Inspired by the lovely work done by Tim Hutton and his colleagues on Ready, I thought I'd try to couple two separate Gray Scott systems together inside my own reaction diffusion application, ReDiLab.
While I haven't been able to replicate Ready's stunning self replicating spots demo, I have created a rather good impression of ink drops diffusing in water which I've added as a preset to my application.
Although Stage3D and dear old Pixel Bender, being GPU based, are the perfect tools for this job, I'm about to hit a hurdle. The most number of components that a GPU based reaction diffusion system would be able to handle is, obviously, 4 (using the red, green, blue and alpha channels). If, for example, I wanted to couple a pair of three component systems together (e.g. Belousov Zhabotinsky), I'd be stuck.
The application, now at a heady version 0.3, lives here and the source code lives here.
Add a comment