n871

I have a patch including multiple sound oscillators. When I switch them all on simultaneously and play with my midi keyboard some weirdness is happening.

The note C for example has, just like any other note, a very clear sound since all oscillators are phase-locked. but when I play some other notes and return to this same C afterwards, it's timbre is often slightly different from the first time I played it (more hollow)...but again constant and non evolving as long as I play it. It's like the mixing of my oscillators changes slightly while playing. Their volume is however fixed.

Has someone encountered this problem too? My patch is about 1GB so difficult to upload.

unexplainable-timbral-changes-in-sound-anybody-recognizing-this-scenario

Sounds a lot like their phase relationship is changing. What objects are you using and how are you driving them? Perhaps you can simplify the patch and post it?

I will try to simplify the patch once I get back home. The objects/oscillators I'm using are 6 "2d.wave" objects. Their speed is controlled by the kslider.

In theory they speed up and slow down together so that there is no phasing. But maybe a few millisecond difference is enough to produce this unwanted problem. Maybe I need to make one BIG 2d.wave object instead of 6 smaller ones, does that make sense?

The 6 oscillators come together in a live.gain object, this is the first point in the patch where timbral discontinuity is measured.

I will try to simplify the patch once I get back home. The objects/oscillators I'm using are 6 "2d.wave" objects. Their speed is controlled by the kslider.
In theory they speed up and slow down together so that there is no phasing. But maybe a few millisecond difference is enough to produce this unwanted problem. Maybe I need to make one BIG 2d.wave object instead of 6 smaller ones, does that make sense?

The 6 oscillators come together in a live.gain object, this is the first point in the patch where timbral discontinuity is measured.


If they're all driven from the same phasor~ then they should stay in perfect phase. I don't use 2dWave~ very often, but just doubled up the object in the helpfile, loaded in a single cycle waveform and this was the case (see picture). If you are driving them from different phasors off your kslider then the different message timings are probably throwing it off.

thanks for the info, they are however driven by the very same phasor..

It mostly occurs when I'm playing really fast, the phasor is driven by a numbox, is it possible that this numbox is slowing everything down because of it's UI?

that could certainly be the problem. But we'd have to see the patch to know.

hmm I think I solved the problem...the second inlets of each 2d.wave were indeed connected to same phasor, but the first inlets were not...

I forgot that the first one determines the speed and therefor the pitch..

designing these oscillators took a lot of time...each loaded wave-recording is divided into 84 slices. Each slice has a slightly different timbre so that not only pitch but also the colour of the oscillator changes when moving from the lowest to the highest note. 2d.wave is truely a magnificent object.

Concerning numboxes I have one last question. Are there faster alternatives without UI? :) in some situations I really need them between objects so that their value can be "banged" with a button.

hmm I think I solved the problem...the second inlets of each 2d.wave were indeed connected to same phasor, but the first inlets were not...
I forgot that the first one determines the speed and therefor the pitch..

Concerning numboxes I have one last question. Are there faster alternatives without UI? :) in some situations I really need them between objects so that their value can be "banged" with a button.


Unexplainable timbral changes in sound. Anybody recognizing this scenario?