Forums > MaxMSP

Chorus-Like Multi Voice Synth Effect

February 7, 2014 | 12:12 pm

I’m trying to make a synth effect and am

A) Looking for advice on building the effect
B) Trying to figure out the goddamn name of it so I do some searches for material.

It’s a sort of chorus-y like effect where when you hit a key on the synth multiple voices play the same note, but slightly out of tune giving it a bigger, fuller sound.

I’m not sure if there’s an easy or a "right" way to do this, but I managed to do something close by just taking the midi-notes that were incoming to my synth patch and generating 4 new notes that were slightly off-pitch, so if the note played was 69, I’d send 68.9, 68.95, 69, 69.05, and 69.1 to get 5 voices spread over 20 cents.

This worked, but it didn’t sound quite right, which makes me think there’s something more to the actual effect.

Also, I just used + and – objects and had everything hardcoded, I’d rather have something where I could choose the number of voices and spread of de-tuning.

Advice?


February 7, 2014 | 1:21 pm

you can call it "chorus" (like korg once did) or you an call it just "beating" (at least when we are talking about a cosine or at last a somewhat simple waveform).

on the nord lead 3 it is called "stack detune" – but there it is not produced by layering several instances of the oscillator, it is applied to only one. (or in other words, it is not really an "effect")

what you describe in your last sentence is what i would suggest anyway. this can be done more or less easily by using [poly~] (while the calculation from the "spread" parameter to the note number or frequency values will still have to happen outside the poly~)

-110


February 7, 2014 | 1:22 pm

one can also use a set of tapping buffer delays with some clever windowing.


February 7, 2014 | 3:14 pm

Stack Detune! That’s the words. Thank you very much.


February 7, 2014 | 5:48 pm

Isn’t that called unison in many synths?


February 7, 2014 | 6:06 pm

Yes, unison is another term for it that I’ve seen. Not sure if there’s a difference.


February 8, 2014 | 4:21 am

it is good practice to use fixed frequency offsets, which are in a harmonic relation to the base frequency (say like 70/71 on the Hz layer), then only add more of these oscillators to increase the effect. aka jp-8000 if am not wrong.


February 8, 2014 | 5:49 am

Yes. The jp8000 (and possibly other Roland Synths) have iirc seven sawtooth oscillators with variable spread control. In their parlance it is called ‘Supersaw’

It should be easy enough to get something close in max by running say seven sawtooths but varying the pitch of each by a proportional amount.

There is an example (among other roland oscillator emulations) in this old thread: http://cycling74.com/forums/topic/iterative-feedback-oscillator/

And here is another go (better) using gen: http://cycling74.com/forums/reply/re-gen-patch-a-day-68/


February 8, 2014 | 2:26 pm

In addition to the names listed above, you often see it called something like polysaw, polysynth, or polywave.


Viewing 9 posts - 1 through 9 (of 9 total)