written by Jason Sadural (email@example.com) comments and suggestions always welcomed
Our purpose is to explore the realm of computational physics and physical tendancies through acoustical perception. In our inate desire to explore who we are and where we come from, physicists have explored the cosmos as far as light and the age of the universe will allow for answers. From data gathered at the furthest spectrum of the universe such as the Cosmic Microwave Background to relatively closer phenomena, a standard model is devised in order for us to understand what we observe and why. Through observational Astro-physics, data from telescopes and satalites are systematically collected to confirm and converge to coefficients in our physical models as well as tendancies certian systems. Computational Astrophysicists then recreate and simulate these systems in order to test the stability and consistency of these models. Within this process, we attempt to create meaningfull sonification techniques of these simulations in 3-dimensional spatialized sound in order to better understand the physical tendancies not easily seen with current Visualization techniques.
What is Computational Astrophysics?
Computational astrophysics is the simulation of astrophysical phenomena on a computer by numerical integration of the relevant governing equations. Such simulations produce detailed solutions to highly complex problems in stellar evolution, galactic dynamics, numerical cosmology, and many other fields.
The simulation data we are going to use is from Laboratory for Laboratory for Computational Astrophysics at UCSD directed by Michael Norman. Professor Norman was one of the first innovators to apply computational physics to simulate star formation from evolving over time the initail conditions of our universe. Our goal is to develop meaningful sonification techniques and sonify phenomena through data mapping of simulation experiments done by his graduate student Dave Collins. Later we hope to be able to adapt this technique to other simulations fundamental to our understanding of physics.
Links to simulations we hope to Sonify:
- http://lca2.ucsd.edu/~dcollins/MagDensityJPGs.tar (Dave Collins current data)
- http://cosmos.ucsd.edu/~mnorman/movies/siggraph01-amr.mpeg (Mike Norman's Simulation)
- http://cosmos.ucsd.edu/~mnorman/movies/normanDFest864.mpeg (Mike Norman's Simulation)
Software currently used to create simulation data:
- ZEUS-2D: A one- or two-dimensional explicit Eulerian grid code for astrophysical radiation magnetohydrodynamics.
- ZEUS-3D: A one-, two- or three-dimensional explicit Eulerian grid code for astrophysical magnetohydrodynamics.
- ZEUS-MP: A parallel, three-dimensional explicit Eulerian grid code for astrophysical magnetohydrodynamics.
- TITAN: A one-dimensional implicit adaptive mesh code for radiation hydrodynamics.
- [KRONOS]: A three-dimensional grid-based hydrodynamics cosmology code combining the piecewise parabolic method (PPM) with the particle-mesh (PM) algorithm for collisionless particles.
- Enzo: an adaptive mesh refinement (AMR), grid-based hybrid code (hydro + N-Body) which is designed to do simulations of cosmological structure formation.
- [MGMPI]: A parallel, multigrid linear system solver for second order PDEs.
- 4D2: An interactive tool for visualizing and animating 3D data (array, particle) on Silicon Graphics workstations.
- [LCA Vision]: The portable rewrite of 4D2 including support for adaptive mesh refinement data.
The software we will be using for sonification playback is Pure Data (PD) created by Miller Puckette, currently Associate Director of the Center for Research in Computing and the Arts. The core of the research will be conducted and developed in the listening room at CCRMA. The current approach is to create a virtual environment where the listener is in a position or along a path in the simulation and in real-time be able to explore the space sonically. Many projects in parallel will be developed with the intention of sonifying astro-physical data.
Links to projects:
In the first approach in linking simulation data with usable data to current computer music tools, we will be using a 2D slice of the 3D simulation. The 2d slice in jpg format will be analyzed by Raster Scanning developed by Woon Seung Yeo. Once implemented this process though helpfull will prove time consuming and cpu intensive. Script is currently being developed in conjunction with Dave Collins and myself in order to extract 3D data directly and efficiently from the current simulation data interpreter. This data will be used to simulate a constantly changing inhomogeneous ether in virtual space representative of various types of physical densities one would encounter in space. The user can then purturb this virtual space using tools developed in parallel.