Sunday, April 8, 2012

First Steps

Sooo we decided we needed to change how we were doing things a little bit! After more reading (we read through Dr. Christopher Batty's notes found at http://www.cs.columbia.edu/%7Ebatty/teaching/general/Fluids2.pdf and looked over his base code for a liquid sim ) we realized we needed to implement marker particles before we did anything else (like level sets...). So this weekend I added a vector<vec3> that holds the locations of the particles that will mark the location of the liquid and functionality that advects their positions according to the velocity field of the grid.   I also added two vectors of booleans for tagging that hold whether a given cell is empty/not empty and solid/not solid. When the positions of the particles are advected, the cell that the given particle is in is tagged as "not empty". I've also added a function for initializing the particles' positions and for right now we are just manually placing them. For the future I'm concerned that the bounds checking that I'm doing when we advect the particles is too simple (it's just check their locations against the dimensions of the grid*cell size) because it's not generalizable to other shaped containers. I hit some strange value initialization issues last night that I think might be a scoping mistake, but I handed it off to Jason (who is traveling back home as we speak)!

Happy Holidays! Animation played a major role in my religious holiday education so here's some tidbits for you:






No comments:

Post a Comment