LSSP Patches in Practice
In practice the exact form of an LSSP patch depends on what you want to achieve. This open-ended approach is of course the beauty of modular systems.
At one extreme a simple setup might use a subset of the functionality and consist of just a handful of modules that patched together operate like a neat little drum machine or a simple bass line sequencer for instance.
At the opposite extreme, systems that implement entire symphonic scale compositions can be built providing that you have the massive computing power required to use hundreds of modules simultaneously and of course the ingenuity and patience to build such large scale patches.
Depth and Freedom
Deep exploration, technical creativity and a certain amount of serendipity are arguably what enthusiasts find so attractive about modular synthesis in the domain of sound design. LSSP attempts to extend this modular philosophy to large scale musical composition.
LSSP is not a simple system, it’s designed to model the internal structure of musical compositions and music isn’t always the simplest of things.
On the other hand everything possible has been done to make the modules straightforward and easy to use. Some novel terminology is introduced here and there but only out of necessity – the goal has always been to keep things transparent and simple. LSSP modules have generic names that reflect their purpose as directly as possible.
LSSP does not make any assumptions about the genre you would like to work in. This means that everything is kept as flexible as possible. The cost of this freedom and sophistication is some complexity but the benefit is that you should be able to use LSSP to work in just about any genre imaginable from cliché pop to extremely experimental generative electronica that uses micro-timing, polyrhythms and micro-tonality.
Unlike a piano roll editor where melody, harmony, rhythm and form are all entangled; LSSP attempts to expose these elements to independent manipulation by making the underlying musical structure explicit. This opens up fantastic possibilities but requires a deeper level of engagement than a system based on a piano roll (or a series of CV step sequencers).
I use the analogy of a graphics editor. In principle a program that enables you to set the color of any pixel in an image is extremely easy to understand and is capable of producing any image. But in reality this conceptual simplicity is of little help when you are trying to create complex shapes and shading.
Piano roll editors have the same conceptual simplicity and are easy to use but they offer little or no assistance in creating music that has real-world structure built from scales, chord progressions and large scale form. Such structural content is only implicit.
So, one of the key concepts in LSSP is disentanglement with Melody Sequencers handling melody, Rhythm Sequencers handling rhythm, Progression modules handling chord progressions, Scale modules handling scales, Chord modules handling chords, Groove modules handling micro-timing and dynamics, CV Sequencers handling CV modulation, Song Control and Song Part modules handling large scale structure and time-splitting modules handling intermediate level form.
Sophisticated interaction between these modules enables these essential elements to be independently manipulated yet coalesce into a whole in a manner that transforms the way you can create music.
Scale and Complexity
Small scale patches of LSSP modules can be useful and you don’t need to build a new patch from scratch for each new composition – moderate sized LSSP patches include sufficient flexibility that you can create a wide variety of compositions using the same arrangements of modules.
But one of the goals of the LSSP architecture is to enable large scale projects to be constructed that model complete compositions with multiple instruments. Perhaps using hundreds of modules and taking weeks to build.
To this end…
- The extendable nature of the Song Control / Song Part pairing means there’s no practical limit to the length of a song.
- There are limitations in the time splitting module set that make it difficult to build song parts that use sequencer chains longer than 32 sequencers (or 512 steps), but as you can have as many song parts as you like then it’s not a real issue. In practice song parts rarely need to be longer than 16 bars.
- V/Bar sequencing is very efficient – dormant Adroit sequencers use almost no CPU.
So, we are left with two major limits. The computing power required to deal with a large amount of polyphony and the problems that come from managing complexity in very large patches.
The CPU load when a large number of simultaneous voices is required is an issue that will slowly be solved by more and more powerful computers. In the meantime one way to address this is to offload some sound generation to external hardware. This is one of the reasons why some Adroit modules concentrate on MIDI interfacing.
The more difficult problem is dealing with complexity. Large scale patches look scary and become brittle and difficult to modify. When you need to scroll even when fully zoomed out then perhaps you’ve hit the limit of practical patch size.
Ultimately the only way to go further is to use hierarchical patches. Think “black boxes”. Then boxes embedded within boxes. This encapsulation requires certain organisational skills but is a well proven technique for managing complexity.
The down side is that encapsulation can be demanding not just on CPU processing power but also on the amount of memory needed to stop everything grinding to a snail’s pace due to memory swapping. Still with a reasonably powerful computer a lot is possible.
The Managing Complexity page discusses some techniques for constructing large scale projects and the Performance Tips page discusses how you might optimise the settings in Voltage Modular in order to achieve maximum efficiency.
The tutorials and extensive documentation on this website should enable you to eventually fully grasp the technical ideas behind LSSP but the creative potential of LSSP takes time to adjust to because it offers a radically different workflow – one based on the organisation of large scale musical structure.
This is normally the domain of a DAW but LSSP gives you the option of bringing things like song structure, chord progressions, motif form and complex interaction between harmony, melody, rhythm and groove firmly into the modular world.
LSSP is deep. Although you should be able to get plenty of near instant gratification there’s a lot of potential that will take months of exploration to uncover. At one level this is because of the technical inadequacies of the design but at another it’s simply a reflection of the fact that composing music is far from straightforward.