Particle collision problem

[artist-fx]

when particles are coliding with ground plane the particle is going in side of the plame it means half of the particle was out side and remaining half was in side(render type "sphere") is there any way to put particles exactly on top of the sculpted plain

[thnkr]

That's where a runtime after dynamics expression comes in really handy, You can push the particles out by their radiusPP. You need to figure out their position, the clostes point on the mesh, the normal at that point, and then you displace them along that normal in the expression.

[artist-fx]

hi michiel thanks for the reply i tried runtime before i given anoffset of .5 but it was caliculating every frame and i tried the state ments also if u dont mind can u give the example scene file..

[joojaa]

Or make a surface thats offset and collide to that, mark it hidden. But yes this is exactly the entire point of particles, not to account for the shape of the object attached.

[artist-fx]

hi joojaa  offset is good for flat surface but what if my surface is un even or sclupted surface. i wnat the particle collisions same as the nparticle collisions

[joojaa]

You can offset any shape of a surface. In any vcase dont use particles where you want to use rigid bodies.

[artist-fx]

rigidbodies are  not a simple things to caliculate i have more that 1000 particles

[joojaa]

Yes but unfortunataly particle by definition means does not have any form so itr can not collide with a form.

Particle n.
  A body whose spatial extent and internal motion and structure, if any, are irrelevant in a specific problem.

So either you fake it with ofsetting the surface in question*, or you need to calculate it all, yes sorry if its too costly to not do idealisations but  youy cant have both at the same time. See you can not get A without paying for A, you have B, B dont do A. Particles dont do this and let me be clear should not need to care, if they do they arent particles no more.

*This makes the problem particelisitzable, If that doenst work then no deal. You simply can not get the higher order solution for the cost of a simplification. Sorry. Life sucks.

PS: you might use ndynamics they work with spheres, but then if your actuall shape is a duck then all bets are off.

[jiang-jieqi]

you can   instance for particles and then reverse noraml of ploygon.you can try do this