Wednesday, January 28, 2009

First post in 3 or so years

I had a talk with Simon Wittber the other day, and he convinced me to start a blog. Then I remembered I already have one, and managed to recover the account and password. Hurrah.

I'll probably try and back-blog many of the things that have happened since I left for TUM, I've been there and back, and also at UniBW and ECU, now Transmin. Worked on great projects involving autonomous vehicles in the grand challenge and robot manipulators in space and augemented environments in germany. Back home, I've been working on genetic algorithms, physics simulations, automotive simulators, fluid dynamics, CUDA and GPU's, computer vision, immersive training simulators, cryptography, and 20 meter robots. Fun! (Oh yeah, and the thesis. *sigh*)

The reason I wanted to post was to collect some thoughts on an Eulerian based fluid simulation I'm currently working on (2d) for PAL.

Found a great link that explains Euluers equations:
http://galileo.phys.virginia.edu/classes/311/notes/fluids1/fluids11/node10.html

I'm basically following an algmated approach based from buoyancy equations and shallow wave equations.
Reference material:
  • Physics Based Animation (Kenny Erleben's book) (which gave me the idea of trying simple wave equations)
  • Fast Water Animation Using the Wave Equation with Damping by Y. Nishidate and G. P. Nikishkov (which is actually just an 'academicerized' version of the standard water drop demo effect of the mid 90's, and Jeff Lander's implementation thereof (good website btw)
  • Stable Fluids by Jos Stam (just for reference/background material)
    and
  • Hardware Accelerated Real-Time Fluid Surface Simulation by Hellstrandh and Bystrom. (for how to apply the water changes coupled to rigid bodies, a nice diagram in there)

    My approach is just the good old 2d ripple effect with buoyancy coloums by raycasting the surface of rigid bodies.. anyway more on that some other time.

    I hope to blog more regularly now. :)

    Global game jam soon.
  •