For a while now I have been trying to develop a code to produce images of interacting binary stars. Initially I'm working on low-mass X-ray binaries and cataclysmic variables, but eventually I'd like to extend it to high-mass systems, Algols and other kinds. A number of 'artists impressions' circulate around (check out Mark Garlick's site for some beautiful examples), but until recently there has been no readily available code to produce a picture of a particular system of interest (i.e. with mass ratio, inclination, stellar temperature, etc to order). What I've been trying to do is to combine the image quality possible with current 3D graphics software libraries and hardware, with as much scientific input as possible. The eventual aim is to produce a general purpose program that can produce images and animations of any binary of interest. A similar but independant code (also based on OpenGL) has recently been made available by Andrew Beardmore here. Dan Rolfe has another code (which also does teapots!) and Jens Kube has yet another.
I have experimented with several implementations (IDL, Java-3D) but have finally settled on writing C++ code using the Mesa3D library. At the moment I'm using Redhat Linux 7.3 with Mesa 5.0. There is the facility to dump PPM and JPEG images and produce MPEG animations.
BinSim is proving to be rather portable. Although it was developed for Linux it can be readily compiled on other Unix variants (e.g. Solaris) and Unix-like operating systems (e.g. Mac OS X). The latest version also works under Windows!
From BinSim 0.7, full animation support is included. As well as orbital motion, there is Keplerian rotation in the disc, flickering at the hotspot, motion along the stream and 'bubbling' of the granulation on the companion star. Animation can be done in real time if you have a fast enough system, or to an MPEG file. I hope to add flickering to the X-ray source which lights up the disc and companion star soon. An old sample animation of a CV can be downloaded here. This has been MPEG compressed down to 1Mb, so the image quality is not fantastic. I will try to put some more and better movies on here in due course. Using the Berkeley mpeg_play program, I've found I need 'mpeg_play -dither color binarysim.mpg' to get it to display right on my 24 bit display.
I plan to expand this page soon with more images in the gallery and sample parameter files. A manual is about half written and this will describe some of the methodology as well as the use of the code. When I have a collection of samples and the manual then I might finally feel happy to release something called Version 1.0! After that I'm going to get back to doing fun new features.
rih [at] phys.lsu.edu