re: 22tet Keyboard example

[Clark wrote...]
>For me there is no question that instruments influence the kind of music
>they can produce . . .

Right on!

>I will also note that the design of keyboards for mechanical instruments
>differs greatly from that of electronic ones;

Absolutely!

>even the 7/8 reduction made by the 12-tone DS piano keyboard is afforded
>only by rather an extreme re-engineering of traditional practice, while a
>similar (full) xenharmonic application would necessitate much further
>departure

Depending on what kind of xen tuning you're into, and if you care about
spanning a reasonable portion of the tuning with a single hand.

>I'd venture too much for any manufacturer even to consider anymore, even
>for low tension instruments.

Norman Henry is just wrapping up work on a fortepiano with 33 notes per
2:1, and my hand can reach most of the 2:1's without trouble. Harpsichords
are far easier, as he showed with his 1974 instrument. Then there's
Michael Harrison's idea of using the una corda pedal to access a large
scale 12 notes at a time (again Harpsichords, and especially pipe organs,
would be easier). Never say never!

-Carl

Carl wrote:

> Norman Henry is just wrapping up work on a fortepiano with 33 notes per
> 2:1, and my hand can reach most of the 2:1's without trouble. Harpsichords
> are far easier, as he showed with his 1974 instrument. Then there's
> Michael Harrison's idea of using the una corda pedal to access a large
> scale 12 notes at a time (again Harpsichords, and especially pipe organs,
> would be easier). Never say never!

Whoops - I didn't mean that any acoustic xen keyboard wasn't possible, but
rather instruments resembling Carillo's pianos are less likely nowadays.

My 22-tone keyboard lives in a Flemish-style virginal, a more difficult graft
than harpsichord would have been due its oblique stringing. Each instrument
type will have design constraints, though - piano action parts can't be scaled
down as radically as harpsichord jacks.

I'd love to know more about Norman Henry's instruments: a 33-tone keyboard is
3/2 larger than anything I've attempted! I believe Harrison's idea has
historical precedent in a number of transposing designs, including some with
more than 12 possible notes.

Clark

[Clark wrote...]
>Whoops - I didn't mean that any acoustic xen keyboard wasn't possible, but
>rather instruments resembling Carillo's pianos are less likely nowadays.

You think? Haven't the costs of such a project gone down, like the cost of
everything else?

>My 22-tone keyboard lives in a Flemish-style virginal, a more difficult
>graft than harpsichord would have been due its oblique stringing.

I imagine. Say, did I miss a URL for your keyboard?

>Each instrument type will have design constraints, though - piano action
>parts can't be scaled down as radically as harpsichord jacks.

Yup. There ~is~ the idea of a play-by-wire piano, where scaling isn't
necessary. Imagine calling a normal piano a "tracker" piano. :)

>I'd love to know more about Norman Henry's instruments: a 33-tone keyboard
>is 3/2 larger than anything I've attempted!

It's a double-strung fortepiano with three 2:1's of range. The scale is a
Partchian tonality diamond on the set [1 3 5 7 9 11 15], which is just the
15-limit diamond with ratios of 13 removed. The keyboard is a graphic
representation of the diamond -- just like the recent thread on the diamond
marimba. It goes sort of like this...

/ \ /\
/ n \ / /
/ o / \ / \
/ t \ \ / /
\o / / \
/ \ / /
\ \ \ \
/u / / /
\ t \ \ \
/ o / / /
\ n \ \ /
/ / \ / / ^
\ / ^ \ \ / |
\/ | \ / 8/1
2/1
1/1 4/1

The vertical zig-zag lines are columns of 1/1's. Otonalities go up and to
the right; utonalities down to the left. Please excuse this asci diagram,
which is severely compressed along the x-axis. The real keyboard is spread
out quite a bit. The keys themselves are round, topped with opaque colored
plexiglass. Each key is two-toned, and each identity has a color
associated with it. So if you want to play a 5 utonality, you'd stick to
keys that were green (say) on the top half. Some keys, such as those
within a column of 1/1's, share strings (are mounted on the same lever).
The variation in force and travel distance amongst such keys is yet to be
tested. Henry's 1974 harpsichord (based on the 29-note 11-limit diamond)
is of very similar design, and the lever-sharing turned out not to be much
of a problem. But the greater mass of a piano action may change things. . . .

-Carl

(Carl:)

> Haven't the costs of such a project gone down, like the cost of
> everything else?
>
If anything, I think they've risen. For the Kelley foundry to cast a custom piano
plate (or 10 of them), even if the customer does all of the preliminary work
reportedly would cost around $10,000 six months ago, but Steinway just bought
Kelly & Co. Other options (unpriced so far) are machining large blanks or
fabricating from smaller ones.

New piano actions are very expensive, at least to rebuilders; lumber prices are
rising, while acceptable timber is dwindling.

> did I miss a URL for your keyboard?
>
Nope, a site's still in the works, with all of my hybrids.

> the idea of a play-by-wire piano, where scaling isn't
> necessary.
>
Patrick Ozzard-Low suggests this in his paper, specifically with electronic
linkages between the keys and action(s). Unfortunately the current solenoid
operated player mechanisms have a very limited dynamic range compared with
pneumatics, and while electro-pneumatics have been applied historically they are
even less servicable than regular pneumatics (actually there do seem to be some
promising products in this field). Such a system would need a keyboard acceptable
to performers, with a tracker like Gulbransen's modified to work with a new
pattern - as well as a player action and framing for one or multiple
instruments.$$$

> [In Norman Henry's forte-piano] The variation in force and travel distance amongst such [shared lever] keys is yet to be
> tested.[...]the greater mass of a piano action may change things.
>

Someone familiar with the performance of a Janko keyboard could perhaps give
insight to tier-to-tier variation in touch weight and dip; if only some levers are
shared this might feel rather strange. Presumably Henry's instrument has a
Viennese action so dip and touch weight (and friction) should be very slight,
whereby variations would seem much less than they would with a more modern set-up.

Thanks for the description; I think I read something about the harpsichord, but
I'd love to see either of Norman Henry's instruments in person.

Clark

>If anything, I think they've risen. For the Kelley foundry to cast a
>custom piano plate (or 10 of them), even if the customer does all of
>the preliminary work reportedly would cost around $10,000 six months ago,
>but Steinway just bought Kelly & Co. Other options (unpriced so far) are
>machining large blanks or fabricating from smaller ones.

And what would it have cost, in today's dollars, to have one cast 50, or
100 years ago?

Even if the prices in the piano industry have gone up, things in general
have gone down, I think (based on the fact that the average man-hour
produces so much more today than ever in the past). Which means there
should be a way to get an iron plate for a reasonable price (which $10,000
most certainly is not).

>Unfortunately the current solenoid operated player mechanisms have a very
>limited dynamic range compared with pneumatics

I don't believe that! And even if it is true, it's some sort of mistake on
the part of the people who design player pianos. I find it hard to believe
that there's an inherent shortage in the dynamic range of solenoids.

Pianodisc samples key velocity in 8-bit (128 levels of dynamic range),
according to their rep at Steinway Hall. Also according to him, this gives
greater dynamic range than the old "reproducing" pianos.

>New piano actions are very expensive, at least to rebuilders; lumber prices
>are rising, while acceptable timber is dwindling.

There's no doubt it would be a costly project. But I have it from a
reliable source that pipe organs are still being built in this world, and
my play-by-wire piano would cost less than most of them!

>Presumably Henry's instrument has a Viennese action so dip and touch
weight >(and friction) should be very slight, whereby variations would seem
much
>less than they would with a more modern set-up.

What's a Viennese action? If Henry's is, I doubt he knows it. He designed
and built the action himself. Unlike Steinway, he has had no problems
working with Teflon.

>Thanks for the description; I think I read something about the harpsichord,
>but I'd love to see either of Norman Henry's instruments in person.

The harpsichord was briefly written up in 1/1, in the 1980's. I made
Norman promise to write up the piano when it's finished. In the meantime,
where are you located? If you write me off list, I may be able to put you
in touch with him.

-Carl