I have a patch, where I record a few seconds of audio into a buffer.

The I use a poly~ to access several different positions of the buffer at the same time. I.e. each voice instance represents a position in the buffer. I play this instrument with Mira. The position to be read from the buffer is selected according to the position of where I press on the x axis. 

This is fine for slow attack pads, but when using it percussive, I've noticed, that the poly shoots of a voice instance before updating the position of the buffer. So the attack of the sound is the "old" position of the buffer, that the voice instance used before being brought to life again.

Is this just because it takes a while to read the buffer or have I programmed this sluggishly?

message

toggle

gate

mira.mt.touch

slider

midiparse

route

midiin

outlet

send~

live.dial

svf~

inlet

receive~

comment

gain~

loadmess

loadbang

umenu

pack

line

scale

translate

pattr

prepend

sprintf

live.toggle

switch

join

unjoin

button

fromsymbol

zl.join

zl.rev

unpack

timestamp

sfrecord~

rslider

live.button

selector~

number

pipe

snapshot~

append

change

hzd-lfo

bpatcher

beatstepin

gran-rec

adc~

clip

flonum

mira.multitouch

deferlow

zl.4096

zl.nth

jit.matrixinfo

jit.fpsgui

jit.matrix

maximum

jit.submatrix

jit.3m

jit.buffer~

poly~

mira.mt.fingers

buffer~

mira.frame

multislider

meter~

attrui

The patch is quite messy, as it has been around 5 years in the making and constantly being changed.

I have a patch, where I record a few seconds of audio into a buffer.
The I use a poly~ to access several different positions of the buffer at the same time. I.e. each voice instance represents a position in the buffer. I play this instrument with Mira. The position to be read from the buffer is selected according to the position of where I press on the x axis. 
This is fine for slow attack pads, but when using it percussive, I've noticed, that the poly shoots of a voice instance before updating the position of the buffer. So the attack of the sound is the "old" position of the buffer, that the voice instance used before being brought to life again.

The patch is quite messy, as it has been around 5 years in the making and constantly being changed.


Hi.
I have a patch, where I record a few seconds of audio into a buffer.
The I use a poly~ to access several different positions of the buff

poly~-using-a-single-buffer~

It's a question of sequencing the numerous messages you sent inside the poly~ voices.

I would suggest instead of sending the parameters, that you use the 'note' message followed by all of your parameters.

If you keep your existing architecture, you'll have to make sure you send the 'midinote' message after you send the position. 

In any case, use and abuse [trigger] / [t].

It's a question of sequencing the numerous messages you sent inside the poly~ voices.
I would suggest instead of sending the parameters, that you use the 'note' message followed by all of your parameters.
If you keep your existing architecture, you'll have to make sure you send the 'midinote' message after you send the position. 
In any case, use and abuse [trigger] / [t].

watch out, sometimes it could be required that audio is turned on in the voices before they start "playing" something, i.e. you should leave them all turned on instaed of using automatic voicing.

watch out, sometimes it could be required that  audio is turned on in the voices before they start "playing" something, i.e. you should leave them all turned on instaed of using automatic voicing.

"Leaving them all turned on" does that mean, getting rid of the "thispoly" object?

I think the problem lies outside of the poly object. The finger instance of the mira.frame first activates the poly instance which then reports back to "the finger" which poly instance was activated and then links the finger to that insance.

I guess I should try to figure out, how to activate the poly instance without activating the adsr~ Sounds very simple now, as I write it :)

poly~ using a single buffer~