Inside DOP cluster grid source

[ItsMeM]

Hi Friends,  I need your help!!

I'm playing with DOP, trying to understand it and feel comfortable working with data.. I come across one issue that I have trouble with..

Is there a way to create cluster grid inside dop?

The problem is that I need to timeshift whole range to get initial data for clusters..

 

Here is the hip file, you can take a look..

I would appreciate some extra feedback about the way I approached with this setup.. I'm I doing something wrong?

Thank you!!

Regards, M

 

in_dop_clustering_help.hipnc

Edited February 18, 2018 by ItsMeM

[pedohorse]

a really quick mod to your setup ( i broke some things as well ). now emit points are actually created by impact points, and every new bounce create a new smoke object.

as i told you on vimeo - it's very easy to instantiate smoke objects from impact points, for example, through a sopnet (though it's not an ideal approach, more like less invasive)

elaborating on this idea - you can generate creation_time+ptnum hash from impact point, store it on container and particles to emit smoke just from particles from impact that generated smoke object, yet not having to spawn new particle objects (yet that would be another way, but you will still have to somehow keep track of which smoke object correspond to which particle object.)

Overall for my taste - the setup is dirty and overcomplicated, extra objects here and there, everything is a mutual affector for everything, custom resize solver that doesn't do better than standard one (btw, you did not include hda for that)

don't understand the reason for so much "custom" subnetworks - you even recreate default smoke fields and their visualizers for some reason )

 

I would suggest you keeping things really simple since your goal is not making a perfect curls on the smoke there, but exploring the possibility of solvers' interconnections and general dop workflow.

in_dop_clustering_help_mod.hipnc