chapelier fou

is there a way to specifically target one particular voice of a poly~ patcher ? I mean with the "midinote" message, not with "custom" messages. The goal is to intentionally sequence the order of voices.

Hi, 
is there a way to specifically target one particular voice of a poly~ patcher ? I mean with the "midinote" message, not with "custom" messages. The goal is to intentionally sequence the order of voices. 

Hi, 
is there a way to specifically target one particular voice of a poly~ patcher ? I mean with the "midinote" message, not with "custom" m

target-"midinote"-to-a-specific-poly~-voice

Also (as a side question), is it possible for one instance of poly~ to send a message to the "parent" patch containing a prefix that can allow to identify from which voice the message comes, in the parent patch ? 

In other terms, is there a way, inside [poly~], to get the voice number ?

Also (as a side question), is it possible for one instance of poly~ to send a message to the "parent" patch containing a prefix that can allow to identify from which voice the message comes, in the parent patch ? 
In other terms, is there a way, inside [poly~], to get the voice number ?

thispoly~ gives you the voicenumber, yeah.


thispoly~  gives you the voicenumber, yeah.

Thanks, that's encouraging, I'll dive deeper into this. 

So in i a nutshell : if I use the midinote message, I have no way to act on how voices are managed. If I don't use it, I'll have to make my own voice management and use a target message for each note on and off. Right ?

Thanks, that's encouraging, I'll dive deeper into this. 
So in i a nutshell : if I use the midinote message, I have no way to act on how voices are managed. If I don't use it, I'll have to make my own voice management and use a target message for each note on and off.  Right ? 

yes. though poly~ has three different modes since quite a while now, target is always more flexible.

in my opinion it has a lot of advatages to be able to do all the voicemanagement yourself from outside using target.


yes.  though poly~ has three different modes since quite a while now, target is always more flexible.

in my opinion it has a lot of advatages to be able to do all the voicemanagement yourself from outside using target.


Yeah I can see the advantages indeed ! It's just that I'm struggling to make it.

Wait...just read what you wrote about the "three modes", and "legacymode" could be just what I need....!

Yeah I can see the advantages indeed ! It's just that I'm struggling to make it.
Wait...just read what you wrote about the "three modes", and "legacymode" could be just what I need....!
Thanks 

if you control voices yourself you basically write abstractions which put a "target" message before the input.

for example you would make an object which sends target #1 for a single target, another one which has 32 inputs for voice 1-32, and then you have one for target 0 of course.

then you do one for round robin, one for random, one for reverse or whatever you think you need.

there is one thing which is a bit more work, and that is when you want voice stealing modes, because finding out when a voice is no longer busy is something which happens inside, and you have to get info out somehow and watch out for the message order when you want it to influence the next input.


if you control voices yourself you basically write abstractions which put a "target" message before the input.

for example you would make an object which sends target #1 for a single target, another one which has 32 inputs for voice 1-32, and then you have one for target 0 of course.

then you do one for round robin, one for random, one for reverse or whatever you think you need.

there is one thing which is a bit more work, and that is when you want voice stealing modes, because finding out when a voice is no longer busy is something which happens inside, and you have to get info out somehow and watch out for the message order when you want it to influence the next input.


My issue is more the process of figure out precisely the voice allocation algorithm and then how to translate it into Max.

Oh I do get the principle very well. 
My issue is more the process of figure out precisely the voice allocation algorithm and then how to translate it into Max. 

I am not clear on what you are trying to do but you could use the busymap message to figure out what voices are busy.

for "classic mode" you´d use a counter which can be reset, and for round robin you could use [cycle].

if you want to cycle AND be able to reset you could steal someones 20 years old basic abstraction.

grafik.png


algorithms?

for "classic mode" you´d use a counter which can be reset, and for round robin you could use [cycle].

if you want to cycle AND be able to reset you could steal someones 20 years old basic abstraction.



Thanks a lot. Still having a headache but I'll find my way through it.

target "midinote" to a specific poly~ voice ?