Hi all,

in this patch non-integer numbers between 60 (C4, 261.63Hz) and 61 (C#4, 277.18Hz) are generated:

midi.pd
If it is run, the same 60 note is always played (as expected, in the midi logic there is no middle ground between 60 and 61).
But is there a way to "bend" the sound to match the exact value? (practically with the exact frequencies coming out of [mtof]).

When in puredata I write a midi file, is there any parameter to take into account these micro-variations in pitch to be used later in a DAW in which I import the .mid file (for example Ardour)?

Thanks,
a.

[I use Purr-Data + Qsynth and Ardour DAW on Ubuntu 18.04]

@atux I think [bendout] is the object you are looking for.
It takes values from -8192 to 8191 so 16384 steps including 0.... no bend.
A little maths to translate the output of [mtof] will be necessary.
David.

It looks as though [bendin] in Purr Data is compatible? (or not?) with vanilla though........ https://git.purrdata.net/jwilkes/purr-data/-/issues/245

That will output the whole note before the [bendout] message which is probably what is required, but you might need to reverse the message order if the DAW doesn't respond correctly.

@whale-av @atux I remember having this problem once and trying to solve it with pitchbend, but IIRC the problem is that pitchbend is global to all on notes on that MIDI channel, so you can't really do simultaneous microtonal notes unless you distribute the notes across several MIDI channels. I think I gave up on the external MIDI synth and just wrote something in Pd, since Pd supports fractional pitch.

@jameslo Per-note pitch bend is part of the new MIDI 2.0 spec, not sure how much stuff supports the 2.0 spec yet but it has been out for over 2 years now so it should be getting out there. 2.0 also supports microtonal tuning in general so no more needing to futz with bending to get microtonal tunings. Now we just need pd to adopt the 2.0 spec.

@oid How raw is [midiout].... ??
Did you establish that it cannot build a midi2 message...?
https://forum.pdpatchrepo.info/topic/13662/raw-midi-midiout

Just wondering because it can build NRPN... but does it have list constraints? RAW suggests "anything" in the message string that is just going to be converted to HEX.
It looks as though midi2 can carry both midi1 and midi2 messages, so maybe there is just one initial status byte to change.
David.

@oid Ha! I should've seen that coming. In addition to being old, I only have access to ancient synths....

@whale-av said:

I think [bendout] is the object you are looking for.

Thanks, it works perfectly in Purr-Data.

This is my initial patch with your add:

midi2.pd

Soon I will study your link "Raw midi + midiout" to understand how to "fix" the microtonal informations in a midi file.

Bye,
a.

@whale-av [midiout] can handle a sysex dump so list length should not be an isssue, I assume it is just dripping the list since midi is serial. Probably just need to throw together a simple [makenote2.0] abstraction. I was going to throw one together and try it out, but it seems midi.org decided that with the release of midi 2.0 they also had to update the website as well and I got tired of clicking links and scrolling looking for what I wanted (their old website). Maybe another day if no one beats me too it.

@whale-av said:

@atux I think [bendout] is the object you are looking for.
It takes values from -8192 to 8191 so 16384 steps including 0.... no bend.

Hi,

by performing a MIDI discrete glissando, for example starting from the midi note 60 and going upwards: when you go from 60.99 to 61, from 61.99 to 62 and so on... there is a downward leap.
I understood (by trial and error) that by replacing 8191 with 4095 (which is the whole part of 8191), the problem is solved and the glissando is regular and without discontinuity.
But I didn't understand the reason.

Bye,
a.

midi3.pd

Hi all,

I state that in Purr-Data my patch above works perfectly, it performs a clean discete glissando of microintervals, from midi note 60 upwards.

Out of curiosity I tried to open the same patch in Pd version 0.51.4: nothing works! The counter object stops at 10 (why?), the bendout part has no effect, as if it were not there.
I didn't realize it, but there is a fairly high level of incompatibility.

[OS: Ubuntu 18.04]

Bye,
a.

@atux I think the bug here is with your install of pd, not pd in general itself. Works fine for me and should work on even quite old versions of pd.

@atux said:

I state that in Purr-Data my patch above works perfectly, it performs a clean discete glissando of microintervals, from midi note 60 upwards.

Out of curiosity I tried to open the same patch in Pd version 0.51.4: nothing works! The counter object stops at 10 (why?),

I can't reproduce this problem in my system -- opened your exact patch file and [counter] is fine.

IIUC [counter] is from cyclone. Purr Data bundles cyclone within its own installation. Standard Pd requires it to be installed as an external. Possibly there's a problem with this installation of cyclone.

the bendout part has no effect, as if it were not there.

I can't reproduce this either. On my system, [bendout] is sending and other apps receive it, no problem.

hjh

@atux Here are some simple vanilla counters...... counters.zip ....... [count] or [count-1to] will work for you except that you have to set the upper limit (maybe 127...?) and the [reset( message goes to the left inlet.
[bendout] not working in Linux doesn't seem to have been flagged by anyone else...... so??
David.