Hello list,
I run into a special problem with MIDI input under osx...
I am getting some MIDI inside Pd, the MIDI is coming from an interface-z card that is transducing a sensor. The midi info is on two control number for higher (12bit) precision. Sometimes the MSB value is not updated well, but the LSB is, and that gives me weird behaviors.
I have tested the same patch with Pd-extended 0.42.5 under mac osx lion and on another computer with osx snow leopard, both have the same problem, but the same patch on a third computer with windows xp is working well. I have tested as well two midi->usb interface (a basic one and a motu micro lite) with the same behavior.
So it seems to me that the problem looks like Pd midi driver on osx. I have tested to modify the audio latency, to change the MIDI latency, but that gives me no better result.
Do you have an idea of where i can look, or a workaround i can find?
Best, n
Hello i would try to force synchronization of incoming midi streams inside pure data according to my needs, ie downsampling a little if necessary so that MSB and LSB are consistent ... and give a call to interface-z for feedback. cheers JM
Le 6 févr. 2012 à 21:15, Nicolas Montgermont a écrit :
Hello list,
I run into a special problem with MIDI input under osx...
I am getting some MIDI inside Pd, the MIDI is coming from an interface-z card that is transducing a sensor. The midi info is on two control number for higher (12bit) precision. Sometimes the MSB value is not updated well, but the LSB is, and that gives me weird behaviors.
I have tested the same patch with Pd-extended 0.42.5 under mac osx lion and on another computer with osx snow leopard, both have the same problem, but the same patch on a third computer with windows xp is working well. I have tested as well two midi->usb interface (a basic one and a motu micro lite) with the same behavior.
So it seems to me that the problem looks like Pd midi driver on osx. I have tested to modify the audio latency, to change the MIDI latency, but that gives me no better result.
Do you have an idea of where i can look, or a workaround i can find?
Best, n -- http://nim.on.free.fr
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
thanks for your reply, The problem is that midi data is a position feedback for the control of a motor, and when msb is stuck, the position command I give is wrong. I don't think the card is to blame cause the same card/patch/interface is working well on win XP. n
Le 06/02/12 21:57, Jean-Marie Adrien a écrit :
Hello i would try to force synchronization of incoming midi streams inside pure data according to my needs, ie downsampling a little if necessary so that MSB and LSB are consistent ... and give a call to interface-z for feedback. cheers JM
Le 6 févr. 2012 à 21:15, Nicolas Montgermont a écrit :
Hello list,
I run into a special problem with MIDI input under osx...
I am getting some MIDI inside Pd, the MIDI is coming from an interface-z card that is transducing a sensor. The midi info is on two control number for higher (12bit) precision. Sometimes the MSB value is not updated well, but the LSB is, and that gives me weird behaviors.
I have tested the same patch with Pd-extended 0.42.5 under mac osx lion and on another computer with osx snow leopard, both have the same problem, but the same patch on a third computer with windows xp is working well. I have tested as well two midi->usb interface (a basic one and a motu micro lite) with the same behavior.
So it seems to me that the problem looks like Pd midi driver on osx. I have tested to modify the audio latency, to change the MIDI latency, but that gives me no better result.
Do you have an idea of where i can look, or a workaround i can find?
Best, n -- http://nim.on.free.fr
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
So I cant help further ... but Ive got the same card that i use with a motor in low precision for now, on OsX !!! thus I might run in the same problem later ! I will follow the link ! :) JM
Le 6 févr. 2012 à 23:20, Nicolas Montgermont a écrit :
thanks for your reply, The problem is that midi data is a position feedback for the control of a motor, and when msb is stuck, the position command I give is wrong. I don't think the card is to blame cause the same card/patch/interface is working well on win XP. n
Le 06/02/12 21:57, Jean-Marie Adrien a écrit :
Hello i would try to force synchronization of incoming midi streams inside pure data according to my needs, ie downsampling a little if necessary so that MSB and LSB are consistent ... and give a call to interface-z for feedback. cheers JM
Le 6 févr. 2012 à 21:15, Nicolas Montgermont a écrit :
Hello list,
I run into a special problem with MIDI input under osx...
I am getting some MIDI inside Pd, the MIDI is coming from an interface-z card that is transducing a sensor. The midi info is on two control number for higher (12bit) precision. Sometimes the MSB value is not updated well, but the LSB is, and that gives me weird behaviors.
I have tested the same patch with Pd-extended 0.42.5 under mac osx lion and on another computer with osx snow leopard, both have the same problem, but the same patch on a third computer with windows xp is working well. I have tested as well two midi->usb interface (a basic one and a motu micro lite) with the same behavior.
So it seems to me that the problem looks like Pd midi driver on osx. I have tested to modify the audio latency, to change the MIDI latency, but that gives me no better result.
Do you have an idea of where i can look, or a workaround i can find?
Best, n -- http://nim.on.free.fr
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
for archive: the midi flow coming from the card was too high... for osx but not for win. good news, best,
n
Le 07/02/12 10:36, Jean-Marie Adrien a écrit :
So I cant help further ... but Ive got the same card that i use with a motor in low precision for now, on OsX !!! thus I might run in the same problem later ! I will follow the link ! :) JM
Le 6 févr. 2012 à 23:20, Nicolas Montgermont a écrit :
thanks for your reply, The problem is that midi data is a position feedback for the control of a motor, and when msb is stuck, the position command I give is wrong. I don't think the card is to blame cause the same card/patch/interface is working well on win XP. n
Le 06/02/12 21:57, Jean-Marie Adrien a écrit :
Hello i would try to force synchronization of incoming midi streams inside pure data according to my needs, ie downsampling a little if necessary so that MSB and LSB are consistent ... and give a call to interface-z for feedback. cheers JM
Le 6 févr. 2012 à 21:15, Nicolas Montgermont a écrit :
Hello list,
I run into a special problem with MIDI input under osx...
I am getting some MIDI inside Pd, the MIDI is coming from an interface-z card that is transducing a sensor. The midi info is on two control number for higher (12bit) precision. Sometimes the MSB value is not updated well, but the LSB is, and that gives me weird behaviors.
I have tested the same patch with Pd-extended 0.42.5 under mac osx lion and on another computer with osx snow leopard, both have the same problem, but the same patch on a third computer with windows xp is working well. I have tested as well two midi->usb interface (a basic one and a motu micro lite) with the same behavior.
So it seems to me that the problem looks like Pd midi driver on osx. I have tested to modify the audio latency, to change the MIDI latency, but that gives me no better result.
Do you have an idea of where i can look, or a workaround i can find?
Best, n -- http://nim.on.free.fr
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Jean-Marie Adrien -
Nicolas Montgermont