Stone Devlog: November (Lack of) Progress

Last time we spoke, I mentioned that I hadn't made much progress in getting a surface extraction algorithm up and running. This post is going to be more of the same. I can't even say it's for lack of trying, because it pretty much is. Life has been busy and I haven't had time to devote to the project.

That said, I did come across a 97 page master's thesis on Cubical Marching Squares that covers an implementation with everything except the part I actually need the algorithm for. It should get me started but once again, we're back to that pesky sharp feature preservation that no one seems to know how to implement. At this point it seems I will have to dust off the my old math textbooks, learn the underlying math from scratch, and then figure out how to implement this functionality based off the research papers. It isn't something I can chip away at, I will just have to dig in and dedicate the time and effort into figuring this out before I can continue on.

I will admit this isn't exactly a blocker, as I could make progress on other systems in the meantime. It was only a couple posts ago that I showed functional incremental subtraction. However, being able to properly extract these surfaces from volume data is the make-or-break for this approach. The advantages it presents over my previous attempt are tremendous, but only if I can consistently and accurately generate meshes for my objects. I also have some smaller projects that I would like to pursue at some point using this technology. On the flip side, I do acknowledge that there are still a lot of unknowns, and sinking this much time into something that may not work out in the end is not an ideal situation. That's the main reason I wanted a working implementation that I could try out and dissect. Unfortunately, it does not appear that I will get that luxury.

No sense bemoaning my misfortune. There's still a massive amount of work to be done and most of it hinges on me figuring something out here. We'll chat again soon.