ok to make the patch more clear: The makenote part is obsolete, and onebang and midiflush are workarounds, but they flush all the notes upon channel change. apart of that, the patch is still correct.

anyone can help?

You could store the location value in a table with every note-on and retrieve that value when a note-off happens.

_ johan

Hi Johan,

thanks a lot for your help here at the cycling74 forum.

however, i have a problem, because I dont want to got random midi channels with my patch, but advance with a stepsequencer. i have attached this, and tried to implement your solution you indicated. I am somehow stuck on how to send the notes to the channels indicated (they are not midi channels), and the note offs do not seem to be send back to the right channels, but to different channels when the step sequencer advanded its steps.

I hope you can have a look at it many thanks. Emanuel

I think you were pretty close. The midiparse and midiformat just after each other is not necessary. As a matter of fact, I never tested the patch, but it seems conceptually correct. The gate and many sends could be replace with a forward object, that can be set dynamically.

_ johan

Hi Johan,

really almost there, thanks so far.

The remaining problem is, that note offs are not sent correctly. The note on triggers note number 1, for example, on channel 1. Then the stepseq moves on to channel 2. Then the note off of note number 1 should be triggered on channel 1, because it originally was sent on there. The patch doesn't do this now, note ons stay always way too long, result is arbitrary.

What do you think that is? Thanks!! manu

hey johan, did you already have a look at the last issue here? greatly appreciated... greetings manu