First time here? Check out the FAQ!

Stereo Spread

0 votes

(Nice sounding piece you got there Alan, would have loved to be a part of the audience,
but my journey in Kymaland only just began !)

I run into an issue using SCC's technique in the pan field:

?VoiceNumber - 1 / (?NumberVoices - 1) - 0.5 * !Spread + 0.5

when the Replicator is using 5 voices, everything sounds like a proper supersaw, but with 10 instance or more, it's starting to sound like it's 'drifting' in a weird way, how's that possible ?


also maybe instead of putting all the Sounds that are detuned below !KeyPitch on the left and those above on the right, or even intermingle them, could we find a way to evenly distribute lows and highs on the left and the right ? so, i can increase or decrease the number or replicated oscillators/tweak my sound as i want without damaging the stereo image ?

i attached my sound, if this can help you in any way:

thanks a lot,
(ps: i'm not sure if i answered your question or created a new topic, please,
feel free to put this answer where it belongs)



related to an answer for: Stereo Spread
asked Feb 5, 2019 in Sound Design by nicolas-petitfrere (290 points)
Hi Nicolas, that detuning expression relies on there being an odd number of oscillators; the center oscillator is in tune, half of the remaining oscillators are sharp, and the other half are flat. If you have an even number of oscillators, no oscillator will be exactly at the center and exactly in tune.
Hi ssc, thanks for your reply.

The !Detune function is working perfectly,
(the audio exemples above are using odd numbers)
i just had an issue with the stereo spread, but i just found out that the nasty drifting is only happening when the detune fader is really low which makes totally sense,
i should probably have both of them under the same fader in my VCS layout

have a great day,


1 Answer

+1 vote

Hi Nicholas,

Kyma is great at being precise but I think in this case you need some imprecision ;)
I've modified your supersaw, adding (below in bold) some 'Drift'; scalable, smoothable, frequency randomisation, seeded per voice in the oscilator pitch field via:

?VoiceNumber - 1 / (?NumberVoices - 1) - 0.5 * !Detune nn + !KeyPitch + ((!DriftAmt * ((1 bpm: !DriftRate *16) nextRandExpWithSeed: ?VoiceNumber - 1 / (?NumberVoices - 1) - 0.5) nn smooth: !DriftSmth) *2-1)

A simillar version of the above is in the panner providing pan drift. An amplitude drift might be interesting but I havent done that yet. 

I've also created a few presets demonstrating the effect.

The Freq Drift depth hot value is intentionally log and you can get some very swarmy effects at high values.

Best regards,


answered Feb 11, 2019 by sean-flannery (Adept) (1,490 points)
Sweet ! will try your version as soon as i can, thanks.


Just tried it, i can't believe how good it sounds, it's alive ~ !!
thanks a lot for your time Sean.