Significant Acousmatic Works
- James, S. (2016). Particle III for solo laptop and 16.1 surround loudspeaker array.
- Hope, C. & James, S. (2015). Feather for 24.4 surround sound (3D Ambisonic).
- James, S. (2013). The Overview Effect for Laptop (24.4 channel).
- James, S. (2012). Veden Ja Tullen Elementit (8-channel)
- James, S. (2001). The Realisation that was Everything and that Nothing Ever Was.
- James, S. (2000). Terminal Voltage.
Particle III (2016-7)
|
For Solo Laptop and 16.1 channel sound system
This work explores examples of spatial motion as described by Trevor Wishart, such as harmonic motion, en masse. Certain psychoacoustic effects are explored through the medium of space, such as an effect I describe as 'spatial beating', due to the spatial movements being modulated at frequencies below or within the lower audible frequency range (0-40Hz). These LFO-like effects are experienced as a spatial shimmering or pulsation of the sound source. This work largely explores spectral spatialisation, and involves the spatial positioning of spectral bands at the micro-level (audio-rate). The images show instantaneous spatial positions of different spectral bands colour coded according to their frequency. The work also explores temporally disconnected spatial displacements of frequency bins, and other bizarre and strange effects resulting through numerical modifications and morphologies/distortions of the coordinates of frequency in space. All of the processes explored in this work involve sound shapes where all spectral bins have independently computed trajectory curves. |
|
Feather (2015)
A collaboration with Australian composer Cat Hope
The Overview Effect (2013)
For Solo Laptop and 24.4 channel sound system
Veden Ja Tullen Elementit (2012)
For Solo Laptop and 8.1 or 16.1 channel sound system
While in residency at the SpADE facility at DMARC, University of Limerick, my focus shifted towards applying timbre spatialisation from found sounds to environmental noises rich in harmonic content. These proved to be effective when diffused across a multichannel system. The exploration of different trajectories had led to me discovering the immersive effects of high-frequency spatial texture by using a range of asynchronous and noisy trajectories for timbre spatialisation. The immersive effects reminded me of natural environments and ecologies where the source of sounds appear to be coming from many different directions around the listener, with no clear localised point of directivity.
This experience in Ireland led to Veden Ja Tulen Elementit (2012) (the elements of water and fire), an acousmatic work that explores several different approaches to spatialisation practice. The composition began largely as an experiment in sound design, in virtual immersive environments and spaces. It was inspired by the listening modes discussed by Pierre Schaeffer (1977), Michel Chion (1983), Barry Truax, Katherine Norman (1996), David Huron (2002), and Smalley (2007). This piece, like other acousmatic music, encourages what Schaeffer (1977) describes as the reduced listening mode, exploring various kinds of immersive sound shapes generated through timbre spatialisation.
The work was presented at the 2013 ICMC.
Two environmental source sounds were used: a rippling river and burning embers. The premise of the work was to begin with the sound of the babbling brook, spatialising this immersively as to encourage a referential listening mode. With an interest in realism and hyper-realism, the opening is intentionally aimed at creating a real sense of place. However, the piece is not intended to be a soundscape, and as I have an interest in the surreal, very quickly this familiar environment begins to distort into the unfamiliar. As a composer, I cannot say I was interested so much in what these sounds represented for the listener, but in taking the listener from the familiar, and gradually and increasingly towards an unfamiliar sound world. The opening sound sample is processed in accumulating and more abstracted ways, gradually shifting this sound away from its original associated references. The piece concludes when increasingly processed water samples finally undergo a metamorphosis into the samples of the burning embers.
This work related to a concept of the phantasmagorical written about by Curtis Roads (1996) with respect to physical modelling synthesis. Some physical models can extend real-world possibilities to dimensions of the absurd: for example, a musical instrument that involves plucking the cable wires of the Golden Gate Bridge in San Francisco. In the real world these are reaching impractical possibilities; however in the algorithmic world we are not bound by these constraints. Similarly in sound spatialisation, we have spatial movements that would be impossible in the natural world, and rather than avoid them, we can explore these as further possibilities of expression above and beyond what is possible given what is possible in the physical world.
The work was conceived on the SpADE speaker configuration shown below, so was written with the elevation of sounds in mind. As this was implemented before 3D timbre spatialisation was possible, the work focuses on the use of two different circular systems: a lower and upper array of speakers.
This experience in Ireland led to Veden Ja Tulen Elementit (2012) (the elements of water and fire), an acousmatic work that explores several different approaches to spatialisation practice. The composition began largely as an experiment in sound design, in virtual immersive environments and spaces. It was inspired by the listening modes discussed by Pierre Schaeffer (1977), Michel Chion (1983), Barry Truax, Katherine Norman (1996), David Huron (2002), and Smalley (2007). This piece, like other acousmatic music, encourages what Schaeffer (1977) describes as the reduced listening mode, exploring various kinds of immersive sound shapes generated through timbre spatialisation.
The work was presented at the 2013 ICMC.
Two environmental source sounds were used: a rippling river and burning embers. The premise of the work was to begin with the sound of the babbling brook, spatialising this immersively as to encourage a referential listening mode. With an interest in realism and hyper-realism, the opening is intentionally aimed at creating a real sense of place. However, the piece is not intended to be a soundscape, and as I have an interest in the surreal, very quickly this familiar environment begins to distort into the unfamiliar. As a composer, I cannot say I was interested so much in what these sounds represented for the listener, but in taking the listener from the familiar, and gradually and increasingly towards an unfamiliar sound world. The opening sound sample is processed in accumulating and more abstracted ways, gradually shifting this sound away from its original associated references. The piece concludes when increasingly processed water samples finally undergo a metamorphosis into the samples of the burning embers.
This work related to a concept of the phantasmagorical written about by Curtis Roads (1996) with respect to physical modelling synthesis. Some physical models can extend real-world possibilities to dimensions of the absurd: for example, a musical instrument that involves plucking the cable wires of the Golden Gate Bridge in San Francisco. In the real world these are reaching impractical possibilities; however in the algorithmic world we are not bound by these constraints. Similarly in sound spatialisation, we have spatial movements that would be impossible in the natural world, and rather than avoid them, we can explore these as further possibilities of expression above and beyond what is possible given what is possible in the physical world.
The work was conceived on the SpADE speaker configuration shown below, so was written with the elevation of sounds in mind. As this was implemented before 3D timbre spatialisation was possible, the work focuses on the use of two different circular systems: a lower and upper array of speakers.
The rendered work consists of a series of audio stems, shown in the following table, are mapped to specific designated speaker channels at SpADE.
The first stage of this composition, given the limited source material, was to build a corpus of sounds through various signal processes in the frequency domain and time domain. The manipulation and distortion of the source sounds were achieved using a range of tools including MaxMSP 6, Metasynth 5, PaulStretch, and Izotope RX.2. Once working with these tools, the compositional intent quickly shifted from simply manipulating these sounds to processes that allowed for the metamorphosis and transformation of them. The CataRT software is a real-time implementation of corpus-based concatenative synthesis and was one way of deconstructing the source sounds into segments, and finally reconstructing them based on their spectral similarities. This became one of the most widely used tools for exploring this morphology of the original found sounds.
Once this corpus of sound had been derived, it was possible to view the scope of this sound world. Considering the corpus as a collective, it was necessary to decide on a structural arrangement of these different sounds. The audio files were imported into Logic Pro and arranged on the timeline allowing for immediate audition and evaluation of the structure. These audio layers were subsequently mixed with some level of automation.
In the almost 18 months between the writing of this piece and the explorations made in Particle 1, significant software development and investigation into performative approaches of timbre spatialisation was undertaken using Model A. This investigation involved the exploration of various kinds of terrain and trajectory combinations, and later explorations tended to favour the use of circular trajectories, rather than the linear trajectory used in Particle 1, allowing for a more perceptible correlation between the Wave Terrain Synthesis and timbre spatialisation processes. Nevertheless there remained some inherent flaws. One of these was that the highest frequency bins of one spectral frame naturally would spill into the lowest frequency bins of the next, creating an effect very reminiscent of the Shepard tones: an infinitely ascending or descending spectrum. Although this can be a desirable effect, it is not indicative of the features presented in the terrain or trajectory structures, and therefore does not promote the concept of a tight and intuitive semantic relationship between this control structure and the sound shape generated. The question ultimately boiled down to aesthetic interpretations, and whether the terrain could be a visual indicator for timbre spread across space and affect the spectral qualities generated across space. Early approaches to Model B take into consideration that local colours of the terrain surface affect the audible colour spectrum of their nearest speaker.
Although the earlier pieces utilise the mapping strategy in Model A, it was a visit to the ICMC in Slovenia in 2012, and preparation for a lecture I gave at the Naples Conservatory of Music that ultimately led me to pursue the histogram method. Just short of these new developments, I had the opportunity to test the software at the SpADE facility at DMARC. Here it was possible to examine the effectiveness of timbre spatialisation using the WTS control strategy over a 16- and 32-channel loudspeaker configuration. It was with this setup that the detail of spatialisation and immersiveness became apparent. Up until this time, most of the source sounds I used for testing purposes were restricted to static oscillators and noise generators. This was because I was interested in establishing various archetypes of sound shapes possible, as well as exploring the morphology of these shapes by manipulating the terrain and trajectory structures. Navigating between various immersive states could be controlled easily and effectively as I began to explore interpolating between different trajectory states as a means of morphing the sound shapes generated.
Although the sounds spatialised in the lab were invariably mono sources, the results were evolving, engaging and immersive environments. These experiments involved varying degrees of random distribution of spectra across SpADE’s 32-channel speaker configuration while experimenting with natural found sounds I had recorded. Timbre spatialisation appears to be extremely effective in a 16-channel environment for creating immersion, despite being an artificial and abstract recomposition of an auditory scene. Normandeau (2009) describes this phenomenon as “specific to the acousmatic medium is its virtuality: the sound and the projection source are not linked” (p. 278).
Normandeau (2009) also states that timbre spatialisation recombines the entire spectrum of a sound virtually in the space of the concert hall, and is therefore not a conception of space that is added at the end of the composition process, an approach frequently seen, but a truly composed spatialisation. Although the structure and basic mix for Veden Ja Tulen Elementit were in progress, I was considering the kinds of spatial treatments I would apply to specific layers of the piece. As I had already heard some of these sound shapes generated at DMARC, it was not so difficult to imagine these effects when applied to the sounds generated.
The piece explores these immersive spatial attributes—envelopment, engulfment, presence and spatial clarity using a variety of movements from slow circular correlated motion, sound slowly unfolding through space and slow fractal dispersion to fast chaotic movement and the random dispersion of sound. I found fusing standard point source approaches worked well against the more complex, evolving and immersive sound shapes generated through timbre spatialisation. This allowed the spatialisation to have momentary points of focus, influenced by the multi-levelled space forms and listener zones discussed by Smalley (2007). In this way the timbre spatialisation generates a circumspectral and immersive sound scene, from which other sound shapes and point source sounds emerge within. Complex sound shapes were also created using fractal terrains and trajectory curves, creating some very abstract and unpredictable sound shapes resulting in low- and high-frequency spatial texture. This asynchronous quality of trajectory curves produced irregular evolving and fluttering movements in the resultant sound shape. The Rössler strange attractor, a continuous differential equation, was particularly effective in exhibiting these qualities.
It was necessary to categorically list and separate these different audio files in order to decide on the spatial treatment appropriate for each sound. These were grouped into five distinct treatments. This table lists coherent sound shapes down to the most inherent and noisy of sound shapes respectively. For example, the original stream sample is spatialised using an incoherent noisy distribution, giving the sound an immersive and non-directional quality; however the descending water samples towards the end of the piece are treated with the first category, slow rotation of linear ramp function with a circular synchronised trajectory. This particular combination was effective in generating the slow rotational effects applied.
Multichannel spectral analysis performed by the Flux Pure Analyzer Essential application (with the multichannel option) show a graphical representation of this spiral- like formation generated after applying the timbre spatialisation.
The final stage of this work required a systematic rendering out of individual audio layers. This proved to be a time-consuming process, as each pass has to be done in real time. Each layer was rehearsed and performed allowing for a musical response to each layer. Initially after trying to save these multichannel audio files in MaxMSP, the most crash-proof method was bussing the output of MaxMSP into Logic Pro using Soundflower. For the purposes of time synchronicity I added a very short click at the start of each sample so they could be precisely time aligned within Logic Pro.
At this time I explored similar kinds of spatial transformations Mace Francis’ When Traffic Rises (2012), Stuart James’ Additive Recurrence (2012) and Erin Coates and Stuart James’ Merge (2013).
Once this corpus of sound had been derived, it was possible to view the scope of this sound world. Considering the corpus as a collective, it was necessary to decide on a structural arrangement of these different sounds. The audio files were imported into Logic Pro and arranged on the timeline allowing for immediate audition and evaluation of the structure. These audio layers were subsequently mixed with some level of automation.
In the almost 18 months between the writing of this piece and the explorations made in Particle 1, significant software development and investigation into performative approaches of timbre spatialisation was undertaken using Model A. This investigation involved the exploration of various kinds of terrain and trajectory combinations, and later explorations tended to favour the use of circular trajectories, rather than the linear trajectory used in Particle 1, allowing for a more perceptible correlation between the Wave Terrain Synthesis and timbre spatialisation processes. Nevertheless there remained some inherent flaws. One of these was that the highest frequency bins of one spectral frame naturally would spill into the lowest frequency bins of the next, creating an effect very reminiscent of the Shepard tones: an infinitely ascending or descending spectrum. Although this can be a desirable effect, it is not indicative of the features presented in the terrain or trajectory structures, and therefore does not promote the concept of a tight and intuitive semantic relationship between this control structure and the sound shape generated. The question ultimately boiled down to aesthetic interpretations, and whether the terrain could be a visual indicator for timbre spread across space and affect the spectral qualities generated across space. Early approaches to Model B take into consideration that local colours of the terrain surface affect the audible colour spectrum of their nearest speaker.
Although the earlier pieces utilise the mapping strategy in Model A, it was a visit to the ICMC in Slovenia in 2012, and preparation for a lecture I gave at the Naples Conservatory of Music that ultimately led me to pursue the histogram method. Just short of these new developments, I had the opportunity to test the software at the SpADE facility at DMARC. Here it was possible to examine the effectiveness of timbre spatialisation using the WTS control strategy over a 16- and 32-channel loudspeaker configuration. It was with this setup that the detail of spatialisation and immersiveness became apparent. Up until this time, most of the source sounds I used for testing purposes were restricted to static oscillators and noise generators. This was because I was interested in establishing various archetypes of sound shapes possible, as well as exploring the morphology of these shapes by manipulating the terrain and trajectory structures. Navigating between various immersive states could be controlled easily and effectively as I began to explore interpolating between different trajectory states as a means of morphing the sound shapes generated.
Although the sounds spatialised in the lab were invariably mono sources, the results were evolving, engaging and immersive environments. These experiments involved varying degrees of random distribution of spectra across SpADE’s 32-channel speaker configuration while experimenting with natural found sounds I had recorded. Timbre spatialisation appears to be extremely effective in a 16-channel environment for creating immersion, despite being an artificial and abstract recomposition of an auditory scene. Normandeau (2009) describes this phenomenon as “specific to the acousmatic medium is its virtuality: the sound and the projection source are not linked” (p. 278).
Normandeau (2009) also states that timbre spatialisation recombines the entire spectrum of a sound virtually in the space of the concert hall, and is therefore not a conception of space that is added at the end of the composition process, an approach frequently seen, but a truly composed spatialisation. Although the structure and basic mix for Veden Ja Tulen Elementit were in progress, I was considering the kinds of spatial treatments I would apply to specific layers of the piece. As I had already heard some of these sound shapes generated at DMARC, it was not so difficult to imagine these effects when applied to the sounds generated.
The piece explores these immersive spatial attributes—envelopment, engulfment, presence and spatial clarity using a variety of movements from slow circular correlated motion, sound slowly unfolding through space and slow fractal dispersion to fast chaotic movement and the random dispersion of sound. I found fusing standard point source approaches worked well against the more complex, evolving and immersive sound shapes generated through timbre spatialisation. This allowed the spatialisation to have momentary points of focus, influenced by the multi-levelled space forms and listener zones discussed by Smalley (2007). In this way the timbre spatialisation generates a circumspectral and immersive sound scene, from which other sound shapes and point source sounds emerge within. Complex sound shapes were also created using fractal terrains and trajectory curves, creating some very abstract and unpredictable sound shapes resulting in low- and high-frequency spatial texture. This asynchronous quality of trajectory curves produced irregular evolving and fluttering movements in the resultant sound shape. The Rössler strange attractor, a continuous differential equation, was particularly effective in exhibiting these qualities.
It was necessary to categorically list and separate these different audio files in order to decide on the spatial treatment appropriate for each sound. These were grouped into five distinct treatments. This table lists coherent sound shapes down to the most inherent and noisy of sound shapes respectively. For example, the original stream sample is spatialised using an incoherent noisy distribution, giving the sound an immersive and non-directional quality; however the descending water samples towards the end of the piece are treated with the first category, slow rotation of linear ramp function with a circular synchronised trajectory. This particular combination was effective in generating the slow rotational effects applied.
Multichannel spectral analysis performed by the Flux Pure Analyzer Essential application (with the multichannel option) show a graphical representation of this spiral- like formation generated after applying the timbre spatialisation.
The final stage of this work required a systematic rendering out of individual audio layers. This proved to be a time-consuming process, as each pass has to be done in real time. Each layer was rehearsed and performed allowing for a musical response to each layer. Initially after trying to save these multichannel audio files in MaxMSP, the most crash-proof method was bussing the output of MaxMSP into Logic Pro using Soundflower. For the purposes of time synchronicity I added a very short click at the start of each sample so they could be precisely time aligned within Logic Pro.
At this time I explored similar kinds of spatial transformations Mace Francis’ When Traffic Rises (2012), Stuart James’ Additive Recurrence (2012) and Erin Coates and Stuart James’ Merge (2013).
