Fogging in progress

So I'm not sure this is going to look good ever.  But, still, I'm creating soft, 3D shapes procedurally, which is kind of cool.

Right now there are two main problems.  The first is that I'm not doing a good job of combining the depth-based fogging with the interference-based fogging.  The second is that I'm sampling discrete parts of the world to get fog depth, which creates those circles on the wall-- those are where the sampling function jumps to a new number of samples, and so it creates discontinuous values.  Sampling also means that it's slow.  It takes a lot of calculations.

I think that 3D sine waves like these could be integrated, by somebody smart with calculus, rather than sampling.  I'm not smart with calculus.  But!  I was thinking that triangle waves might look better-- and then I realized that triangle waves would be probably be a hell of a lot easier to integrate as well.  So maybe it's not hopeless.

Beyond that, there are still problems.  The wisps move through the model, rather than around her, and I don't think there's anything I can do about that.  Second, there are insufficient depth cues.  I might be able to provide those with some grain from a texture.  Not sure.  Finally, the density just doesn't seem right, which is why I started thinking about triangle waves to begin with-- but even then, I might have to do something to exaggerate the wisps in the foreground, and minimize them in the background.

So, interesting experiment so far, more to investigate, but nothing worth using, at least as of yet.  The bugs are always fun though.  Of course, not bothering with turbulence or physics until I get something worth it.

If anybody knows of a great MMD fog effect, let me know.  I've seen ones like DepthOut and NightFog, which are okay but have no complexity to the fog.