While waiting for Whimsical Raps to restock their new Silhouette module I thought it might be fun to attempt to build something similar in VCV as a patch as it’s a very unique module.
Figure the goal is not even a 1:1 recreation but something that works in from what I understand in a similar spirit. Being able to scan around a sound source while sending some frequencies to some effects and highlighting others with other effects.
I’ve been attempting it with separating out a stereo input into high, mid, low frequencies with the Mindmeld EQmaster and from there figure each band has to go out to a morphing mixer that would allow to scan around the stereo and frequencies of the input and send to various effects.
I had a short look at SILHOUETTE – WHIMSICAL RAPS at the manufacturers website. Including the short demo video. At the core it looks like a knob/phasor driven 6 (light) + 1 (background) router/switch/mixer/crossfader.
Not sure how the BBD delay options work. In the demo video it simply creates a time delay between left and right channel. Also not sure how feedback is routed internally. Also not sure how the filtering is routed.
For the core input channel mixing you could maybe use VenomPoly Fade. It can crossfade between 1 - 16 (polyphonic) chanels. You can use a phasor (or in fact any voltage source, e.g. LFO, audiorate oscillator or sequencer) to control the crossfading (in Silhouette lingo: the ‘spotlight’).
On top of that, Venom Polyfade gives you several other very usefull options relating to the crossfading VCA and phasor behaviour. Also, you get outputs for phasor, gates and envelopes, e.g. to control other modules, to implement the additionally needed functionality.
Thanks @kwurqx & @DaveVenom you guys are the best. The Venom Poly Fade is exactly the right module.
Having so much fun with this. Attached is a patch kinda inspired by the scanning aspect of the Silhouette and by no means an emulation or recreation. Also I don’t have a Silhouette ( yet ) so maybe i even got that part totally wrong but who the fuck cares.
Anyway here you go for anyone interested - it’s setup to process audio from my modular so you’ll have to run something into it, there’s pretty detailed notes and I’ll prob add to it more and will post some updates here when I do. Switch out the modulators and effects to suit your needs. If you do something cool with it or understand the Silhouette please add / fix / update and post back here again, could be fun.
Also note this is almost made entirely of modules that would work on the 4ms Meta Module. I’ll adapt it more later as it’s my goal to get it running there as an effect.
Marsona VCV Euro Silhouette v6.vcv (9.4 KB)
and here’s a little clip of it in action, subtle generative morphing of a dub patch coming from the euro. Marsona VCV patch test
First of all, most thanks should go out to Dave for creating this great module (and many more…).
I did some patching experiments using Poly Fade (but didn’t save any). Especially adding delays and feedback loops.
I hooked up a regular delay for longer delay times back into inputs. Also routed some direct feedback paths (module output directly back into to input). Also hooked up Sckitam Waveguide Delay
SckitamWaveguideDelay is a precision delay (or delay line) (with feedback option). It is more suitable for Karplus-Strong like (tuned) Delay structures or Resonator structures (generally very short delay times and often at very high (near 100%) feedback). Sckitam Waveguide Delay also offers sample delay to compensate for all 1 sample per cable delays in the chain, that would cause detuning if not compensated for (since you effectively lengthen the delay/feedback line with 1 sample for each cable in the out-to-insignal path). Also offers output and control over a pickup point on the Delay line. Also has a Feedback input with phase invert option.
Slightly detuning delaylines is also fun. Chorus/Flanger effects. Sckitam WaveGuide Delay can also precisely control (relative) Phase, since it is a precision delay. Phaser/Comb filter effects.
I guess another fun thing would be to feed in some pretty unrelated audio tracks (in same key and at same tempo) into the inputs and scan though that with various voltage sources, feeding various patterns and rates into the phasor input. I would expect suprising results, especially when manipulating width, rise and fall.
Also, audiorate scanning of various inputs generates interesting results. Blending spectra from the various input signals. Here also, Poly Fade, with its Phasor, and it width and rise/fall options, offers way more control then basic ‘binary’ and ‘triggered’ switching from one source to another.
This audiorate signal switching options somewhat remind me of the concepts behind the ‘Granular filter’ in Klavis Grainity. But…that’s a whole other rabbit hole.
BTW, I also found this long and in-depth video on Silhouette, explaining and demoing it’s features and behaviours in way more detail than the manufacturers demo/explanation.
Amazing! I was playing with the audio rate scanning last night an your idea of feeding it complete different yet related audio is really cool, I’m going to add in some control options for the phasor / width, rise / fall options too. Had some really great results just playing with those settings where the semi repeating loop on my modular was morphed into a sort of percussive pulse kind of like a dub version of SND if you are familar with them. Yes Venom and the other devs get mass respect for their contributions but also very happy you put me in the right direction, I’m fairly new to all this and was feeling stuck so very grateful for the tip as I think this is becoming one of my favorite patches on vcv so far.
Also added in an effects chain with the Stochastic Telegraph Memory modules, they are being triggered by some of those envelope / gate cascades and randomly record and playback snippets of audio. I’ll post here once it’s refined a bit more but it’s pretty fun. Would love to see how you patched up the feedback loops.
Drifting a bit off-topic…towards sampling and granular and general audio mixing and mangling and feedback/delay lines…
Your profile shows plenty of experience with audio/synthesis (including Reaktor an Max/Msp). So, probably not afraid to dive in deep.
This is also reflected in your choice of the Stochastic Telegraph modules for audio recording/buffer manipulation.
A practical tip: Have you tried any of the NYSTHI samplers, especially for more ‘straightforward’ recording/playback?
About feedback, waveguides, delay lines, Karplus-Strong, resonators and such. Basically it is just connecting output back into input, with or without introducing a delay time (either musically tuned or not). Also with or without any other processing in the feedback line (e.g. amplitude dampening and/or filtering).
Maybe have a look at the extensive research by Julius O Smith III at CCRMA at Stanford. He more or less pioneered the field and published a lot on many concepts and implementations.
Julius O Smith III is much a legend and computer/electronic music pioneer as Miller Puckette (author of Max/MSP and Pd).
Julius O Smith III Homepage at CCRMA Stanford: JOS Home Page
The JOS Home Page is a portal into many, many subjects. Just plowing through the Global JOS Index gives you an impression of the extent…
But…we were focusing on waveguides/delaylines and feedback…
You can mostly ignore all the math and just focus on the concepts and the actual implementations. Often visualized in basic block diagrams.
Also, maybe have a look at the Faust Programming Language. Especially click the ‘Try it Online’ button, check out examples and experiment with your own constructions. Many concepts can be (re)constructed in VCV Rack.
The Prototype page indicates “Unavailable”. It also says it only supports javascript and lua (and more languages coming “soon”, but maybe not, if the plugin is no longer supported).
Wandering off topic here, but I did get ModScript to install after building it from source (it’s not in the VCV library), and it provides Lua support.
