Dull. The midi polling code (run once a tick) was happily giving up after handling one incoming MID message per device - so as soon as there were more than about 700 messages per second they piled up in teh driver and eventually started getting dropped.
I put in a throttle of 100 messages per device per tick, just to keep incoming MIDI from bringting the rest of Pd totally to a halt. cheers Miller On Sun, Jun 24, 2012 at 12:18:24AM -0400, chris clepper wrote: > Miller > > Out of curiosity, what was the fix? > > On Sat, Jun 23, 2012 at 8:29 PM, Miller Puckette <[email protected]> wrote: > > > Yep.. I already responded to the other thread but will mention it here too > > just to be thorough... I think I've got this fixed - it's up on the > > usual http://crca.ucsd.edu/~msp/software.html ... .43-3 test1. > > > > cheers > > Miller > > > > On Sat, Jun 16, 2012 at 01:20:05PM -0700, Miller Puckette wrote: > > > I went ahead and ordered a nice nanoKontrol2 - it should get here in a > > > few days and I'll report whether I can finally see this myself. > > > > > > cheers > > > M > > > > > > On Sat, Jun 16, 2012 at 10:11:02PM +0200, Ángel Faraldo wrote: > > > > Hi, > > > > > > > > thanks for your quick reply, Miller. > > > > > > > > On Jun 16, 2012, at 9:39 PM, Miller Puckette wrote: > > > > > > > > > Hi Ángel - > > > > > > > > > > I've seen a similar problem reported in Windows: > > > > > > > > > > http://lists.puredata.info/pipermail/pd-list/2012-02/094794.html > > > > > > > > > > and tried to recreate it with no success. I think the Max problem > > you > > > > > found is something different (because at least there the values > > settle > > > > > correctly when the sliders stop moving while for you it seems like > > something > > > > > is ending up in the wrong state). > > > > > > > > Exactly. > > > > > > > > > > > > > > Anyway, with apologies I have to say I'm delighted you can get this > > to > > > > > happen on a Mac - if indeed it's the same problem as on the PC it > > means > > > > > it's in teh Pd code and so if I can ever get it to happen here I can > > probably > > > > > fix it :) > > > > > > > > I'm glad you're delighted, specially if a fix can come out of it. > > > > > > > > thanks again! > > > > > > > > ángel > > > > > > > > > > > > > > Miller > > > > > > > > > > On Sat, Jun 16, 2012 at 07:24:09PM +0200, Ángel Faraldo wrote: > > > > >> Hi list, > > > > >> > > > > >> Yesterday I bought a new korg nanoKontrol2. To my surprise, after a > > few basic testings, i've discovered that when I move as little as three > > faders sending CC values simultaneously --and not incredibly fast--, pd > > doesn't receive a big part of the values (with [ctlin]), staying half the > > way in the midi-value range. > > > > >> > > > > >> ---------For example, moving three faders from minimum to maximum, > > that is, from 0 to 127, the values in the number boxes connected to [ctlin] > > end up in random numbers in between (like 63, 74 and 70). Even more weird > > (or maybe not!) is that when I move any of the faders again, the next value > > that is thrown out is 126 (as if the object had actually gotten the > > previous 127 and is going down from there!)--------- > > > > >> > > > > >> Indeed, using a midi monitoring program ('MIDI monitor' on my mac), > > I can see that the computer receives all the midi values as expected, so I > > guess that something happens on the [ctlin] level in Pd. > > > > >> > > > > >> Apparently this is/was also an issue on Max's implementation of > > [ctlin]: http://cycling74.com/forums/topic.php?id=19562 > > > > >> > > > > >> I have worked with other Midi controllers in Pd and I never had a > > similar problem. > > > > >> > > > > >> i've also tried [midiin]->[midiparse], with exactly the same > > results. > > > > >> > > > > >> (mac osx 10.6.8, pd-extended 0.42.5, pd-extended 0.43.1, pd 0.43.2) > > > > >> > > > > >> any light on this?? > > > > >> > > > > >> cheers, > > > > >> > > > > >> Ángel Faraldo > > > > >> > > > > >> _________________ > > > > >> www.angelfaraldo.info > > > > > > > > > >> _______________________________________________ > > > > >> [email protected] mailing list > > > > >> UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > > > > > Ángel Faraldo > > > > _________________ > > > > www.angelfaraldo.info > > > > > > > > > > _______________________________________________ > > > [email protected] mailing list > > > UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > _______________________________________________ > > [email protected] mailing list > > UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > _______________________________________________ [email protected] mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
