterpret every American Duo-Art
roll so much differently that one would not recognize the artist.
Strange, seeing as how the artist signed both releases.

Frankly, I strongly doubt that all British Duo-Arts are salon
instruments as implied.  I think they are good instruments and capable
of a very realistic performance, so I don't believe it, Julian.  I'd
say the ball is firmly in your court because you and Paddy have made
the claims that refute what Aeolian actually did.  Remember, we have
their original equipment to see if what you say is correct, and it
isn't a matter of artistic interpretation but physical numbers,
measurements, and the laws of physics.

It is now incumbent upon you both to prove that American Aeolian
Duo-Arts were wrongly designed and that they not only didn't know
much about music, but proved it by putting springs on their boxes
having rates differing by a factor of 2 when they should have been
equal, and that they were wrong to design a basically linear box
(easily shown), their music is offensive with that silly, supercilious
crash valve, and that Duo-Art rolls never used a theme expression equal
to the accompaniment, but always raised or lowered it.  "It's your
serve."

What else can we can say  to someone who simply ignores the actual
working device except, "prove it." You both owe us that.  You've stuck
your necks out.  Now, back it all up, or just tell us that perhaps you
should reconsider.  Don't worry -- that happens to everybody and Brits
are not exempt.  I do not pretend to tell you about your Duo-Art
systems, but you insist that ours could not possibly work
realistically, as originally designed.  However, we are all ears.

Craig Brougher