Making Generative Music With Voltage Modular - Volume 3
Making Generative Music With Voltage Modular - Volume 3
Content
Chapter 0 Filling the Toolbox 001
Chapter 1 The Generative Choirmaster -
A Rather Classic Approach 025
Chapter 1.1 No Limits Nor Masks On the Pitch 030
Chapter 1.1.1 Envelope “Glue” & Volume “Glue” 030
Chapter 1.1.2 Rhythm “Glue” 039
Chapter 1.1.3 Timbre “Glue” 043
Chapter 1.2 Limited Pitch Successions 051
Chapter 1.2.1 Choosing Scales 051
Chapter 1.2.2 Pitch Clusters 059
Chapter 1.2.3 Pitch Dependencies 062
Chapter 1.3 Masked Pitch Successions 067
Chapter 1.3.1 Continuous Masking 068
Chapter 1.3.2 Scale Masking 070
Chapter 1.4 The Choirmaster´s Example 083
Chapter 2 Formulas, Formulas, Formulas -
One For Each Musical Element 095
Chapter 2.1 About Limits, Masks And Shapers 099
Chapter 2.2 The Simplest Formula Ever – But In
Multidimensional Spaces 106
Chapter 2.3 The Sonic Beauty of Mathematics 115
Chapter 2.4 The Sonic Aesthetics of Curves,
Lines And Coloured Areas 123
Chapter 2.5 A Sonic Game of Domino 137
Chapter 2.6 Layering Algorithms 147
Chapter 3 Phase Shifts, Event Delays, and
Coincidences 153
Chapter 4 Generative Evolutions -
From Musical Protozoa to
Creative Monkey Business 177
Chapter 5 The Probability Theorist´s Paradise -
About Weak, Strict, Clear and Diffuse
Masking Techniques 184
Chapter 6 Musical Wall Papers -
Thinking in Colours and Pictures 224
Chapter 7 Driving Without a Navigation System -
Let Inspiration Flow 247
List of Paintings 263
Appendices 264
Some excerpts of the chapters:
Chapter 0
Filling the Toolbox
When you hear the term “Block diagram” in connection with modular synthesizers you probably think of something like this.
There are the “usual suspects”, common building blocks of modular synth patches like VCO, VCA, ADSR, VCF, GATE etc.
Those diagrams are quite helpful when documenting a patch, or even when planning a patch.
But when it comes to composing a patch, composing a piece of music for modular synthesizers – not only generative music of course – then we need a higher level of abstraction to begin with.
Why is that?
Let´s take a VCO for example. A VCO is a sound source – sometimes. But sometimes it is also a modulation source – think of audio rate frequency modulation.
But there are more sound sources than just VCOs. There are NOISE modules, there are fast looping envelope generators (or function generators), and there are even whole networks, which as a whole fulfil the function of one sound source.
The idea of a network, which contains VCOs but must be seen as a sound “unit”, with the contained VCOs not being handled or understood as sound sources, needs an explanation.
Let´s look at the following diagram.
The sound source – from the view of a composer – is the whole orange block. It´s not VCO 1, because the audio coming from VCO 1 is not used in the rest of the patch. It´s not VCO 2, because VCO 2 doesn´t produce its outcome alone. It´s the whole block that must be seen as one single sound (audio) source, with its output being used as audio (and not as modulators shapers etc.).
The block delivers the audio that is going to be further used, e.g. for building melodies or what ever. The elements and goings on inside the block belong to the realm of sound design – not of composing.
Designing the timbre, the sound is – of course it is – a part of the work of the composer/producer of music for modular synths. But on the highest level of abstraction, when starting to develop the idea of a piece, I may draw or write something about which kind of sound/timbre I want, but not how I create this kind of sound/timbre. That´s going to be a job for later.
Alright then, what are these “first level” building blocks of a composition then? Which of these components do we have to put into our toolbox?
I´ve thought out some identifying pictograms for each of them:
The Fix Level Source can be a DC offset, or a VCA which is not modulated, or an unmodulated mixer channel strip etc.
Each of these blocks can be a single module or a network of modules, and all of them, but for the Fixed Level Source, can physically contain one or more of the others, which then are no longer seen as a block, but as components of a block, which – on this most abstract level of composing – don´t receive any further attention. The “modulators, shaper, etc.” in the picture on page 4 are such an example.
I have to mention two last terms before I can show a practical example of this all. These terms are “primary output” and “secondary output”.
From a primary output signals are send out of these blocks, whereas the signals from a secondary output are only used inside a block. The picture shows the two secondary outputs and the one (there are more than only one sometimes) primary output in the structure that is shown in the picture on page 4:
Let´s have a look at a practical example at last. ….........
Chapter 1
The Generative Choirmaster -
A Rather Classic Approach
The four-part chorale setting is probably well remembered by every former music student as the professor´s favourite torture instrument almost at the beginning of their studies.
The term “generative polyphony” seems to oppose nearly everything that we have been told to do in those younger days of ours. Isn´t “generative polyphony” rather like telling a couple of musicians (let´s say 4 to stick to the example)
“You can do whatever you want as long as you all do it at the same time”?
So what? Everything is allowed in art, isn´t it? Even something like the patch called “chapter 1 preset 1”.
But for how long are you willing and able to listen to the video of this patch:
And how often would you return to it, and listen to it again and again (I mean without some chemical help)?
Have we to say “good bye” to generative polyphony then? Of course not!
But the generative aspect of a polyphonic piece has to accept either some limitations - or on the contrary some add-ons. Let´s proceed systematically and nicely step by step.........
Chapter 2
Formulas, Formulas, Formulas
-
One for Each Musical Element
This compositional technique or approach is completely different from everything which is shown and written in chapter 1.
It´s not about pitch, it´s not about polyphony – even if we can (and will) apply it to polyphonic parts, it´s not about voices and their coexistence.
The technique (or call it an approach) is far from being classic or old – and at least in my time nobody taught it at any music school.
We need a completely different way of thinking and planning – and composing – here.
We don´t even have to see it as something which is related to music in the first place.
And the word “one” in the title should not be taken too literally.
The formula approach can be applied to nearly everything:
to music (luckily – otherwise I couldn´t write about it here),
to performing arts like dancing and playing theatre,
to painting, to sculpturing, to fashion, and to whatsoever.
Some say, you can even apply it to your way of living itself.
Wow! If that is not a big ass introduction to a chapter, then there isn´t anything “big ass” at all!
Hey, Rolf! Come down to Earth again, and return to serious writing!
To make it absolutely clear:
This type of producing generative music is about composing layers of modulations, where our main focus is the structure of these layers in the first place, and the targets of the modulations only take secondarily our attention – not saying no attention......
Chapter 3
Phase Shifts, Event Delays
and Coincidences
The compositional approach, which this chapter 3 is going to deal with, focusses on the temporal distribution of – not only sonic - events and processes, as well as on the effects of different durations of similar processes, and coincidences.
A rather simple example shall introduce the matter. The patch is based on one and the same melody being played at 3 different speeds – in other words: it lasts differently long. The sonic events, which I mentioned above, are pitches therefore. The principle is as old as electronic music itself. Not very interesting so far.
But there´s an additional twist in the patch: always when two identical pitches are played in versions 1 and 2 at the same time, then the playback speed of the otherwise slowest running melody (version 3 in the lower cabinet) changes to that of the fastest running melody – but only as long as the pitches of versions 1 and 2 are identical.
And what´s more: the speed of versions 1 and 2 randomly change (Sample & Hold module).
The patch is called “chapter 3 preset 1”. The Manual Triggers module lets us reset the 3 sequencers as well as the 3 clock LFOs, so that all start at the same time.
The sequencer of version 1 of the melody (upper cabinet on the left side) is clocked at a rate of 1.0Hz.
The sequencer of version 2 (upper cabinet next to version 1´s sequencer) is clocked at a rate of 1.2Hz.
The sequencer of version 3 (lower cabinet) is clocked at a rate of either 0.8Hz or 1.2Hz.
But the clock LFOs of the sequencers of melody versions 1 and 2 are rate modulated by the Sample & Hold module in the lower cabinet.
Let´s have a look at the mechanism that switches version 3 between 1.2Hz and 0.8Hz. My weapon of choice is the Comparator module in the lower cabinet......
Chapter 4
Generative Evolutions -
From Musical Protozoa To Creative Monkey Business
This compositional approach starts with a single, and rather simple, musical idea, the stem cell, but builds up mechanisms, which develop this (simple) idea to more and more complex patterns and sonic goings on.
Opposed to the approach in chapter 7 these mechanisms are planned in advance and part of the compositional concept. There is no improvisational aspect in the process of patching.
It all starts with a simple 4-note motive:
G3-C3-Bm2-C4.
This is my “stem cell”, which shall develop into the octopus (the one in the picture of the last page :-)).
The structure is as simple as the motive itself: Sequencer into Partial Oscillator, audio through Chorus and Digital Reverb (there are 3 different reverbs in the patch – the Digital Reverb is called “reverb 3” here).
But I develop this “stem cell” across more than only one sonic dimension, not only in the pitch domain, but also including sonic growth into different timbres, into noises, into effects, into dynamic dimensions etc........
Chapter 5
The Probability Theorist´s Paradise
- About Weak, Strict, Clear and Diffuse Masking Techniques
What´s that?!
Isn´t this topic the same as the topic of chapter 2?
And haven´t I already talked about masks and limits – e.g. on page 9, and then in chapter 2 of course?
No and yes. No, it is not the same, and yes, I have talked about masks and limits.
But this chapter 5 is about the RELATIONS of differently masked or limited MUSICAL PARAMETERS, and about the (kind of) rules and (surely given) facts we have to consider when composing these relations.
Or in more artful words: it is about painting a picture across all available sonic parameters, a multicoloured picture, in which the interplay of colours represents these relations.
It´s not only about randomizing frequencies (and I intentionally don´t use the term “pitch” here), but about randomizing everything (timbres, rhythms …) all at the same time, with masks and limits taking care that there is musical sense and meaning in the whole picture.
A (perhaps much too) simple example:
The Sample & Hold modulation in the patch called “chapter 5 preset 1” is meaningless, it doesn´t make any sense. The high pitches in combination with the rather high cutoff frequency make any modulation of the resonance (“emphasis”) toothless.
Well, let´s modulate the cutoff as well – but in the opposite direction (it´s only for reasons of clarity that I use the Invert module instead of simply turning the modulation depth knob to the left). The patch “chapter 5 preset 2” shows the result.
There is a kind of effect audible now, but it´s the occurrence of a (probably not intended) rhythm, not a change of the timbre.....
…..Another important thing to be aware of is the fact, that not all parameters react equally strong to a certain change of CV, and even if they did, we won´t always perceive the effects as equally strong......
Chapter 6
Musical Wallpapers
- Thinking in Colours And Pictures
The second part of the chapter title is to be taken literally. In this chapter I´m going to introduce a special way of approaching generative compositions, a way which starts with a – mostly abstract – picture in my mind.
The picture which initiated the following patch example is shown above: slowly mixing colours in a glass of water with bright, colourful star-like speckles.
There are two groups of colours in the patch. Each group consists of 4 sound sources. Each of these groups has a sonic character of its own.
Whereas group 1 is rather calm and solid like real (quite dark) ink from an inkjet printer, the second group is more shiny, even a bit incoherent, and each of the 4 colours is permanently changing its hue – like with my grandchildren´s set of watercolours.
Each group plays the same intervals, but based on different pitches, which are 7 semitones apart.
Group 1 plays intervals consisting of 2 randomly chosen notes, which only be C, Eflat, Gflat and A. Possible intervals are therefore C-Eflat, C-Gflat, C-A, Eflat-Gflat, Eflat-A, Gflat-A.
The notes on offer for group 2 are G, Bflat, Dflat and E.
There are 4 Quantizers, all of which have these identical settings, those of group 1.
Each quantizer gets its input from a Sample & Hold module of its own. These Sample & Hold modules are triggered by an LFO each. The LFOs run at rates of 0.4Hz, 0,5Hz, 0.6Hz and 0.7Hz.
Each of the 4 Quantizers sends its pitch CV to its own VCO.
The left most Quantizer sends its pitch CV to a Vintage Oscillator-Ladder Filter block in voice group 1, as well as to the spectral oscillator CHEBZ in voice group 2.
The second Quantizer (counted from left to right) sends its pitch CV to a Partial Oscillator in voice group 1, as well as to a WARPSQUARE oscillator in voice group 2.
Quantizer number 3 sends its pitch CV to a Soft Oscillator in voice group 1, as well as to the left of two Super Oscillators-Filter blocks in voice group 2.......
Chapter 7
Driving Without a Navigation System
- Let Inspiration Flow
This chapter is not about live patching – but it´s not far away from it. Let me call it “live patching with yourself as the only audience”.
Sounds cryptical, doesn´t it?
OK, another try.
This chapter is about building up generative patches without a lot of prior planning and drawing.
No, Fred. Not that way!
Last try then:
This chapter is about having only a rough idea of what the patch shall do.
It is about being prepared to change these rough and rather blurry plans in the course of patching.
It is about being open for leaving the path of the next intended steps, and rather following sudden new flashes of inspiration.
It is about “Wow! That sounds great too!”
It´s even about “Let me try what this module can do, this one, which I bought ages ago, but never really used.”
And it is about doing all this without getting lost in endless fiddling around, never finding an end, until your wife, girlfriend, husband, boyfriend, partner – or all of them together – enter your studio saying unisono:
Even though Fred looks like a prince who got thrown against the wall a few too many times when he was still a frog, he´s finally got it! Yay!
But now it´s me who has a problem. How shall I write something that has not to be written down, not to be planned, not to be drawn?
Once being written, it is no longer spontaneous.
I think I´ll write this chapter in the form of a case study, a field report or something like this.
In other words, right now, as I am writing this very sentence, I don´t know, what I´m going to do and write next. I´ll simply start doing and writing.
A rough idea is not no idea at all, and a vague imagination of how a page shall look like is not no imagination at all.
Let me lean back for a while and think about it......
… well, what about a probabilistic bass line, let´s say 16 sequenced notes, each 5th or 6th of which may randomly change from time to time, but only within a range of 1 octave, and only to a set of possible quantised notes, which belong to a parallel scale.
I imagine some strongly LP filtered bass sound.
And above that a completely random kind of melody, which is not generated by a standard VCO, but occurs from filtered resonant noise. Here I imagine quite a lot of reverb.
Let´s try something........