MIDI Voice Allocator (polyphony bug)

The (MIDI Note Allocator) macro uses 8 instrument voices. I have set 8 voices in the instrument properties.
Notes gets stuck if playing more than 8 notes at once.



Edit:
I have discovered another bug too.
First press note 1.
Then press note 2 while note 1 is still holding.
And then press note 3 while note 1 and 2 is holding.
Then release note 2 while all 3 notes are holding.
And now there is hole on note 2.
The note 3 does not go the note 2 if note 2 has released.

 

*all* structures using 8 voices will fail if you play more than 8 voices at once.

and the second 'bug' you discovered is a feature - the allocator is designed for making finger-type ensembles. you don't want to suddenly switch which core macro is receiving a note on, it'd cause all sorts of coding problems.

 

Then i need to create my own midi voice allocator. To have it work the way i want it to behave.

 

Now i have created a lowest note sorter.

 

i don't understand why the audio voice combiners and a/e modules? it seems to work but it takes way too much CPU for the task it is performing.

 

Try this version without a/e modules:

 

... crashes reaktor on my win 7 machine.

 

Reaktor is buggy.

 

Reaktor does have a few bugs, but it's also pretty easy to bring things to a standstill by coding a very inefficient macro, can't blame Reaktor for that.
What exactly are you trying to do?
what are the requirements of the sorter?
why are there 8 outputs?

 

I am trying to build a sorter that works like this:
voice 1 = 60
voice 2 = 49
voice 3 = 70
voice 4 = 58
voice 5 = 65
voice 6 = 63
voice 7 = 81
voice 8 = 45
------------------------------------------
voice 1 = 45
voice 2 = 49
voice 3 = 58
voice 4 = 60
voice 5 = 63
voice 6 = 65
voice 7 = 70
voice 8 = 81

 

Can you respond?

 

Why do you need a sorter that works like that?
What exactly are you trying to do?
Whay can't you use Don's allocator from the UL?

I'm asking this because often we try to make things unnecessarily difficult for ourselves by trying to make complicated devices when with a bit of a re-think, simpler one's will work as well or better.
e.g. I built a very complicated system of proxy knobs and a custom soft takeover system for one of my devices. In a later project I achieved a similar but better functionality with a very simple approach using Reaktor built-in components.

 

well, my allocator certainly won't work for this.

i have one like this sitting around somewhere though, i think it's on my desktop but i'll upload it next time i'm nearby.

 

Recovering an old thread it seems.... I am looking for exactly this feature for an ensemble i'm doing.
Jedinhopy, did you find your way to an algorithm that performs the kind of voice sorting you described there?
I'm trying to build some kind of arpeggiator that needs this feature to sort the incoming notes. I need/want to work it out using polyphonic voicing (for an arbitrary number of voices), so it would be great to find this sorting code!

did you have any luck?
or anyone else?