International Computer Music Conference 2016 Papers Track

Full Program »

The Paradox of Random Order

Period-by-period synthesis aims at suggesting an order in randomness. Eventually, through the system, an arbitrary probability distribution is being turned into a concatenated set of waves. Statistics segregates discreteness and continuity among probability-distributions as being opposite. Here, a devised chance function entitled “noise transformer” generates continuously distributed random numbers, however, the probability-distributions are always defined by an array including a finite amount of probability-density information. Such a concept could be realized by using filter functions, however, the noise transformer will operate in the time domain. By combining the noise transformer and a statistical analysis method, an arbitrary distribution can be transformed into random numbers and vice versa. Next, generated random numbers can control ‘period durations of waves’. Shapes of the waves can be static or determined by feeding other random number streams into a prepared polynomial function, nevertheless, they are independently being made. However, respective probability-distributions can be transformed dependently and diachronically.
Based on random data generated by autonomously changing probability-distributions, period-by-period synthesis has been used for making live-performances. A fixed-media piece also has been made with the synthesis process in the context of using analog studio equipment. This experiment therefore attempts to realize a paradoxical concept “utilizing meticulous inaccuracies” in creative processes.

Author(s):

Jeyong Jung    
Institute of Sonology
Netherlands

Peter Pabon    
Institute of Sonology
Netherlands

Graham Flett    
Institute of Sonology
Netherlands

 

Powered by OpenConf®
Copyright©2002-2015 Zakon Group LLC