jonp Posted December 2, 2014 Share Posted December 2, 2014 Just curious how the smoke / pyro solver works... does it try to solve the equation pressure = density * temperature? In other words, if I create a sim with a constant initial density, and varying regions of high and low temperature, will the density advect to the lower temperature regions, or will the high temperature dissolve into the low temperature region? Which drives which? Would be cool to create "wind" in a realistic way (by changing the pressure). Cheers, Jon Quote Link to comment Share on other sites More sharing options...
loopyllama Posted December 2, 2014 Share Posted December 2, 2014 someone should chime in if I am wrong... I am pretty sure the only pressure in the solver is the pressure field used to make the new velocity field divergence free. The density is only being pushed around by the velocity field and also diffused into the container. In a smoke solver, temperature will affect the velocity through the buoyancy. There are buoyancy controls for scale and direction. If you have a smoke solver with varying temperature in the container, and some amount of buoyancy, the solver will advect the velocity field and any present density. Quote Link to comment Share on other sites More sharing options...
Annon Posted December 2, 2014 Share Posted December 2, 2014 I believe you are correct, the pressure field is to make the sim non divergent. Also used to calculate the collisions as well as gas expansion (divergence) in a pyro based sim. You should be able to inject pressure however (I believe) to force it, but I've only ever used pressure to affect other things (like turbulence or masking sources by it), not injecting it straight in. I'll have a go tomorrow. Quote Link to comment Share on other sites More sharing options...
freaq Posted December 2, 2014 Share Posted December 2, 2014 one could test it easily,have a volume with old gas,add one heat emitter (say a bar on one side.)if the gas starts swirling, you could do wind that way, if not you cannot.that said you could create a vector field (velocity) yourself from this as well offcourse. Quote Link to comment Share on other sites More sharing options...
eetu Posted December 2, 2014 Share Posted December 2, 2014 Ohh yeah. Trying to get some global weather patterns going would be awesome fun! Here's quick radial gravitation and buoyancy vop forces for starters. It would probably be quite a bit faster with a velocity-modifying volume vop, but I always feel a bit of a brute when modifying velocity directly. Then this needs a rotating frame of reference to get some coriolis going, and then density and temperature profiles for the air column,..too bad that Houdini density isn't really density in that sense.. ee_meteo_v003.hip ee_meteo03.mov 2 Quote Link to comment Share on other sites More sharing options...
jonp Posted December 4, 2014 Author Share Posted December 4, 2014 Thanks, seems as though it's not fully straightforward to translate physical values. I guess it shouldn't surprise me too much as fully efficient fluid simulation is still not a solved problem! Nice, Eetu. A while back I actually put in a RFE for spherical voxel coordinates (for me it was to render some scientific sims), which would help out your demo quite a lot. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.