Disklavier

How do you go about changing midi files to direct pedaling to channel 
3 (for continuous pedaling)?  I assume you use Cakewalk to do this.

Thanks.

--- In disklavier@y..., "Robert E. Welcyng" <rwelcyng@a...> wrote:
> I have used a DAT and Sound Forge to measure the timing accuracy of 
my Mark II. 
> At note velocities over 40, the timing error 90% of the time is 
less than a
> millisecond for single notes played in succession.  Statistically 
speaking, I
> found no difference in timing accuracy between MIDI from a 
sequencer and E-SEQ
> in the Mark II drive.  With older slower computers there can be 
noticeable
> timing degradation of MIDI under Windows if you give Windows other 
jobs
> concurrently.
> 
> Driving the DKV through the MIDI input does have another catch that 
is not a
> problem with E-SEQ: You should take note of the pedaling.  If the 
MIDI file has
> continuous pedaling directed to channel 3, you will want to set the 
DKV's MIDI
> IN for "CH=HP" so that the DKV responds to the continuous pedal.  
If only on/off
> pedaling is present and directed to, say, channel 1, then you will 
want to set
> the DKV for MIDI IN as "CH=01".  Of the PianoSoft disks published, 
some use one
> scheme and some use the other.  (I've modified my files so that 
there is always
> pedaling directed to channel 3.)
> 
> Putting all one's files on a CD is not going to be everyone's cup 
of tea.  It's
> still a lot of work and the operation is not "iron ball" simple as 
with a
> diskette.  Yet even operating with diskettes is not fool-proof--
insert the
> diskette upside down and, I understand, you will destroy the floppy 
drive.
> 
> "Mark A. Fontana" wrote:
> > 
> > I wonder if there might be a technical disadvantage to the 
approach of
> > driving the Disklavier via MIDI from a computer.  Namely:
> > 
> > Suppose a piece contains a chord of three notes struck 
simultaneously.
> > This can be represented in ESEQ or MIDI format as three note-on 
events with
> > delta times of 0 between the second and third events.  The 
Disklavier's
> > embedded controller can process a string of events like this as 
quickly as
> > possible, ensuring all three notes strike at the same time (or as 
close to
> > it as possible).
> > 
> > Whereas if you send MIDI data into the Disklavier, there is no 
notion of
> > delta time; notes are played as they are received (actually 500 
ms after
> > being received, so that the Disklavier can compensate for varying 
solenoid
> > strike times at different velocity levels).  As MIDI is a serial 
protocol
> > at 31250 bits/sec, 10 bits/byte (8 + start bit + stop bit), a 
certain
> > amount of time is required to transmit each byte.
> > 
> > With notes being played as they're received, it's impossible to 
specify
> > that multiple notes be struck simultaneously.  Using running 
status, a
> > three-note chord would require a minimum of almost 19 ms to 
transmit via
> > MIDI, or about 28 ms without.  A larger chord with note-offs and 
pedal data
> > mixed in could slow things much more.
> > 
> > Has anyone been able to tell a difference between playing the 
same material
> > using the console and from a computer?
> > 
> > On the other hand, it could be that the convenience of storing 
one's music
> > library on the computer outweighs any concerns about timing 
accuracy.
> > 
> > On Thu, 28 Jun 2001, Robert E. Welcyng wrote:
> > 
> > > You've done a service to those with an affection for more of 
those pesky jewel
> > > cases.
> > >
> > > As an alternative, some might wish to convert their PianoSoft 
material to MIDI
> > > and put the files on a CD-ROM.  Let's say you devote 600 MB of 
a CD-ROM to
> > > data.  That would allow you to store at least 800 each 720 KB 
albums on a single
> > > CD.  Of course, you'd put the files in appropriately labeled 
folders and name
> > > the files descriptively.  You'd devote a folder in the CD to 
playlists and use a
> > > sequencer to play your DKV.  Rather than search and shuffle 
through jewel cases,
> > > you use your Windows "Find" facility to locate files by 
composer, title, or
> > > performer.  You don't have to bother about returning disks and 
cases to their
> > > proper places.