Live coded AV set. Both the audio and visuals are analysed and fed into each other, in feedback loop system.
For this project we crafted an immersive performance, utilising multi-channel audio and video. Three separate computers were used, one for the visuals and one for each audio feed (one per performer). All elements of the performance were created within Max, using the application MadMapper to project to three screens. We strived for high quality projection, using three Canon WUX450ST projectors and spending time correctly aligning them. To document the performance we filmed on three separate cameras. All aspects of the performance, including the recording and documentation, were conducted by myself and my performance partner.
Various interactive visualisations and installations.
Yee-King, Grierson, Zbyszynski @ Resolution 03
Loraine James @ Resolution 04
Roly Porter @ Resolution Aures
Part of the Max/MSP patch
Through the Resolution project we have worked with many great musicians and artists, designing bespoke visualisations for each performer. We work with them to design a set of visuals that compliment and react to their musical performance. Often audio reactive, the visuals incorporate audio analysis features and combine various methods of design, including shaders, film manipulation, and custom written code.
All Resolution photographs by Rupert Earl.
Tools for splitting up and remapping visuals
We have also designed tools for manipulating visual artist's designs. Taking their work and remapping them for projection over multiple screens.
Analog and digital modular synthesiser controller, using machine learning to control synth's parameters
With this research project I aimed to design a novel and creative method of controlling parameters on a modular synthesiser. I focused on the ability of the computer to generate a vast array of textures on the synth, then used machine learning to compose according to live user input. I chose to investigate connectivity with modular synths due to a desire to control these machines remotely. Whilst there are methods of control already available, I found them to lack the dynamic element and responsiveness that I desired. These methods don’t allow for dynamic transitions, as I was striving for an intelligent system that can learn and explore.
More information can be found on the Dissertation page.
Max/MSP
Wekinator
Processing
C++
