The "wiggle" is actually something all *analog* electronic oscillators do when outputting a square or saw tooth wave. In the electronics world it's called "ringing" or "edge ringing". In the physical world it's caused by both the inherent capacitance within the oscillating transistor itself and the inductance (think of it as moving electron inertia) in the copper traces connecting the transistor. *The VCV VCO, in analog mode, actually includes and simulates this "edge ringing" fairly accurately.* The interesting thing, and something VCV does NOT simulate, is that no two transistors, even if they are the same series number, have the exact same "ring" parameters because of slight chemical differences that occur in the manufacturing process - even for transistors made on the same wafer. Your "sampler" is just operating at a high enough frequency and small enough fractional ratio of the sampled wave frequency that it can detect and output the values of the simulated ring oscillations. The "glide" or "plateau decay" is caused in the physical world because the oscillating transistor heats up and cools off with each oscillation causing a minute change in it's internal resistance, changing the output voltage over the duration of each cycle. It can also be a result of "leaky" capacitors in the resonant tank circuit ... or a little of both. If you switch the VCV VCO to "digital" mode, the ring is reduced and the slope will go away. I'm not 100% sure why the ringing doesn't go away completely in digital mode since the wave form is not actually being generated by an oscillating transistor, but by an algorithm that mathematically calculates the values of the wave form (in both modes). It's a credit to the developer of VCV that this subtle physical nature of electronic circuitry is included and simulated in the VCV modules. The reason third party oscillators in VCV don't have the edge ringing is probably because the people programing the modules are not electronic techs and don't realize that it exists. It is extremely difficult to filter out of high frequency audio and radio signals and exist in nearly all real world signals to some degree. With this "discovery" you have stumbled on to a very important aspect of Electronic Warfare ... which I will NOT explain.
The actual ringing usually comes from inevitable limit of bandwidth in analogue circuits, so even if very precise, very wide-band signal generator would generate a perfect square wave with hundreds of MHz of overtones, as soon as this signal travels through regular cables and comes into input filter of a normal ADC, it will be limited to 20 kHz band before digitization. And we’ll see these wiggling around every fast slew because if signal won’t be limited below Niquist frequency, we will get parasitic sidebands aka aliasing. At the same time, when you look onto the same signal through good oscilloscope with properly compensated probes, like mine, 400 MHz analog Iwatsu, there are no visible ringing as long as you observe it at reasonable vertical zoom.
The noise within a system is fundamental to all signal based mediums and can be referred to as the latent space. I used to experiment with video feedback, and recognized that the patterns that seemingly emerge from nothing were an amplification of that subtle noise within the circuits, and that depending on the chain of devices, you will get different noise characteristics due to each device having its own signature. More recently I've been working with generative AI image tools, and similarly, at the base of it is literally what is termed a latent image, which is simply a noise pattern. I believe our own cognitive processes work in the same way. When we look at clouds and see forms, stare at waves and let our mind wander, or gaze into a fire in order to transport us into our imagination. This fundamental relationship of ordered systems that filter a noisy signal I think is integral to the nature of things.
This is amazing. I'm new to VCV Rack, but this is exactly the level at which I like to learn about different creative tools, examining the ghost in the machine. Thank you so much for sharing!
Just a heads-up: You can instead use a second VCO with same freq as the "alias" one as a trigger in the Sample & Hold, so you don't need to type out the sampling rate.
Another thought-provoking video, very interesting! The Shapemaster playheads idea was particularly impressive, partly because it's very efficient. Why run multiple sequencers when you can run one? :) I'm picturing the time dimension like a physical one, and you'd see a pattern laid out on a terrain, with diagonal slices across it. So beautiful.
Thank you! You just introduced me to a whole new way of thinking about sequencing.
Not only do you shine light on obscure patches and techniques, you do a great job of explaining! Thanks Jakub!
The "wiggle" is actually something all *analog* electronic oscillators do when outputting a square or saw tooth wave. In the electronics world it's called "ringing" or "edge ringing". In the physical world it's caused by both the inherent capacitance within the oscillating transistor itself and the inductance (think of it as moving electron inertia) in the copper traces connecting the transistor. *The VCV VCO, in analog mode, actually includes and simulates this "edge ringing" fairly accurately.* The interesting thing, and something VCV does NOT simulate, is that no two transistors, even if they are the same series number, have the exact same "ring" parameters because of slight chemical differences that occur in the manufacturing process - even for transistors made on the same wafer. Your "sampler" is just operating at a high enough frequency and small enough fractional ratio of the sampled wave frequency that it can detect and output the values of the simulated ring oscillations. The "glide" or "plateau decay" is caused in the physical world because the oscillating transistor heats up and cools off with each oscillation causing a minute change in it's internal resistance, changing the output voltage over the duration of each cycle. It can also be a result of "leaky" capacitors in the resonant tank circuit ... or a little of both. If you switch the VCV VCO to "digital" mode, the ring is reduced and the slope will go away. I'm not 100% sure why the ringing doesn't go away completely in digital mode since the wave form is not actually being generated by an oscillating transistor, but by an algorithm that mathematically calculates the values of the wave form (in both modes). It's a credit to the developer of VCV that this subtle physical nature of electronic circuitry is included and simulated in the VCV modules. The reason third party oscillators in VCV don't have the edge ringing is probably because the people programing the modules are not electronic techs and don't realize that it exists. It is extremely difficult to filter out of high frequency audio and radio signals and exist in nearly all real world signals to some degree.
With this "discovery" you have stumbled on to a very important aspect of Electronic Warfare ... which I will NOT explain.
Thanks for the explanation!
great writeup, thanks.
I’d love to read a summary on the “electronic warfare” you mentioned….
The actual ringing usually comes from inevitable limit of bandwidth in analogue circuits, so even if very precise, very wide-band signal generator would generate a perfect square wave with hundreds of MHz of overtones, as soon as this signal travels through regular cables and comes into input filter of a normal ADC, it will be limited to 20 kHz band before digitization. And we’ll see these wiggling around every fast slew because if signal won’t be limited below Niquist frequency, we will get parasitic sidebands aka aliasing.
At the same time, when you look onto the same signal through good oscilloscope with properly compensated probes, like mine, 400 MHz analog Iwatsu, there are no visible ringing as long as you observe it at reasonable vertical zoom.
The noise within a system is fundamental to all signal based mediums and can be referred to as the latent space.
I used to experiment with video feedback, and recognized that the patterns that seemingly emerge from nothing were an amplification of that subtle noise within the circuits, and that depending on the chain of devices, you will get different noise characteristics due to each device having its own signature.
More recently I've been working with generative AI image tools, and similarly, at the base of it is literally what is termed a latent image, which is simply a noise pattern.
I believe our own cognitive processes work in the same way. When we look at clouds and see forms, stare at waves and let our mind wander, or gaze into a fire in order to transport us into our imagination. This fundamental relationship of ordered systems that filter a noisy signal I think is integral to the nature of things.
This is amazing. I'm new to VCV Rack, but this is exactly the level at which I like to learn about different creative tools, examining the ghost in the machine. Thank you so much for sharing!
you're a mad scientist, dude; these vcv hacks are the BEST! -i appreciate all your time and research.
Just a heads-up: You can instead use a second VCO with same freq as the "alias" one as a trigger in the Sample & Hold, so you don't need to type out the sampling rate.
Great idea, I was thinking how this could be done in hardware domain / eurorack... and your idea is solving the "typing" issue. 😊
It fun and interesting how VCV handles these little tricks you're playing with. Good stuff.
Great video Jakub, you're out of the box thinking is very inspiring, and this actually made sense too!
Oh God this is really something to try had so fun trying this
Nothing magical but I really like this approach, thanks. Also this is very useful for analyzing audio signals.
Another thought-provoking video, very interesting! The Shapemaster playheads idea was particularly impressive, partly because it's very efficient. Why run multiple sequencers when you can run one? :) I'm picturing the time dimension like a physical one, and you'd see a pattern laid out on a terrain, with diagonal slices across it. So beautiful.
INCREDIBLE!!!
Great job on the insights, an amazing tutorial. Thanks for sharing.
Your videos are always amazing and arcane.
Wow! What a patch!
Excellent tutorial
yeah ... everything is around voltage ... 💙
Big Applause 👏👏👏
This is birds screams on the beach and it's under many seconds reverb effects, try to use sample and get the same result faster quickly
mind blown... :)
❤
dobre
This is deep
Jakub, you are the David Blaine of VCV Rack! In that ShapeMaster demo, I felt just like Harrison Ford here :-) ua-cam.com/video/rB0wzy-xbwM/v-deo.html