Hi Jane,
I tried to reproduce your error and couldn't. Here's what I did:
1) create a file with the sysex message:
F0 7E 4C 6 2 18 4 4 5 0 32 2E 32 36 F7
2) run pm_test/mm to monitor input
3) run pm_test/sysex
Use send (s) to send the file
4) look at the data printed by pm_test/mm:
F07E4C06 System Exclusive
02180404 System Exclusive
0500322E System Exclusive
3236F7 System Exclusive
I tried over the IAC Driver Bus and through a loop out and back into a
USB MIDISport 2x2, and also sending with 0 latency and non-zero latency.
It is still possible that some other driver is forming packets in a
different way and PortMidi is not interpreting them correctly. Can you
try this loop-back test on your hardware? You could also run pm_test/mm
to receive the Proteus sysex message to make sure that at least 2
programs (yours and pm_test/mm) agree on what data is actually delivered
by PortMidi. I'm running OS 10.4.11 on a MacBook Pro.
I don't know about the 'Service name already exists' message. If anyone
has looked more at the Core Midi internals and has any ideas, please let
us know.
Thanks,
Roger
Jane wrote:
> hello,
>
> im using portmidi to implement a cross-platform (linux/mac/win) sysex
> editor for the proteus 2000 sound module family. im implementing on
> linux and portmidi works great! ")
>
> problem: on the mac i get a different midi response for a hardware
> info request.
>
> on linux i get (correct reply):
>
> >>> midi <<< Prodatum_MIDI::Process_SysEx(15)
> FFFFFFF0 7E 4C 6 2 18 4 4 5 0 32 2E 32 36 FFFFFFF7
> received MIDI standard universal message
> device code/member-code: 516/5, os-rev: 2.26
>
> on mac (osx leopard) i get:
>
> >>> midi <<< Prodatum_MIDI::Process_SysEx(19)
> FFFFFFF0 7E 4C 6 2 18 0 0 4 4 5 0 0 32 2E 0 32 36 FFFFFFF7
> received MIDI standard universal message
> device code/member-code: 0/516, os-rev:
>
> messages side-by-side:
> FFFFFFF0 7E 4C 6 2 18 4 4 5 0 32 2E 32 36 FFFFFFF7
> FFFFFFF0 7E 4C 6 2 18 0 0 4 4 5 0 0 32 2E 0 32 36 FFFFFFF7
>
> as you can see, the length of the replies is different because on the
> mac four "0"-bytes have been added. they are always added at the same
> position.
>
> i tried the latest svn revision (82) on the mac (normally using rev.
> 42 as everyone seem to package it by default for some reason (at least
> debian, ubuntu and macports)? why is that?)
> anyway, the results are the same with both versions.
>
> any ideas?
>
> (
>
> this is not so important but if anyone knows what this means and howto
> fix it, that would be great:
> on mac osx leopard, when i initialize pm, i see this:
>
> >>> midi <<< Prodatum_MIDI::Prodatum_MIDI(0)
> 2008-10-11 04:51:26.077 prodatum[72002:10b] *** CFMessagePort:
> bootstrap_register(): failed 1103 (0x44f) 'Service name already
> exists', port = 0x150b, name = 'Apple MIDI Client'
> See /usr/include/servers/bootstrap_defs.h for the error codes.
>
> )
>
> ty,
> jan
>
> (not subscribed, pls cc)
>
> _______________________________________________
> media_api mailing list
> media_api@create.ucsb.edu
> http://lists.create.ucsb.edu/mailman/listinfo/media_api
>
>
_______________________________________________
media_api mailing list
media_api@create.ucsb.edu
http://lists.create.ucsb.edu/mailman/listinfo/media_api
