source file: m1523.txt
Date: Thu, 03 Sep 1998 12:58:35 +0530

Subject: MIDI  implementation & Yamaha XG

From: Drew Skyfyre <steele@bom2.vsnl.net.in>

Ahoy !

Another note about my (mis)adventures with the Yamaha PSR-530
XG compatible keyboard. 

BTW, you can microtune from the front panel, and set master tuning in 
cents too. Very limited no. of parameters are accessible from there
though.

Last night I suddenly realised that each channel can be microtuned
differently!
Yup, ea. of the 16 channels can hold a different "scale tuning". I
checked.
So, though the method of shifting the pitch of one side of the keyboard
split works for a number of tunings, I don't have to do that, I can just
set up
a different tuning for each side.

Now:
Beware, beware, be very, very, ware of your synth's pitch bend and 
RPN fine tune parameters. It would appear that manufacturers are pretty
nonchalant with their implementation of these parameters. 
No using pretty spreadsheets to generate values.

Following my boo-boo mistaking the +/- 64 scale tuning for 64ths of a
semitone,
and then realising Yamaha had actually documented it correctly as 
actual cents, I decided to check ALL pitch parameters using SysEx and
Hansruedi Baer's Digital Oscilloscope Version 2.1 . I recommend everyone
with
a Mac (PowerMac&68k) get it (& those without go look for one). It's
freeware ! 
An Oscilloscope and Frequency meter rolled into one. Very nice. Not a
toy, but
simple enough for the Xenharmonic/Allotonic neophytes among us,
 like yours truly.

I figured out the SysEx messages for all the Pitch related parameters in
the PSR-530,
there are a total of  8. And most seem to work as advertised in the
manual. 
The multipart detune parameter is (-12.8..+12.7 [Hz]), and being Hz, it
works
logarithmically, i.e., when set to the max +12.7, all keys don't go up
12.7Hz, only
the B key does. The C key goes up @ 6.7Hz.  

Pitch bend, I checked with the range at 1-semitone and 
found no clear pattern in the data values that caused each successive
pitch change.
Graham Breed mentioned he does this and
David Finnamore posted a note cautioning about it, you HAVE to check 
the frequency of the bends against actual cent frequency equivalents.
The 
pitch bend implementation in the PSR appears to be quite coarse. 
The RPN channel Fine Tune parameter is very odd, completely ignoring
the LSB, and going from the equivalent of (-100  to +98 cents). 
So, here again, it it means checking for the desired pitch.

 And the SysEx (SYSTEM) Master Tune parameter (-102.4..+102.3Hz) doesn't
appear to yield any more accuracy than the MIDI Master Tune
(-100..+100cent)
parameter. I was trying to use it to get the pitch of the C key to 256Hz
which means 
going down @ 37.63 Hz, but the best it gets to is @256.25Hz.

BTW, I adopted Joe Monzo's system of tuning the fundamental, and refer
to it as the 
Monzo Fundamental, so if there isn't another name for it, and it's o.k.
with you,
Joseph..... We haven't heard a peep out of you lately, must be the road
trip, 
 you around or too busy burning asphalt ?

Cheers,
Drew