First time here? Check out the FAQ!

replicator panning

0 votes

Hi all


I want to create a replication of a sound when the input amplitude exceeds a threshold and then pan that copy to a unique location. 

I have played with the Chopper and differing angles but I can not hear them all at once - I want to be able to hear all positions at once not have to pan to them - I dont see a replicator that allows me to have all multichannel (quad in this case) output channels open at once?

Also I am wondering how to make a new sound happen when the input threshold is exceeded - this would be a snare live drum hit.  It seems to me that in the replicator all instances are made at compile time?


asked Jul 25, 2015 in Using Kyma by garth-paine (290 points)
So initially, you want to hear it from all channels?  And when the threshold is crossed, it should jump to one channel?
Hi - I want to create one sound on the first snare hit at a location in the pan angle - then on the second strike, create another version of the sound at a new angle and then both play - then on the third, another version at a new angle and now 3 play etc
I want to do this over a 4 channel system, but make it flexible so I could do it over 8 if available.  The total number of instantiated algorithms would be set somehow and then would lead to replacing the previous voice - thats a latter problem though.  Thanks, Garth

2 Answers

+1 vote

If you put a MultiSample into a MultichannelPanner and then into a Replicator, then:

• You could set the angle of the MultichannelPanner to depend on ?VoiceNumber (the copy in the Replicator), for example:

?VoiceNumber / ?NumberVoices * 2 - 0.5 (to get the range from -0.5 to 1.5 for the angle).

• You could trigger the MultiSample with a Threshold sound.  To get each subsequent threshold crossing to trigger the next voice, you could use:

([ThreshSound] L countTriggersMod: ?NumberVoices) eq: (?VoiceNumber - 1)

• To trigger slightly different snares each time, you could make the Index of the MultiSample also depend on ?VoiceNumber, for example:

?VoiceNumber - 1 / (?NumberVoices - 1)

This should scale with the number of voices (if you increase from quad to more channels). Let us know how it goes!

answered Jul 25, 2015 by ssc (Savant) (119,300 points)
awesome and so fast as always :-) Ill try these out - thanks

Ok thanks - I want the Snare sound to come from a live source (a musician on stage) - but I will work that out as I go - the challenge there is haveing the pacement happen at the time the input sound is made, and not at compile time.

But to get back to the initial challenge - I have experimented with the replicated in this TEST and can indeed get all 4 running at the same time - I would like the angle to be tighter - as per the 4Output - ie a much tighter diffusion on each channel (althought I do want to be able to pan them to). Any thoughts on that?  I tried varying the Radius without much effect.  Could you also tell me is the first ?VoiceNumber = 0  or 1?

I tried adding a SoundToGlobal controller to show the number in the VCS but it did not like my approach :-( 


ps. I posted this as an answer as I could not upload a sound in the comment field

In that case, SelectableAudioOutput might be a better choice than MultichannelPanner.  Then in the Channel field you could use something like:

([ThreshSound] L countTriggersMod: 4) + 1

to set the output channel?
?VoiceNumber starts at 1 (not at 0).
+1 vote

You may also be interested to learn about the 'Panning Strategies', a suite of Sounds perfectly suited for panning voices in a replication (and more). They are part of the many sounds you can find at the Complexity Lab.


answered Oct 30, 2015 by cristian-vogel (Master) (8,370 points)
Thanks Cristian, this image would be a great resource for the NeverEngine section of the Community Library (!  Do you need any help figuring out how to post there?  Please let us know.