Very specific (i think) fluid sim issue

[florian_DKS]

Hey everyone!

I've been creeping on this forum for a year now, and have had so many questions answered, but this I can't really google, so I'm turning to you!
So here's the problem: I'm working on a ink mixing setup, based on the Taylor Raleigh instability thing. My issue is this: I am using a Sop solver inside my dop net to create "divergent bubbles" that repulse the ink to the sides at specific times, but when I mesh the simulation, those very loosely scattered particles look awful! So I had two ideas in mind:
- create them in a specific color that's not present elsewhere in the sim and use that info to create a group and delete it after the sim (I am very bad at vex and have run into multiple issues trying this).
-create a group inside the Sop that's inside my Dop, and use that group info to delete. But the group info doesn't transfer outside my Dop (at the first sop level). So I tried making a "delete" attribute, same thing, I can't get the info out to the first sop level!

- any ideas you might have.. (I'm pretty bad at houdini and have been fiddling around, so I'm sure there is a better solution than the ones i'm thinking of). Also, I did want to go unmeshed and use pure particles but I haven't yet found a good way to render them, but that'll be for a later time!

Attached are screenshots of my node tree and the result, sorry if there are useless nodes here and there, i've been trying many many things now.

Any ideas ?

 

[ThomasPara]

Cant see where your input geo are merged into the sim and witch attributes you transfer, easier if you support us a file.

[Librarian]

Quote

(I'm pretty bad at houdini and have been fiddling around, so I'm sure there is a better solution than the ones i'm thinking of)

nice render you have on instagram ____All done by comb av cinema 4d and Houdini or ?

 

[florian_DKS]

8 minutes ago, ThomasPara said:

Cant see where your input geo are merged into the sim and witch attributes you transfer, easier if you support us a file.

Hey, unfortunately I can't really share the files as i'm in the RnD process, and i intend to pitch this to a client, but actually someone helped me correct my VEX syntax and so on, so I was able to get rid of those hideous blotches! To answer your question, on the first image there is a null called "OUT_BUBBLES", and in the third the object merge is calling to that specific null, in order to actualize the values on each frame and be able to animate the "bubbles"
Just in case any other noobs here (like me) ever need to make a group according to customizable Cd values, here's the VEX code:
in an attribute wrangle:
"

int group_"yourgroupname" = 0;

if ((@•Cd.r > chf("tolerance_r")) && (@•Cd.g > chf("tolerance_g")) && (@•Cd.b > chf("tolerance_b"))) {

       i@group_"yourgroupname" = 1;

}

"

(just press the little button and the channels appear) (again, for the noobs, I know this is basic stuff)

Very very basic but i couldn't get the syntax right here, and i needed specific values for each Cd channel, easily customizable to tweak and find which points to remove! 

Here's the result without the blotches!

(I was also able to get more definite displacement values, which was awesome)
 

 

[florian_DKS]

15 minutes ago, srletak said:

nice render you have on instagram ____All done by comb av cinema 4d and Houdini or ?

 

Thanks man! Yes most of last year is houdini, only rendered in C4D because thats where i started and feel most comfortable with cameras, lights, etc,.. But I'm actually teaching myself houdini rendering, and cameras, etc.. today, because i'm sick of exporting alembics for hours..

[florian_DKS]

Oh, but if someone has a workaround to have detailed fluid meshes, however low the particle density is, I'd be glad to hear it, I did lose nice details on the divergent bubbles with mine..!

Edited June 11, 2019 by florian_DKS

[Librarian]

florian please..

.Am ( Song-----------Show must go on-------) ..

Quote

intend to pitch this to a client

“You’re so talented!” 

please share your file .