A step closer to particle surfaces

Hi all 🙂

Last night I finished the implementation of surface refinement for particles, this allows to increase the resolution of the mesh for smoother surfaces. I still need to redefine my current surface evaluation from particles because I’m getting the wrong isosurfaces 😛 but still I could show you my progress.
This will be useful not only for particle fluids but also for the whole particle system (Newtonian,Boids,Keyed), that’s why we plan to release first the fluid particles engine independently from the surface generator very soon 🙂

here´s some screenshots:

particle iso surface resolution of 1
particle iso surface resolution of 1
particle iso surface resol 2
particle iso surface resol 2
particle iso surface resol 3
particle iso surface resol 3
particle iso surface resol 4
particle iso surface resol 4
A step closer to particle surfaces

10 thoughts on “A step closer to particle surfaces

  1. great thing. it’s really fun to visit your blog once in a while and see all the progress.

    I’ve got one tiny question. Let’s say if I used this fluid particle system with the famous ‘fill glass with liquid’ setup, would the generated surfaces show up smooth when being rendered or show little steps like the elbeem fluids do at rather high or medium resolutions?


  2. farsthary says:


    It should be smooth , but that is heavilly dependant on the implicit function used, if the function is more metaball like , then it will be smoother but other options are aviable, this is an active field of research


  3. Thanks for your quick answer. I am looking forward to see the progress.

    i imagined it like:
    http://graphics.stanford.edu/courses/cs348b-competition/cs348b-09/finals/wine.png (where fluid and glass mesh meet) or maybe here http://area.autodesk.com/th.gen/?i/u4pbn-b1vap-d9z54-wejev:720×0.jpg (nparticles SPH)

    something i found. maybe you already know or it’s not related to your project: http://levelofdetail.wordpress.com/2007/09/28/screen-space-meshes/

    Yeah, that’s what I do too. I smooth the mesh (smooth modifer) and then (depending on the resolution of the simulation of course) use a subsurfaces Modifier. But if you use a rather high-res sim in e.g. a cylinder shaped glass you get you get visible steps which need to be smoothed a lot and if you smooth it too much it can look odd too. I have experimented with a duplicated, slightly scaled glass duplicate and a boolean modifier, to cut off those steps but this can result in nasty edges when rendering an animation.


  4. Ruddy says:

    Hi Dennis,
    Yes, if Stephen, Farsthary and Janne manage to do that, you will be able to simulate fluid in blender in a similar way than Realflow (http://www.realflow.com/), that is to say without being constrained by a domain for your simulation
    –> very convenient when you spread fluid all over your scene, as it would require a “huge” domain size with the current flow simulator (too cpu hungry)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s