Clean PWM in Bazille

[Urs]

Oh and a last one, as much as I LOVE current Baz's oscillator (I do, a LOT!), is there anything considered in that regard ? I'd love to hear your take on Serge / Buchla

For a hardware version we would have to make it a dual one and allow for internal routing of feedback modulations.

And when it comes to software ?

Oh, you mean Bazille Osc in ACE? - Doesn't make much sense IMHO unless we build that much larger kind of modular system we originally wanted to do.

[MitchK1989]

I just found this thread while searching for a way to do PWM in bazille - I think the nilhartman is asking about a third u-he cabled modular in the same line as ACE/Bazille that focuses on buchla style sound modules (triangle core based "complex osc" focusing on analog FM/RM into a wavefolder, Vactrol style LPGs, etc) not for bazille's oscillators to be ported into eurorack or ACE.

At any rate, has any tricks been found for actually doing analog style PWM in bazille? I've been guilty of using bazille without FM sometimes as a kinda "ace XL" because it's modulation possibilities are insane and it has four gorgeous filters.

edit: I feel like I owe people a bonus tip for reviving a dead thread - If your host has the ability to assign modulation to host automatable parameters (like you can in M4L or bitwig) adding an envelope (on mono sounds) or aftertouch to the "shake" parameter on the spring reverb can make for some cool effects

[Howard]

User guide, search for "PWM". I prefer method 4.

[MitchK1989]

User guide, search for "PWM". I prefer method 4.

Thanks, I actually like that method a lot too (though I also like the sound of method 3 as well it's not very "analog" in it's implementation and ties up a second osc)

I remember reading the entire user guide back when bazille was new all those years ago but between some updates both to the manual and the synth and my memory not being photographic I could probably use a re-read.

Edit: Just found a way of doing it that ties up fewer modules - Just run the sine through the rectifier and into the quantizer, adjust the oscillator level until you get a square (or whatever kind of pulse you want your starting point to be) and then oscillator level modulation becomes PWM with the caveat that you can go above max/below minimum if you push it too far. - this makes your osc unipolar though so it does change how it mixes with other waveforms or distorts if you drive the filter. I think I'll go back to the default "method 4" but I don't think you need the rectifier on the LFO now that the LFOs can be set to unipolar

[Howard]

Edit: Just found a way of doing it that ties up fewer modules - Just run the sine through the rectifier and into the quantizer, adjust the oscillator level until you get a square (or whatever kind of pulse you want your starting point to be) and then oscillator level modulation becomes PWM with the caveat that you can go above max/below minimum if you push it too far. - this makes your osc unipolar though so it does change how it mixes with other waveforms or distorts if you drive the filter.

Interesting "PWM 5" method

I think I'll go back to the default "method 4" but I don't think you need the rectifier on the LFO now that the LFOs can be set to unipolar

Indeed not - thanks! Updating the user guide...

[gentleclockdivider]

You can do pulse width modulation with just fm (phase modulaion) , I did it recently on my tg 77
Operator 1 , ratio 1
operator 2 : ratio 2 and self feedback ( so it becomes a saw )
operator 3 ; ratio 1 ( also self feedback = saw )
Only caveat is that you have to adjust the phase ofset of operator 3 with 90 degrees ( luckilly the tg 77 does that )
here's the routing , adjusting op3 output defines pulse width
Edit , operator 1 and 2 need at least 2 fm inputs , operator one from op2 and 3 , and operator 2 from itself and three ..and all with individual mod.amount .. in bazille this becomes a mess pretty fast ( trying hard with the multiplexer but no luck getting a nice sounding pw )

[MitchK1989]

another option - you can pull the same trick that zebra + podolski use to generate PWM from sawtooth waves - just combine osc 1's sawtooth with an inverted osc 2 sawtooth (with both set to gate so their phases match) and then LFO the phase of osc 2.

[gentleclockdivider]

Here's an example of pulse width modulation done with phase modulation ( FM method described above )
Done in reaktor .
https://app.box.com/s/ak7c4i4it3jbfi7fub7xoxpgy6bsx2av

[gentleclockdivider]

And here it is in bazille , using fm only