id like to use guide geometry to control the viscosity of a flip sim (i.e. all particles inside an object at the start of the sim (geometry not included in sim) are given a specific viscosity etc.. )

any tips greatly appreciated.

second week on houdini.. my head hurts but im enjoying it so far.

Something like this?

exactly like that. loving this forum.. many thanks.

robinlawrie
second week on houdini.. my head hurts but im enjoying it so far.

When your head is hurting, you're doing it properly

hi, resurrecting this thread as ive noticed that when i crank up the detail in my sim, the geometry intersection test is taking about 98% of the time on the first frame.. with a reasonably detailed sim (but still way, way below my final scene setup) this is about 5 hrs.

is there any way to speed this up? makes testing a bit of a pain in the ass.

thanks, robin.

Try it with sampling an sdf of the object whose bounds you need. Should be ridiculously faster

thanks for that, it looks promising. however my sim is set up generating multiple groups that each then get a viscosity assigned.

could you advise how id use the vopsop to create groups rather than set the viscosity directly?

sorry its like getting someone to do your homework for you, but there is so little training material around for these things.


thanks again for your help, Robin.

just to show im trying.. ive edited the vopsop so the bool output from the compare node goes into a pair of if_then nodes, one set to true, one set to false. these are wired to a pair of new group nodes, each with a different group name.

i dont get any errors, and if i pull up the stats for the vopsop i get confirmation 2 groups have been created (both unfortunately with 0 points in)

am i on the right track? ill have to transfer this setup into my scene as i use a fliptank, so maybe the setup will differ slightly.

I'm sorry to hijack this thread with a stupid - nevertheless unanswered question - …

Looking at the hip provided (Thanks a lot!) there doesn't seem to be a way to keyframe viscosity or influence it after the simulation started.(?)

I would like to “melt” something after it became “frozen”. I guess you could do it with a time based viscosity attribute. But in my case I would like to have it change on impact (from a different volume).

Any chance the particle viscosity can be changed by keyframe or a intruding volume during a running simulation?

Thanks for any help and sorry for interrupting this thread - but it seemed the only one close to covering the issue.