Return to Main Index for the CDP System.
Wellspring Music in association with CDP
A graphic program suite complementary to CDP
~ by Richard Orton & Archer Endrich ~
The CDP software devised by Trevor Wishart is often semi-algorithmic in nature. That is, it does more than just make 'a sound', it makes patterns with sounds that unfold over time. The EXTEND program set is a good example of this, as it presents a number of forms of repetition of a sound source. The TEXTURE set is an even better example, for it is an extremely powerful way to play with and balance the random and the defined. ProcessPack consists of a number of modules, a few built on Trevor Wishart originals, and the rest devised by Richard and Archer working together as 'Wellspring Music'. It is graphic software written for PC in TCL/Tk, so a MAC version is feasible.
We have sought to focus on this pattern-making, inspired by research into 'micro-time-forms', trying to increase the role of behind-the-scenes algorithms. A number of modules were completed before Richard's untimely death in February 2013, with an emphasis on the algorithmic dimension. These modules include:
Follow this link for a more complete list of the ProcessPack modules developed between 2007 and 2012.
- DISPERSAL generates multi-events in expanding or compressing shapes, or along specific trajectories
- HOVER implements a Trevor Wishart idea that generates sound by reading back and forth in a soundfile at varying speeds
- PLAYMIX enables soundfiles to be selected and played in an overlapping or sequential manner in real-time, and the result can be saved and mixed down to a single soundfile.
- SHIMMER extraordinary internal sonic movement based on mixing 100 slices of soundfile according to pattern templates
- DRAWSCREEN a variety of shapes can be drawn onscreen and turned into a soundfile
- SONIFY several ways to turn external data into sound
The following section describes DISPERSAL in greater detail. It is the most powerful module in the set, so far, and points in the direction that we would like to go. New collaborators are needed to take this project further. We have decided to make it a purchase rather than a free download at this time in order to help generate some income for CDP in view of the free download of the main system and to support future development.
Dispersal in ProcessPack – A Composer’s View
The ProcessPack functions are designed to realise a musical passage: i.e., one or more phrases or portions of music with an identifiable formal pattern — an interior event sequence in a piece of music — patterning at a micro-level rather than working at the level either of the single sound or of the overall macro-form of the piece. Hence the term ‘micro-form’, or, more fully, for music, ‘micro-time-form’.
A typical micro-form is ‘transition’: a gradual change of state. Others are ‘repetition’, ‘contrast’ and ‘embedding’. There are many, and composers are always creating new ones!
DISPERSAL is based on the pattern, the micro-time-form, of ‘scattering’: taking something and separating it into (many) pieces and dispersing it, as it were, to the four winds. This is more than a (perhaps randomised) segmentation of a sound because it goes on to reassemble the fragments in a specific way. In its classic form, it spreads outwards, both horizontally and vertically, a bit like the mist from a spray can. One can emphasise the edges of the shape or the fill between the edges. When reversed, this process realises another micro-form: ‘convergence’ or a ‘gathering together’.
This is the Dialogue Window for DISPERSAL, showing the large number of parameters that can be selected and combined:
Realising a classic dispersal or convergence would normally be a very time-consuming process. Suppose you want to have 500 segments, each to be cut from the source, perhaps in a certain pattern, then transposed and placed on a time-line according to the overall pattern to be achieved, and finally all 500 components mixed together to produce the output soundfile.
DISPERSAL does all this in a matter of a few seconds, while offering you a staggering number of options for shaping the result: your intended passage of music. This is what we mean by increasing the algorithmic dimension of pattern-making for musical purpposes. The technical description of the DISPERSAL module goes into more detail about these options. In particular, note the options available
The length of the output can also be specified, and this plays an important part in achieving an appropriate compositional role for the result: a concentrated flurry of activity over 10 or 20 seconds, or a more gradual unfolding over e.g., 1 or more minutes! Richard's programming handles the transpositions and time-locations of all the components while finishing at the duration specified by the user.
- to alter how the source is cut,
- to order segment lengths and inter-segment gaps,
- to select among a the variety of shapes, including Wedge In and Wedge Out as well as a number of contour shapes such as Sine or ‘V’ shapes.
The purpose and opportunities of the exercise also differs according to the nature of the source soundfile:
- If the source is spoken words, a concentration of words can then fan out or v.vs. The sense of the text can be maintained if the pieces are cut from the source progressively, from the beginning to the end; or it could be much more fragmented if the cuts are made at randomised positions in the source. Such an explosion or focusing of voices can be full of energy and intensity. More light-hearted results are possible when randomised transpositions and the natural rhythms in the text start to overlap at different speeds.
- If the source is a single steady tone, there may not be much variety from segment to segment. The segments can be made to overlap so that a fairly continuous tone results, clearly delineating the selected shape. Using the double free-draw facility, upper and lower limits can be created and the space between filled with a scattering of segments. By snapping each segment to a user-defined harmonic grid, harmonies can grow from or collapse to a focal point. Similarly, motivic figures or melodies can be put together using the Sequencer function within Dispersal.
- If the source is instrumental, the instrumental tones can multiply and overlap, creating shaped textures.
- If the source is a timbrally complex or noise-based sound, the Dispersal module can be used to create powerful sonic images.
DISPERSAL comes with a comprehensive manual that details the large number of compositional results that are possible, with sound examples and a chart for each one showing its key parameter settings. A discussion of micro-forms and a list of about 80 of them are also included with the manual, to help with thinking at the level of the musical passage, and also as an indication of possible future ProcessPack modules. The graphic user interface is intuitive to use, and parameter settings can be saved to a configuration file and reloaded, for example for use with a different source.
Our aim overall has been to move beyond the sound to the passage of music, automating the process as much as possible while providing an extremely broad and flexible range of shaping options. The other modules are not as comprehensive as DISPERSAL, but offer a variety of enticing sonic possibilities.
Last updated: 30 January 2014