From:
                                                            1/12/97 12:00 AM
Subject: MAX Digest - 10 Jan 1997 to 11 Jan
1997To: Recipients of MAX digests 

There are 2 messages totalling 55 lines in this issue.

Topics of the day:

  1. sysex question (2)

----------------------------------------------------------------------

Date:    Sat, 11 Jan 1997 09:25:41 -0500
From:    Jim Croson 
Subject: sysex question

Greetings.
In my patch, I have a sysex message out (fine) and receive sysex back
(fine, again). But I can't see a way to make the returned sysex info
enter BEFORE all other pending processes finish. For instance, a bang
box starts the sysex exchange and sysexin goes to a capture box. The same
bang fires a dump message to the capture box. But the dump message (and
whatever else might come off the original bang) always comes before
the sysex info comes in.

Can anyone help?
Thanks in advance,
Jim
croson@theory.esm.rochester.edu

------------------------------

Date:    Sat, 11 Jan 1997 16:55:00 -0500
From:    Stephen Kay <70714.1356@COMPUSERVE.COM>
Subject: sysex question

>In my patch, I have a sysex message out (fine) and receive sysex back
>(fine, again). But I can't see a way to make the returned sysex info
>enter BEFORE all other pending processes finish. For instance, a bang
>box starts the sysex exchange and sysexin goes to a capture box. The same
>bang fires a dump message to the capture box. But the dump message (and
>whatever else might come off the original bang) always comes before
>the sysex info comes in.
>jim croson

If I'm understanding you correctly, there are several things you can do.
The problem here is apparently the amount of time it takes to send/receive
the sysex message.

First, you could try connecting the bang to a trigger "t b b", and make
sure the right most outlet goes to the send part, and the left one goes to
the dump connected to the capture. This may or may not solve the problem.

Secondly, you could use the returning sysex info itself to trigger the
dump, through a trigger "t b l", where first you run the returning sysex
into a "thresh 50" (or whatever time is deemed sufficient) so that it
becomes a single list, then connect the right out of the trigger (list) to
the capture, and the left outlet (bang) to a dump connected to the capture.

Third, you can resort to delays.  Connect the original bang straight to the
sending section, but insert a "del 1000" (or whatever time is deemed
sufficient) between it and the dump and the capture.

Stephen Kay

------------------------------

End of MAX Digest - 10 Jan 1997 to 11 Jan 1997
**********************************************