Disklavier

From: Walter Sharpe

Thank you Fred, Spencer and Piano Bench,
     I will respond to your various comments (copied below) all at once rather
than separately to make it easier for you all to follow. Maybe some others out
there will benefit from some of the information here.
     Since my response is rather lengthy, I have included it in an attached zip
file along with some midi files that demonstrate some of my points. They sound
very nice too so maybe those who are not that interested in the discussion will
at least enjoy the music!
-Walt

***********

W: Also contained in the zip file is a file called "PEDAL THUMP TEST.mid". It
contains nothing but pedal events so you can listen to the pedal action and any
rumble of vibration pedal thump sets off in the strings. I created it with
Cakewalk Home Studio. When you insert a pedal event in Staff view, Cakewalk
writes in a 127 and a 0. (In other words, a down and an up.) On Track one, by
going into the "Event List" view and changing the numbers, I was able to
demonstate that my two year old MX500 Disklavier II interprets Controller 64
values between 85 and 127 the same: pedal down with a thump. It treats values
between 80 and 0 the same also: pedal up with another thump. It's going to be
interesting to put this file on a floppy disk and try it on a few different
Disklaviers on the show room floor.

D: Hi Walter

D: I've been following this thread with interest as I also have been
experimenting with the 'Pedal Thump' problem for some time. I am using a Mark II
XG Disklavier. When I change my controller 64 information so that I have a
'linear' type of pedaling, my sustain pedal follows in a somewhat linear
fashion, and not responding in a hard 'on/off' fashion. Have you tried
'modulating' the controller 64 data, i.e., a series of 'timed' 'on/offs'
gradually moving them apart in time? (A bit like PWM, pulse-width-modulation, in
Engineering terms. I can get a bit more detailed if you need this info.) A
mechanical motion takes time to respond from a stimulus. This *may* 'damp' the
physical motion enough to produce a softer 'on/off' pedaling. (Of course, we
don't know what Yamaha is doing internally when reading this MIDI data; i.e.,
how it responds to varying controller data in real time.) It sure would be nice
if we could figure out how different Disklavier models respond to the same
pedaling data. We could put this info in a .PDF file similar to what Carol added
in our 'files' section a while back showing differences between Disklavier
models.

D: A also tried your "PEDAL THUMP TEST.mid" file on my Mark II XG. As shown in
the first three measures of this file, you have an alternating 80 and 85, then
you go to the alternating 0 and 127. My pedal goes down a small bit and stays
there until measure 4.

D: I am attaching basically the same file with some controller 64 mods. When
playing this MIDI file on my Disklavier, the thump quietens the most with a
release value of 50. I can't hear much of a difference with the release values
of 20 or 30. I was trying the 'on' values of 127 and 100, just for
experimenting. The only time I hear the hard thump on my piano, is the releasing
of the sustain pedal. How about yours?

D: Danny