[Bf-committers] Newton fluid control in smoke module

Patrick Fuerst pat.fuerst at gmail.com
Thu Oct 3 15:27:08 CEST 2013




Hi there, 

I've looked into how meshes as obstacles are added in the smoke module. 

To explain my plan on how to add objects for the target density I will explain how I did the implementation of the algorithm first. 

	+ First I added an array  for the target density, which is the same size as all others (density, velocity ect. ) 
	+ Then I implemented the algorithm which calculates a force in each grid cell to reach the target density, this is calculated right before the forces are added to the velocity. 
	+ For the boundaries I simply edited the obstacle array and the existing code handles the rest. 

So far this works all good but now I want to add more complex objects as target density and don't want to init the target density by code. 

My plan would be to do it the same way as adding a collision obstacle and the inner volume would define my target density. 

Patrick






On Oct 2, 2013, at 2:19 PM, Brecht Van Lommel <brechtvanlommel at pandora.be> wrote:

> Hi Patrick,
> 
> The fluid simulator supports control objects, so I imagine the smoke
> simulation system could get something similar, where you have a
> Control smoke type next to Domain, Flow and Collision.
> 
> Miika Hämäläinen and Daniel Genrich are the smoke experts, they might
> have more specific ideas about how this should be implemented.
> 
> Brecht.
> 
> 
> On Wed, Oct 2, 2013 at 1:34 PM, Patrick Fuerst <pat.fuerst at gmail.com> wrote:
>> 
>> 
>> Hey all.
>> 
>> I'm new here and want to introduce myself first.  I'm a student at the Technical University of Vienna studying Computer Graphics. Currently I'm working on my Bachelor thesis implementing an algorithm to control newton fluids.
>> http://cg.iit.bme.hu/~zsolnai/gfx/fluidcontrol/real_time_fluid_control.pdf
>> Already implemented it in Blender.
>> 
>> Currently I'm testing it, by adding obstacles (serving as my boundaries for the smoke distribution ) like a sphere with the old code of the wave turbulence.
>> The next step would be to add some code, so it would be  possible to add any kind of geometry serving as my target distribution.  Like adding collision objects.
>> 
>> Would appreciate some thoughts about this.
>> 
>> 
>> cheers,
>> Patrick
>> 
>> 
>> _______________________________________________
>> Bf-committers mailing list
>> Bf-committers at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-committers
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers



More information about the Bf-committers mailing list