Formant VCO

Talking to Trent there’s no secret sauce - are you looking at the static technical map or the one with moving diagrams?
Btw it was all done In illustrator :slight_smile: I thought it was probably some matlab or something but nope, illustrator - Trent has this amazing ability like Brian and my old mate nik that did the omnimod to turn everyday tools into magic

I’m trying to copy sections I think will help but I’m on a iPad so once I wake up a bit I’ll go to my computer and give it a crack but I’m pretty sure your looking at the static version?

1 Like

Oh that bug is intentional and helps with the whole sound that makes the mangrove special. It intentially misses some impulses so once it reaches a certain integer it flips to octaves - I’m not sure if I explained that right though

1 Like

Ha I meant a bug in my code, the syncing algorithm I wrote has to handle the simd vectors and it doesn’t properly account for multiple syncs within a single vector. So super high freqs will bug out

Yea I was just looking at the github doc, didn’t know there was another

1 Like

@tomf Yeah you’re very close (video started working). There is only one extra oscillator in my ER-301 micro-case and it’s Mangrove. I left it there because I couldn’t imagine recreating the sound in the ER-301. I was debating if I should keep it or not lately and here we are, you are making the decision quite a bit easier!


Ah here, Mannequins-Technical-Maps/ at master · whimsicalraps/Mannequins-Technical-Maps · GitHub

Even at the extreme utone position of BARREL the impulse generator never releases to its resting state immediately. Instead there is a small window in which the impulse duration exceeds the clock duration, but the impulse is in its rapidly falling state. In this brief falling state, the impulse generator is receptive to triggers: if it receives a trigger, the generator turns around and starts rising again. When this occurs, our ears simultaneously perceive two separate tones - one at the original frequency, and one at the divided frequency in the undertone series. If you visualize this phenomena on an oscilloscope, it’s as though every second cycle slides into or behind its neighbour.

This is what I was missing! My impulse can’t start again until it’s completely done falling, but here it can be re-triggered after the rise stage


Happy to report it’s all finished, I’ll be releasing Formant along with another unit as soon as I can get the docs together.

The other unit I made is another application of this soft-syncing idea that’s so key to how mangrove works. When you sync, the wave just reverses direction so there’s no click.

Applying this to both the rise and fall stage on an oscillator turns out to be extremely useful, especially when using it as an LFO. The syncing action prevents sharp jumps and can add a pleasant rhythm to modulated parameters.

Now that I have the code to do this, I think I’ll change Arc to behave this way as well. The clicks from re-triggering the envelope always bothered me and that means I can also move the filter before the gain stage in the LPG unit for whatever that’s worth (purism?).


wow, can’t wait to give it a try!

1 Like

This sounds amazing already! Can’t wait to give the final version a go :heart_eyes:

PS If you have an envelope + vca + pulse oscillator in your rack you can patch this pretty easily, all analog. It’s a fun exercise for sure :slight_smile:

In my experience it can be a bit finicky but really changed how I’m looking at my layout.


thanks for this oscillator, it’s very versatile – impressively brassy :slight_smile: Here’s a saxy clip – there’s some FM going on for the pitch bending fwiw


This sounds great, how did you patch this?

1 Like

very very cool! is this made with my formant vco (the one this thread is about) or the new version by @tomf made in c++ ?
in any case, great use of the formant oscillator!!I love it too especially for brassesque sounds!

thanks both @hyena It’s tomf’s, but I’d like to try yours out too now :slight_smile:
@doomglue Thanks, I’ll share the chain with you in the next couple of days. Iirc, it’s a Maths looping envelope into the Scale Quantizer (phrygian) for the notes. The osc goes into @tomf’s LPG, which is being trigged by the EOC of another looping Maths env, not synced to the first.
But it’s the modulation of the Osc Freq that adds to the organic brassy feel. And a static sine osc is used to add more harmonics.
Oh and finally there’s a waveshaper at the end of the chain

Thanks for coding this @tomf – I love “uncanny valley” sounds :slight_smile:


beautiful patch, congrats!

1 Like

It really sounds lovely as there is so much expression in how the notes always slightly change.

1 Like

Whoaw, I’m really into this. Longer version available? More music somewhere? :slight_smile:

Technical question, (probably ignorance fuelled) but how exactly does a sine wave add harmonics, or is it tuned differently, or something?

1 Like

thank you! I’m editing and mixing a full version at the moment, will definitely share when it’s done. My other stuff isn’t intimate or as tonal as this. But I think I’ll just keep going with this setup. It’s enjoyable to make canons/fugues with the 301, very relaxing :slight_smile:

The FM sine is tuned an octave above the root freq of the Formant Osc. Actually, I think it’s so subtle it’s not adding any harmonics/inharmonics but rather just a little pitch flutter.


@Bparticle Edited a version of this. Brassy part starts around 3 mins in. The first part is the same patch doing something a little different. FWIW I fall asleep within minutes when listening to it :slight_smile:


Thanks, loved it, and I like the rest of the stream as well. Inspiring stuff.

1 Like

also curious how you gleaned these beautiful tones, not really understanding your notes…