MMD > Archives > May 1998 > 1998.05.06 > 18Prev  Next


MIDI vs. Pneumatic
By Richard Stibbons

Far be it from me to defend MIDI, it's a nightmare, but there are a few
misconceptions which could be cleared up.

John Tuttle raises the prospect of controlling the intensity of every
note individually.  That's exactly what MIDI does, only in MIDI-speak
'loudness' is referred to as 'velocity'.  There's also a perceived
problem with the fact that MIDI uses serial data transmission.

Standard MIDI has a fixed transmission rate of 31.25 kilobaud which is
about 3000 bytes per second.  To play ten notes simultaneously would
require, in a bad case, 40 bytes of data so it would be possible to
play 75 ten notes chord per second over a MIDI link and each note would
be just over one thousandth of a second behind the preceding one.
(That includes specifying an individual velocity for each note.)

That's a good start but it gets much better because virtually all MIDI
devices use buffering to create a mini time machine.  The technique is
to delay all parts of the signal slightly but to delay some parts less
than others.  In this way the timing can be re-aligned.  We are talking
milli-seconds so the fact that the music is a bit late doesn't matter
but it enables the relative timings to be corrected which is far more
important.

One of the joys of using modern technology with piano rolls is that the
required resolution has a clearly defined cut-off determined by the
perforator step rate.  MIDI performs within this by a very wide margin.

Richard Stibbons MIMIT
Cromer, Norfolk
England


(Message sent Wed 6 May 1998, 08:20:57 GMT, from time zone GMT+0100.)

Key Words in Subject:  MIDI, Pneumatic, vs
Enter text below to search the MMD Website with Google

Please Support Publication of the MMD with your Generous Donation
No PayPal account required
SSL Certificate
by
Let's Encrypt