1 Nov
2009
1 Nov
'09
10:14 p.m.
On Mon, Nov 2, 2009 at 12:20 PM, Dan Mills <dmills(a)exponent.myzen.co.uk> wrote:
On Mon, 2009-11-02 at 10:52 +1100, David wrote:
Yes you are correct, I made a factor of 10 error.
The copy of the MIDI 1.0 spec I looked at says "the interface operates
at 31.25 kbaud" so I interpreted this as 31.25 10-bit symbols per
second, which is one possible interpretation of baud, but this is not
what they meant. The slippery old baud claims another victim. They
meant "the interface operates at 31.25 kbit/s", so I made a mistake
and thank you for clearing it up !!
I broadly agree with your rule of thumb, and it is a useful addition
to my attempt to give a better understanding of what issues are
involved here, and it makes sense to add it.
But nothing beats a test once you know what all the issues are, which
was the point I was attempting to make.
Also it would probably not hurt to improve the midi output's EMI
rejection with some ferrite beads and bypass caps if there's going to
be a longwire antenna hanging off it, even if it is made of twisted
pair.
Cheers
David
Another thing about the that MIDI spec is where
it says "optoisolators
... rise and fall times should be less than 2 microseconds" which is
amusing because the total time of one midi bit is 3.2 microseconds. So
don't imagine you have nice square bits driving the system even if
your cable is zero length. So your optoisolators *might* be a limiting
factor depending on their speed (ie age, cost).
I think you make a factor of 10 error! One midi bit is 1/31250 = 32
microseconds, not 3.2 which makes a difference.