Weird Parsec spread behavior

[stp2015]

So when in initialize Parsec 2, ramp up spread and hit random midi notes, something very funny happens.

One of the stereo channels will play a square wave, the other one a saw-like sound.

Moreover, with each new midi note, left and right will switch.

The attached pic shows the signal on of the left channel of 4 short notes played.

What is going on here?

[Abstrax]

Probably because it's not a conventional spread function. Since it's an additive synth, it pans individual partials.

[stp2015]

Not conventional at all... [emoji3]

[stp2015]

By partials do you mean the underlying sine waves? If so, i dont see how that would result in the behavior I get...?

[Abstrax]

By partials do you mean the underlying sine waves? If so, i dont see how that would result in the behavior I get...?

Kind of...Every sound is just a collection of sine waves. The fundamental, and then the added harmonics (in additive synthesis, a harmonic is called a partial). As far as I know, a square wave is a fundamental with all the odd harmonic overtones. A saw is all even & odd harmonics. Since all the partials are panned independently, I'm assuming the partials alternate. So one side is the fundamental and the even harmonics (sounds like saw), other side is odd (sounds like square).

There are quite a few people on this forum that are balls deep in this subject, and know way more about harmonics than I do. Hopefully they can throw a few cents in.

[stp2015]

Ok, so my understanding is the same as yours (two non experts agree...)

So i would have expected to see the fundamental sine of the Saw wave center, then the first partial left, secon partial rigbt, third left, and so on. The position of the spread knob should then determin HOW far off center the partials are.

But again, that is not at all what i am getting.

Can any expert chime in?

[stp2015]

I understand your point now, did not quite get it the first time. Maybe this can be called a spread then, but it does lead to some funny results. Maybe it is not an optimal implementation.

[eXode]

If it spreads all the odd partials you will get a square on that side since a square is built from odd partials.

It's not ideal, but Parsec spreads differently depending on various factors such as if you are using both generators or not, also if you are using a modifier (and which modifier is used, iirc). But at the end of the day you can't really spread a single sawtooth in an meaningful way.

[stp2015]

Ok, makes perfect sense now as to what is going on. You are right about spreading a single saw. However, I still find the spread also sounds funny when i use the unison modifier.

I guess it is best to accept that the Parsec spread may just not yield the type of sound i am looking for in a spread function.

[stp2015]

Thanks for your responsed