Hey all 
I guess you all saw the frost glass spill video I put online. I’ve just created a new video that shows how frost solve a big issue that’s always a pain in realflow - the mesh flickers. Most common in high viscosity simulations. I didn’t see this advantage noted anywhere in the FAQ or the docs but it is quite a huge advange which makes working with realflow much much easier and fun.
youtube.com/watch?v=tdUTPdEZAZo
Cheers,
Isaac.
Very cool 
I was really surprised how nice a look you could achieve with low particle counts, that is huge.
You made the Thinkbox News again (and made my day, too 
)
thinkboxsoftware.com/news/20 … -test.html
This was not mentioned in the FAQ because frankly we did not know about it. In fact, we consider improving the Frost meshing in the future because some of us don’t think it is good enough…
Great to hear it performs well in your tests!
Thank you both!
Bobo, I was really excited to test Frost just to see if such meshing issues can be finally resolved for realflow stuff. I did spend countless days in the past trying to get clean flicker free meshes and you can just imagine the frustration of having over 100gb of sim data that still flicker in HD. Even when I did kinda managed to get it cleaner the long process it took just didn’t worth it in production value. I know some major companies use Houdini just for the meshing process since it also have better a way to read particle neighbors data but that’s not an approach every company can do.
Yes while Frost is great it’s not good enough like you say. As JohnnyRandom already reported in the beta forum, one of the major things is the phantom particles created between real particles with the Zhu/b mode. It can make funny looking bulges if you don’t have enough particles in certain areas. (attached image).
Anisotropic is great too but you need to be careful with it’s blend value otherwise the mesh turns too flat and lose it’s volume.
Being able to read .pd realflow files would be great too. It gives all the separate channels you just need to make amazing stuff (neighbors, viscosity, pressure, density, age, etc…).
Not that I’m complaining, still an amazing tool. But there’s always room for improvement. 
thats a great test!
Thank you for your post!
Would you happen to know more about this, or does anyone have a related link?
Yes, I would like to improve all of this.
Increasing the Low Density Trimming Threshold or Strength may help trim away the phantom particles created in Zhu/Bridson mode.
Sent you a PM re the houdini thing. I’ll try the trimming density you suggested, thanks!
