using sdf data in particle network for collision

[junaid]

Hi Champs,

I am simulating some good amount of particles and have to collide those particles with some rigid bodies and was wondering, how do I save my rigid bodies into sdfs format(iso offset or something else) and use them in particle simulation for collision?

Cheers,

Junaid

[pclaes]

I'm not sure if you can use them inside a pop network.

But you can use a pop network inside of a dop network and use the pop solver to make them collide with your rbd's.

I've recently posted an example in this thread that also uses the sop solver (you might not need the sop solver part). The file is called "particle_erosion_12"

http://forums.odforce.net/index.php?showtopic=8990&hl=

Good luck

[junaid]

hey peter,

I am doing collision with geometry and its working fine at the momment, but I happen to attend jeff houdini 10 demo and he was suggesting me to use sdf's for much robust/accurate collision and I was wondering is anyone have any experince with it.. bcoz there shitt lotts of particles and dops will die if I use pop solver in dops I guess, but I see what do you mean.

PS: your raven project is looking quite convincing, keep posting good stuff

Cheers

Junaid

I'm not sure if you can use them inside a pop network.

But you can use a pop network inside of a dop network and use the pop solver to make them collide with your rbd's.

I've recently posted an example in this thread that also uses the sop solver (you might not need the sop solver part). The file is called "particle_erosion_12"

http://forums.odforce.net/index.php?showtopic=8990&hl=

Good luck

[pclaes]

I am doing collision with geometry and its working fine at the momment, but I happen to attend jeff houdini 10 demo and he was suggesting me to use sdf's for much robust/accurate collision and I was wondering is anyone have any experince with it.. bcoz there shitt lotts of particles and dops will die if I use pop solver in dops I guess, but I see what do you mean.

PS: your raven project is looking quite convincing, keep posting good stuff

Perhaps in H10 you can do it sdf collisions inside pops - or perhaps with a vop pop, sort of more as a repulsion field instead of hard collision. I can see it'll get slow with the pop solver. I'd love to spend a bit more time on it and look into it, but I'm currently learning Massive and Renderman shader writing - deadline for those projects is in 10 days, so not much time for anything else.

The Raven project was fun to work on, now it's over again. They jam-packed the MSc full of good stuff this year, so it's just moving from one project to the next. I'll gather all my stuff at some point this summer, make a new reel and put it online. I might see you in September in London at some point. Not sure yet where I'm gonna go, but I would like to work on the vfx for the next Harry Potter.

Good luck with all the sand, errr particles .

[junaid]

its to do with advect volume I guess, anyway I am trying something which seems decent but yeah welcome for any suggestions/ideas.

cool, good luck with your assignments, I exactly know your situation with assignment and yeah give us a shout when you happen to come to london, we generelly meet up often on friday evening for drinks. see you soon.

Cheers,

Junaid

Perhaps in H10 you can do it sdf collisions inside pops - or perhaps with a vop pop, sort of more as a repulsion field instead of hard collision. I can see it'll get slow with the pop solver. I'd love to spend a bit more time on it and look into it, but I'm currently learning Massive and Renderman shader writing - deadline for those projects is in 10 days, so not much time for anything else.

The Raven project was fun to work on, now it's over again. They jam-packed the MSc full of good stuff this year, so it's just moving from one project to the next. I'll gather all my stuff at some point this summer, make a new reel and put it online. I might see you in September in London at some point. Not sure yet where I'm gonna go, but I would like to work on the vfx for the next Harry Potter.

Good luck with all the sand, errr particles .

[pushpathadam]

Hi,

Has anyone got the collisions in the curlnoise pop to work? It has a hook for an sdf collision but the file browser doesn't recognize the .simdata suffix. I wrote in the full path but that seems to break something..the advection fails..no warning, nor errors..just no advection.

-Tom

[junaid]

its better to save it in bgeo, its just gonna contain little more info then simdata, haven't tried it though. let me know where you get around just for knowledge.

Cheers,

Junaid

Hi,

Has anyone got the collisions in the curlnoise pop to work? It has a hook for an sdf collision but the file browser doesn't recognize the .simdata suffix. I wrote in the full path but that seems to break something..the advection fails..no warning, nor errors..just no advection.

-Tom

[pushpathadam]

I tried saving the sdf as a bgeo...and reading it into the curlnoise pop. It cancels out the advection just like using the sdf does. Also I tried reading the bgeo back in with file sop as a sanity check. That errors out.

its better to save it in bgeo, its just gonna contain little more info then simdata, haven't tried it though. let me know where you get around just for knowledge.

Cheers,

Junaid

[junaid]

check this out if you just after collision through sdf geo

http://www.sidefx.com/index.php?option=com...p;highlight=sdf

cheers,

Junaid

I tried saving the sdf as a bgeo...and reading it into the curlnoise pop. It cancels out the advection just like using the sdf does. Also I tried reading the bgeo back in with file sop as a sanity check. That errors out.

Edited May 26, 2009 by junaid

[junaid]

check this out if you just after collision through sdf volume

http://www.sidefx.com/index.php?option=com...p;highlight=sdf

cheers,

Junaid

I tried saving the sdf as a bgeo...and reading it into the curlnoise pop. It cancels out the advection just like using the sdf does. Also I tried reading the bgeo back in with file sop as a sanity check. That errors out.