I'd like to speak to the "disappointment" in the DKV's apparent limited
effective MIDI note velocity range. I fully concur with PianoBench's
suggestion of concentrating on listening to evaluate the DKV's technical
performance and not getting hung up on numbers. However, some of you
might be interested in a different angle on the matter.
Some time ago, I made sound output measurements on my DC3 Mark II. I
wrote MIDI files to play a succession of the same single note throughout
the velocity range of 1 to 127. I recorded that succession of sounds to
DAT and measured the initial power of each note using Sound Forge. In
plotting the relative sound power of the notes (in decibels) versus note
velocity, I found a fairly straight line between velocity 25 (-20 dB
relative sound power) and velocity 90 (0 dB relative sound power). The
slope of that line is then 20 dB per 65 velocity units, or, in other
words, 0.3 dB per one unit of velocity.
Now, what do you suppose is the minimum step-change in sound power that
an average person listening to a continuous sine wave can detect? Well,
that depends upon the frequency of the sine wave and the original
loudness. A typical figure (for 1000 Hz and medium loudness) is 0.5 dB.
(See John R. Pierce, THE SCIENCE OF MUSICAL SOUND, PAGE 131).
So, generally speaking, even though the DKV's effective MIDI range (with
the DKV Volume control set at "0") may be only 25 thru 90 (or 30 thru
100), its output sound intensity range is divided into steps
sufficiently fine that most humans could not detect the change from one
velocity step to the next.
Was this design simply fortuitous, or was it brilliance on the part of
the engineers at Hamamatsu? I believe the later.
My question is: Beyond the personal satisfaction of owning a Pro, does
recording and reproducing note-on velocity to a precision of 1000 steps
really buy you anything? I'd love to make some measurements on a Pro
and find out.
>
>> <<My DU1A actually plays from a minimum of 30 up to something like 95
>> or 100 velocity. Very disappointing for somebody who wants a 1-127
>> dynamic range ! PianoBench gives here a very sensible answer.>>
>
>
> In the hope that you might not be perpetually disappointed, I would
> suggest that you compare the effective velocity range of your Disklavier
> to that of another electronic keyboard in two ways:
>
> (1) Record a piano piece on the Disklavier and then record the same
> piece on an electronic keyboard of your choice. Then, play back both
> pieces. Ignoring what knowledge you have of the MIDI velocity numbers,
> ask yourself which plays back more sensitively and accurately and which
> is more pleasant to listen to. My point is simply that the ultimate
> issue is how good the audible results are.
>
> (2) Record a piano piece with really wide ranging dynamics--such as
> something by Chopin, Liszt, Rachmaninoff, or Prokofiev--on both your
> Disklavier and on an electronic keyboard of your choice. Then, examine
> the MIDI data from both recordings. Did the electronic keyboard produce
> a significantly wider range of MIDI velocities? Try the same test using
> a piece with more modest dynamic variation. How do the note-on
> velocities compare?
>
> I think that the results of test #2 will be very interesting. They will
> depend on the characteristics of the particular electronic keyboard, of
> course, the nature and quality of the digital piano sample that you hear
> when you play, as well as the volume level that you set for the audio
> output of the keyboard.
>
> The last issue--the volume level of the electronic keyboard--will have a
> profound impact on the results. If you set it very high, you will have a
> tendency to hold back when you play loudly because the resulting sound
> will too easily get harsh and ugly. The result may be that you don't
> achieve the highest possible MIDI velocities. If you set the volume
> really low, you may similarly not achieve very low velocities in your
> playing because the resulting sound would be too soft.
>
> My point is simply that the issue of note-on velocity gets to be rather
> artificial and arbitrary in the world of electronic keyboards.
>
>> <<Now I guess the "Pro" models have been designed to ameliorate the
>> lowest velocities, those when you finger will go "climb backward" down
>> the key, then just at the end give that little velocity that will just
>> caress the string. There's no other way for a disklavier to do that
>> and it is : controlling the hammer and key movement all the way during
>> its travel tawards the string. Remember when you have a key half
>> depressed in a Chopin Nocturne, and the just feed the little energy
>> that will have the hammer just "humming" at the end of a phrase ?
>> This is something exquisite and normal, but Midi cannot take care of,
>> if only because it was designed as a physical model of movement :
>> you're getting the velocity plus the exact moment you depress the key
>> (Midi keyboards are sophisticated but basically an off-on system
>> plus a velocity the weight of the key when striking the bottom of the
>> keyboard bottom) The movement of even a child's first attempt to play
>> the piano is a much more sophisticated one !>>
>
>
> Actually, the Disklavier Pro is more sophisticated in this regard than
> one might think. Its servo playback system--combined with key release
> velocity data results in key movement on playback that is very close to
> the original performance. And, of course, the speed of the hammer
> hitting the strings during playback is quite accurate.
>
>> <<I really would like to be able to afford a pro system, but I guess
>> it will still not be able to capture life's movements. Anyway, when I
>> hear Rubinstein pays "l'amor Brujo" on a 1920 piano roll, on my DU1A,
>> I shiver and I think I get his very touch on a very deep and moving
>> way (check it on those scanned rolls that Terry Smythe so gently
>> offered us ! http://members.shaw.ca/smythe/library.htm ) . I do not
>> know where that comes from. Maybe it's my imagination ? The Yamaha's
>> technology plus my imagination have been filling the gap, and that's
>> good news !>>
>
>
> Back in the days of the pneumatic (i.e. vacuum operated) player pianos,
> there were a variety of playback systems. I am not sure which system
> Rubinstein recorded on. I can tell you, however, that the recording
> process did not capture his dynamics. The recording process captured the
> note one/off information and pedal on/off information. Any dynamics that
> were encoded onto the roll were put there after the fact by the roll
> editors in an attempt to capture the spirit of his playing.
>
> The roll editors used a number of ingenious techniques and did a
> remarkable job. Typically, the recording artist had to sign off on the
> results before the roll was release, so we can think of the results as
> being close to the original performance.
>
> I have not had a chance to look at Mr. Smythe's converted roll data in
> detail. It will be interesting to see what range of note-on velocities
> are in these files. The Duo-Art system, for example, at 16 "power
> levels" of encoding for "accompaniment notes" and 16 power levels of
> encoding for "theme notes." You might think that this would result in 32
> possible hammer velocities, but that is not the case. Some keys have
> more mass to move than others. Also, the amount of energy available to
> move the keys at any power level gets distributed differently depending
> upon whether the dampers are up and depending upon how many notes are
> being played at the moment. The bottom line is that 32 power levels were
> capable of producing an amazing spectrum of actual hammer velocities.
>
>> <<I tend to think that this is the way we should consider any musical
>> instrume, be it a mechanical one : This outstanding piece of its
>> century technology (a Stradivarius for example, or a Yamaha DKV) plus
>> our imaginations.>>
>
>
> Good way of thinking about it. When I attend a concert, I notice that a
> lot of the audience's enjoyment of the performance goes well beyond the
> actual note-on velocities that produce the sounds that they hear.
> People's enjoyment has a lot to do with their preconceptions about the
> artist, the artist's personality, the visual trappings of the venue, the
> choice of pieces, the performer's gestures, et cetera.
>
>>
>> <<Music has always had some magic in it. The technology will never be
>> able to capture it all. So when it suddenly comes out of a mechanical
>> pianoplayer, however sophisticated, it must come out of somewhere
>> else. I tend to think it comes from heaven.>>
>
>
> We are getting closer all the time. I have recently been involved in
> recording contestants in the screening round of the International
> Piano-e-Competition. Contestants were recorded on Disklavier Pro pianos
> in Hamamatsu (Japan), LA, New York, and Paris. We recorded them using XP
> data and recorded them visually with a digital video camera (capturing
> the video directly on DVD). During the recording process, we used SMPTE
> and MTC (two types of time code: Society of Motion Pictures and
> Television Engineers and MIDI Time Code) to coordinate the recording
> devices.
>
> Next month, the 1st round jury will assemble in St. Paul, MN and watch
> the video on a large screen and listen to the contestants' MIDI data
> played back on a 9-foot Disklavier Pro in a 350 seat recital hall. This
> is an attempt to create an experience for the judges that is close to a
> live performance. The judges will then reduce the number of contestants
> from about 55 down to 24 who will compete in person in St. Paul in the
> next round.
>
> Regards,
> PianoBench
>
>
>
>
> To Post a message to the group, send it to: disklavier@...
>
> To Post a private message to Todd Muncy, the group's founder and
> moderator, send it to:
> disklavier-owner@...
>
> To reach our group's web site go to:
> http://Yahoogroups.com/group/disklavier
>
> Todd's family web site was completely updated 01/15/03. It contains
> some fun disklavier content and links to midi sites among many other
> things, The url is:
> http://MuncyFamily.com
>
> THINKING OF LEAVING THE GROUP?
> If you are thinking of unsubcribing because you are getting too much
> mail, go the the web site and change your email delivery option
> instead. That will fix the problem, while maintaining your access to
> the group. If you insist on leaving us completely send a blank email to:
> disklavier-unsubscribe@...
>
> Know someone who wants to join? Have them send a blank email to:
> disklavier-subscribe@... or give them this link:
> http://Yahoogroups.com/group/disklavier/join
>
>
>
> ------------------------------------------------------------------------
> Yahoo! Groups Links
>
> * To visit your group on the web, go to:
> http://groups.yahoo.com/group/disklavier/
>
> * To unsubscribe from this group, send an email to:
> disklavier-unsubscribe@yahoogroups.com
> <mailto:disklavier-unsubscribe@yahoogroups.com?subject=Unsubscribe>
>
> * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
> Service <http://docs.yahoo.com/info/terms/>.
>
>
--
Robert Welcyng
Anchorage, Alaska