Dear Sirs,
I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have:
- sysex gets fragmented (also when nothing is sent out meanwhile by other
objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
- sysex just magically does not work on the target machine ;o) I have no
clue why except for this thing with being truncated but it *never* works even if the message seems to go out intact. Even more interesting that the machine shows me that it is recieving sysex! Tried with the mpu-401 of an Audigy and with a MOTU 828 mkII through Jack and a2jmidid (jack-alsa midi bridge). Sysex is
Thanks a Lot!
I haven't actually tested sysex out limits, but sysex in is limited to something like 24 byte messages..
here is a thread where i talk about it: http://lists.puredata.info/pipermail/pd-list/2009-01/067372.html
It is hard coded into the source. You can modify the source to change the hard coded limit.. i have a diff on that thread which shows how i did it.
It would be nice if this limit was either totally removed or dynamically growable.. as it would be really nice to be able to write real sysex editors in PD.
I have been creating my own physical devices with sysex and I use pd to edit the sysex data but i have to make my sysex packets short because of PD's limitations.
-Alex
2009/7/7 András Murányi muranyia@gmail.com:
Dear Sirs,
I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have:
- sysex gets fragmented (also when nothing is sent out meanwhile by other
objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
- sysex just magically does not work on the target machine ;o) I have no
clue why except for this thing with being truncated but it *never* works even if the message seems to go out intact. Even more interesting that the machine shows me that it is recieving sysex! Tried with the mpu-401 of an Audigy and with a MOTU 828 mkII through Jack and a2jmidid (jack-alsa midi bridge). Sysex is
Thanks a Lot!
-- Muranyi Andras
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
2009/7/8 Alex x37v.alex@gmail.com
- sysex gets fragmented (also when nothing is sent out meanwhile by other
objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best
recommendation
knowing that my patch is full of sysex outs (+envelopes on some of them!)
- sysex just magically does not work on the target machine ;o) I have no
clue why except for this thing with being truncated but it *never* works even if the message seems to go out intact. Even more interesting that
the
machine shows me that it is recieving sysex! Tried with the mpu-401 of an Audigy and with a MOTU 828 mkII through Jack and a2jmidid (jack-alsa midi bridge).
I haven't actually tested sysex out limits, but sysex in is limited to
something like 24 byte messages..
here is a thread where i talk about it: http://lists.puredata.info/pipermail/pd-list/2009-01/067372.html
It is hard coded into the source. You can modify the source to change the hard coded limit.. i have a diff on that thread which shows how i did it.
Thanks Alex, it's an important thing to know. My sysex is shorter than 24B it's just broken. What we shall see repeating is 240 67 16 73 8 16 19 (knobvalue) 247
Instead I'm having this grinded mess coming back on physical loopback:
print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 8 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 8 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247
This makes me think I'm trying to send too much data in too short intervals...? Is there a handy way to limit the density of my output? (i think its not not [drip] as it has to stay realtime) I'm not very experienced, sorry if this traces back to something very obvious!
It looks like you may experiencing another thing that i've complained about... pd's alsa midi connects to all MIDI inputs and outputs that it can.. this includes alsa's MIDI through.. so everything that you send out gets echoed back into the input.. SUPER annoying.
run aconnect [or easier, though not great, aconnectgui] and see if that is the case for you.. disconnect pd from the alsa through port.
-Alex
2009/7/7 András Murányi muranyia@gmail.com:
2009/7/8 Alex x37v.alex@gmail.com
- sysex gets fragmented (also when nothing is sent out meanwhile by
other objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
- sysex just magically does not work on the target machine ;o) I have no
clue why except for this thing with being truncated but it *never* works even if the message seems to go out intact. Even more interesting that the machine shows me that it is recieving sysex! Tried with the mpu-401 of an Audigy and with a MOTU 828 mkII through Jack and a2jmidid (jack-alsa midi bridge).
I haven't actually tested sysex out limits, but sysex in is limited to something like 24 byte messages..
here is a thread where i talk about it: http://lists.puredata.info/pipermail/pd-list/2009-01/067372.html
It is hard coded into the source. You can modify the source to change the hard coded limit.. i have a diff on that thread which shows how i did it.
Thanks Alex, it's an important thing to know. My sysex is shorter than 24B it's just broken. What we shall see repeating is 240 67 16 73 8 16 19 (knobvalue) 247
Instead I'm having this grinded mess coming back on physical loopback:
print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 8 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 8 print: 247 print: 240 print: 67 print: 16 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247 print: 240 print: 67 print: 16 print: 73 print: 247
This makes me think I'm trying to send too much data in too short intervals...? Is there a handy way to limit the density of my output? (i think its not not [drip] as it has to stay realtime) I'm not very experienced, sorry if this traces back to something very obvious!
-- Muranyi Andras
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
2009/7/8 Alex x37v.alex@gmail.com
It looks like you may experiencing another thing that i've complained about... pd's alsa midi connects to all MIDI inputs and outputs that it can.. this includes alsa's MIDI through.. so everything that you send out gets echoed back into the input.. SUPER annoying.
run aconnect [or easier, though not great, aconnectgui] and see if that is the case for you.. disconnect pd from the alsa through port.
I've checked and it's ok. Right now have the output looped back with a cable but it only comes back once.
This makes me think I'm trying to send too much data in too short intervals...? Is there a handy way to limit the density of my output? (i think its not not [drip] as it has to stay realtime)
What about dropping some data before making up / sending out the sysex? My best guess was storing into [float] and firing it with [metro] or [pulse] at a constant, decent rate.
András Murányi wrote:
Dear Sirs,
I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have:
- sysex gets fragmented (also when nothing is sent out meanwhile by
other objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
in the patch you posted, the message you send to [midiout] is [240,67,16,73,8,$2,19,$1,247( , which Pd takes as 9 separate messages because of the comma in between each item. Probably [240 67 16 73 8 $2 19 $1 247( would work better.
Martin
2009/7/8 Martin Peach martin.peach@sympatico.ca
András Murányi wrote:
Dear Sirs,
I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have:
- sysex gets fragmented (also when nothing is sent out meanwhile by other
objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
in the patch you posted, the message you send to [midiout] is [240,67,16,73,8,$2,19,$1,247( , which Pd takes as 9 separate messages because of the comma in between each item. Probably [240 67 16 73 8 $2 19 $1 247( would work better.
Dear Martin,
that is how it goes afaik, sysex is kind of a sequence, which is comma separated as discussed here: http://puredata.hurleur.com/sujet-417-sysex
I just cannot prove it as i have never seen sysex out working :o)
András Murányi wrote:
2009/7/8 Martin Peach <martin.peach@sympatico.ca mailto:martin.peach@sympatico.ca>
András Murányi wrote: Dear Sirs, I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have: - sysex gets fragmented (also when nothing is sent out meanwhile by other objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!) in the patch you posted, the message you send to [midiout] is [240,67,16,73,8,$2,19,$1,247( , which Pd takes as 9 separate messages because of the comma in between each item. Probably [240 67 16 73 8 $2 19 $1 247( would work better.Dear Martin,
that is how it goes afaik, sysex is kind of a sequence, which is comma separated
Yes and no: sysex is a sequence that starts with the value 240 (hex F0) followed by a number of bytes, each less than 128, and ends with 247 (hex F7). There are no commas in it unless the message contains commas. (If you really want to send commas you would need to specify the ASCII code for a comma, 44.) As far as Pd goes, separating items in a message box makes them into separate messages, so it's no surprise they get sent out separately.
Martin
I think that pd's midi out doesn't take whole sysex messages though, you simply send a stream of bytes. This is how it has been working for me. I created an abstraction which takes a list of bytes and makes it into a sysex message (stream of bytes) [adds the sysex start and start and then outputs bytes one by one] and it works quite well for me.
I've attached it just as an example.
Btw, the sysex loopback code [using the software through] that you sent initially does work for me.. I get the whole message.
-Alex
2009/7/8 Martin Peach martin.peach@sympatico.ca:
András Murányi wrote:
2009/7/8 Martin Peach <martin.peach@sympatico.ca mailto:martin.peach@sympatico.ca>
András Murányi wrote:
Dear Sirs,
I just have never been successful with sysex in Pd and I'm asking for your kind advice. Attached is an example for a long sysex message which goes out on [midiout] (linux only, afaik). The problems I have: - sysex gets fragmented (also when nothing is sent out meanwhile by other objects). It is always properly terminated with '247' but the content is often truncated. Is this a bug/feature...? What is your best recommendation knowing that my patch is full of sysex outs (+envelopes on some of them!)
in the patch you posted, the message you send to [midiout] is [240,67,16,73,8,$2,19,$1,247( , which Pd takes as 9 separate messages because of the comma in between each item. Probably [240 67 16 73 8 $2 19 $1 247( would work better.
Dear Martin,
that is how it goes afaik, sysex is kind of a sequence, which is comma separated
Yes and no: sysex is a sequence that starts with the value 240 (hex F0) followed by a number of bytes, each less than 128, and ends with 247 (hex F7). There are no commas in it unless the message contains commas. (If you really want to send commas you would need to specify the ASCII code for a comma, 44.) As far as Pd goes, separating items in a message box makes them into separate messages, so it's no surprise they get sent out separately.
Martin
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Alex -
András Murányi -
Martin Peach