JonathanGranskg

I've only done some minor changes to the FLIP solver when I played around with airfields and that was mostly some changes to droplet detection system to make it work properly. I don't think many people build full-blown solvers from scratch. Like Christian and Miles said it's more about adding extra stuff.

Not sure if anyone has posted this before, but these Two Minute Paper videos by Károly are pretty interesting. https://www.youtube.com/user/keeroyz/videos

http://www.sidefx.com/docs/houdini15.0/

http://www.sidefx.com/h15 Feature list is up. Maybe download will be available soon?

You can use: if (condition) { do something } else if (condition) { do something else } It works the same way.

I love you.

Create a particle simulation with the general shape and movement you want for your simulation. Cache these out. Create a pyroFX container and source it from the particles you just simulated. Play with the parameters. Take a look at this as well: http://www.orbolt.com/asset/SideFX::flamethrower

Floraform - An Exploration of Differential Growth http://n-e-r-v-o-u-s.com/blog/?p=6721

https://mediatech.aalto.fi/publications/graphics/GPT/ "We introduce gradient-domain rendering for Monte Carlo image synthesis. While previous gradient-domain Metropolis Light Transport sought to distribute more samples in areas of high gradients, we show, in contrast, that estimating image gradients is also possible using standard (non-Metropolis) Monte Carlo algorithms, and furthermore, that even without changing the sample distribution, this often leads to significant error reduction. This broadens the applicability of gradient rendering considerably. To gain insight into the conditions under which gradient-domain sampling is beneficial, we present a frequency analysis that compares Monte Carlo sampling of gradients followed by Poisson reconstruction to traditional Monte Carlo sampling. Finally, we describe Gradient-Domain Path Tracing (G-PT), a relatively simple modification of the standard path tracing algorithm that can yield far superior results."

The story feels a bit suspicious in my opinion. I find it very immature thing to start drawing cartoons about the people involved in the company and one of the comments on the post are about him basically being an asshole at the company and everyone disliking him. I'm not sure what to believe.

That's interesting... Wish I knew it when I talked to them at FMX. I wonder how popular it is for studios to pirate software. Edit: I think there's more to this story. Read the comments.

Yep. In my experience Houdini on Windows is definitely unstable compared to Linux. I can simulate a lot more particles/voxels on Linux than I can on Windows. It's almost ridiculous. No idea how to make it more stable other than what mantragora said.

Yes, when you convert it to polygons you can plug another vdb into the third input to mask out the areas you want to keep/delete. The vdbcombine node can also be used to do boolean operators on vdbs. Although, this does not achieve the effect you are asking for. Hope it helps!

http://www.sidefx.com/docs/houdini14.0/vex/functions/pnoise https://www.sidefx.com/docs/houdini14.0/nodes/vop/periodicnoise This one should repeat over space so if you offset the input position manually over time by your periodicity it should loop. I say "should" because I haven't had a chance to try it. Edit: This is obviously a very simple noise, but I think you should be able to figure out how to add lacunarity and other features to it.