Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://star.arm.ac.uk/~spm/animation.html
Дата изменения: Wed Mar 8 15:35:54 2000 Дата индексирования: Mon Oct 1 22:37:54 2012 Кодировка: Поисковые слова: ngc 6520 |
Watch as I try to create things that look like solar systems using the laws of physics and a bit of number crunching (or listen to me rae on about it here....
This is an image from a sequence I'm rendering at the moment, most of the movies are by products of my work, movies of planets forming, or being destroyed or whatever I'm currently working on. Unfortunately I can't figure out an easy way of forcing an automatic refresh with each frame, so a manual reload is the order of the day :-(
Of course there's no real way to tell if the frame is just taking a long time to render or whether I just don't have anything running so this is hardly the most interesting 'Live' sites in the world, if you wan't real life then you can get your kicks watching Jennicam.....
All of the animations were rendered on my desktop machine, by a quick and dirty Bulirsch-Stoer integrator which I hope to replace with something a little more specialised (but I'm not revealing that just yet, I hope to get a paper out of my 'neat idea'). The data files are passed through another purpose written piece of code which generates a scene description file for Povray. All the frames are converted into JPEG format for storage and when the time comes they are sequenced together using the Berkely MPEG encoder software.
Actually I'm not entirely proud of the quality of the encoding, mainly because I can't get the system to produce decent sequences without encoding them entirely in I-frames, so the streams are larger than they could be. Also most of the acretion disc work is not well suited to being rendered on Povray since the objects are so far away and small. of course it was much more suited to my series on planets colliding, just a shame that these are hardly realistic :-|
Also because I'm only working on my desktop machine the amount of CPU time is not sufficient to do the number crunching with anything approaching realistic conditions, in this case I make the particle cross sections roughly 100 times what they should be, so the amount of scientific value in these animations is questionable. Of course part of ,my work involves working out haw to do this realistically , withing the lifetime of the universe, and preferably, within the duration of my PhD :-)
Of course those sequences which are more predictable can have the load spread across multiple computers using my Distributed Animation Rendering Scripts for POVRay. These scripts simply let you use many computers to render a sequence of frames, with a single computer coordinating and collecting the images.
So here are some of the results, I hope to get them all linked in with time, and perhpas censored to remove the rubbish ones.
You might also be interested in some older 'experiments' I did with computer graphics. I needed a programme to generate asteroids for movies, and out of some strange (read sad) scientific line of thought I generated some hypothetical light curves for some of these asteroids. If I get time I might do some real research based on this.....
And finally there's a directory full of unsorted stuff here.
That's all for the moment, I should have some more linked in when I find them all, feel free to e-mail me if you have any questions about these, or how to turn your boring scientific data into funky animations.