For anyone who can help..... the patch is here...... https://forum.pdpatchrepo.info/topic/13555/picodrumz-extremely-simple-drum-module
Personally I think it needs a better ADSR than the simple [line~] but maybe that's a matter of taste.
Changing [the message to line to [1, 0 2000( and the [pow 4] below to about 50 works better for me.
With a syncopated rhythm triggered manually the pitch seems to vary, but I think that's psychoacoustic...
But it could be the [phasor~] fm though as you say...?
Very hard for me to be sure.
David.

@mezko The list-abs library has an abstraction for this... list-inter.zip
It might produce the same lists as @oid's patch. I have not tested.
I imagine that it will do what you are looking for if the correct lists are applied to the second and third intets.
David.

@KoaN Thank you for posting what you have discovered. It will surely help someone in the future.

In windows you can start 2 instances of Pd.... one for audio and one for midi.... which can help as although the audio settings do affect the midi lag you can set one of the instances to run with a tiny buffer that would cause audio dropouts..... but use that instance purely for midi.
This can be done in other os's too.

The 2 instances can then communicate control rate messages using [netsend] and [netreceive].
A steep learning curve but you seem to be advancing rapidly.

The instances can be opened with a pre determined setup using command line flags.... http://msp.ucsd.edu/Pd_documentation/resources/chapter3.htm#s3.4.1

Here is an example of a batch file that should achieve this.....

start "C:\Users\David\Desktop\Pd Vanilla\pd-0.55-2.msw\pd-0.55-2\bin\pd.exe -noaudio -midiindevice 2 -midioutdevice 2" start midi_patch.pd
start "C:\Users\David\Desktop\Pd Vanilla\pd-0.55-2.msw\pd-0.55-2\bin\pd.exe -r 44100 -asio -nomidi -audioindev 9 -audiooutdev 9 -inchannels 18 -outchannels 22 -audiobuf 2 -blocksize 64 -callback -nosleep" start audio_patch.pd
exit

The paths will be wrong for your computer, so will need to be changed, and the command line flags will be wrong for your sound card, and probably for the midi devices too. For the sound card and midi devices you should first open your patch and open the Media AudioSettings and Media MidiSettings and write down the device numbers and settings for those found on your system as Ps starts up.
Then create an audio patch (-nomidi) and a midi patch (-noaudio) and change the command flags to match your setup for each.

BUT I cannot (so far) get the flags to work in a batch file opening 2 instances (works fine for one instance so it is my batch programming problem).
So copy the code above into a simple text editor like notepad, save it as e.g. "pd_twice.bat" and then edit it to match your Pd binary location and the Pd files you want to load to each instance. Once the 2 instances are open set the audio and the midi for the instances to something like this and do not turn on DSP for the midi instance.......
Keep the batch file with the Pd files you are loading in the same folder.

Of course your settings for the audio and the midi instances will be different..... different devices, different delay for audio.... but the midi instance can have the delay set to 1 (or maybe even 0) as it will not be running any audio.
As @jameslo says above...... in Pd the "Delay" setting is the buffer for communication with the sound card.
David.

@KoaN Well done getting it to work... and welcome to the forum.
Pd will introduce a small delay. The "control" messages... the non-audio messages like midi... are processed in between the audio blocks. The standard block in Pd is 64 samples.
That gives a latency of about 1.45 milliseconds. If you move one and a half feet away from another musician you hear what they play one and a half milliseconds later..... so that latency is not really noticeable.
It could be that you have a large audio buffer set for your sound card. On a computer one thing has to finish before another can be processed and audio has priority. Try to keep buffers as small as possible. If you make them too small you will start to get clicks as the computer fails to process audio in time... so it's a compromise.
With a pro soundcard with good drivers you should be able to reduce the buffer to 128 samples or maybe even less. Using the computer on-board sound card you will need a much bigger buffer to avoid clicks and so some noticeable latency.

You are unlikely to get help here with those exact plugins..... as someone would have to have purchased them. But other forums dedicated to plugins might well give you some pointers.
David.

@porres Wow! Thank you to all the contributors..
David.

@Balwyn Personally I wouldn't bother with $0 as the clones and nested clones can have more arguments like this......
this.zip

Edit... I have not passed on an argument for the first clone.... so 20 messages received....obviously incorrect.... but no time to fix it now. But easily doable...

David.

@impression All clones are automatically assigned $1..... normally from 0 but here starting at 1.....
Clone is [clone -s 1 6voices_part 800 800]

Put a [receive] inside the receiving abstraction and [route] your message to the correct abstraction using [route $1]

The message will be e.g. [6 cutoff 0.378(
Where "cutoff 0.378" will be received at the outlet of [route $1] by the abstraction number 6.
Sent to [s curve] (in this case)......
David.

@impression Good news..!
Yes, the MrPeach objects take care of string, float etc formats without us needing to get involved in such details.. They are very useful.
David.

@impression If it is definitely sending then tcp and udp should work if you have specified the same port and the same format (tcp or udp) for both the send and receive..
If you put a [print] on the left outlet of [netreceive] you should see some data arrive.
Assuming you are not trying to receive on another (second) RPI, in which case you will need to tell the GPIO RPI the IP address of the second RPI as well as the port.
David.

@atux You can try that....... pd2py_v2.pd
Does it work?
If not then post your Python code and someone will tell you why not.
David.