So I haven't been doing too much on the game front in the past few months - at least, nothing that is really 'visible'. I've mostly done back-end stuff, like code cleanup, refactoring, etc. I am currently playing around with another algorithm for smoothing over voxels called Marching Cubes. I think it's probably the more common one used in games. My understanding is that it isn't as 'accurate' as Dual Contouring, but it is much faster (as it's mostly just a giant lookup table).
My current problem is that my terrain isn't smooth. There are jagged edges scattered around the terrain, no matter how much I 'smooth' the generated noise. I think the problem isn't that the Dual Contouring algorithm isn't working properly. I think the problem is that the noise I'm generating isn't smooth enough - that is, the jagged edges are actually part of the isosurface that my noise algorithm is describing. Nuts! So my thought is I can try Marching Cubes, which doesn't preserve sharp edges, to smooth these rough spots. So in effect, the loss of accuracy in Marching Cubes vs Dual Contouring is an advantage! We will see how this experiment comes out though. I still really like the idea of having a 'high precision' algorithm - I don't really know why, but it seems like to use a less accurate algorithm is the wrong solution. I think it ultimately feels like I failed to get it to work right, and that bugs me - a lot :) Maybe if I worked on the noise generation to be smoother, but I really am not sure how I would do that. And I'm not sure I really want to invest the time to figure that out. Oh well, we shall see what I ultimately decide.
In other news, the job hunt is on! I've made the decision to leave my current company (after 3 years!). So if anyone knows of any good opennings in the KW area, let me know! :)