Hi, dear All:
Has anyone implemented fluid dynamics introduced by Jos Stam in his paper named Real-Time Fluid Dynamics for Games on AMD's GPU?
If someone did the work, i wonder if I could get some help. Thanks in advance.
If I can get around the problems and further into this project, I'd be very interested in implementing such a thing.
I made a comment on the Infinity-Universe forums about simple fluid dynamics shaders.... I was told there's not enough time in a frame to fit that stuff etc. etc.
I believe CAL is really unlocking doors. Obviously a lot of the procedural geometry of the I-Novae engine is based on the cpu, because I can get fBM going at 200 octaves and the shader still completes in a very fractional portion of a second on maps of up to 2048x2048 (as it seems to be close to the limit for me to calculate in one pass). This alone is evidence that the technology has arrived to do a lot more... technical things in games... Physics are no problem on a massive scale, mapped to SIMD architecture. Fluid Dynamics, Cloud Volumes, you name it, it can be prepared before rendering even occurs. Who needs a programmable graphics pipeline... that's just for the fireworks and stuff....
Retrieving data ...