theor's

I recently implemented the famous Marching Cubes algorithm (Wikipedia) using density function to describe the terrain (GPU Gems chapter about that). Having to recompile the code to tweak the density function was too slow, so I wrote a fast expression evaluator for Unity, available on GitHub.

I was curious to see if I could implement belts like in Satisfactory or Factorio with DOTS in a performant manner. In the end, my implementation is highly parallel and works at 60fps for a million items, but let’s start with a deep dive into how both games do it.

In the last part, we ended up with a collection of random 2D points. Now it’s time to triangulate them. After an overview of what a Delaunay triangulation is, we’ll describe the Bowyer-Watson algorithm and write a custom data structure to store the triangles, then we’ll generate a mesh from it.

We’ll start by generating the random points we’ll triangulate later and benchmark that, but first, we need a bit of boilerplate :

• a Monobehaviour holding the generation parameters
• a matching inspector to add a few debug facilities
• a DrawGizmos function to visualize the generated points

I always loved procedural generation, and was really impressed by mewo2’s fantasy maps. I decided to give it a try using Unity and the Burst Compiler, an amazing piece of tech developed at Unity. It was also the opportunity to dig into SVG and implement a few algorithms and white papers: Delaunay’s triangulation, a depression filling algorithm, a cartographic labeling algorithm based on simulated annealing, water network computation and a few others. I’ll write a (probably long) series of articles detailing the interesting parts and algorithms of the project and my general process to approach this kind of topic.

This is an opinion piece. YMMV

Once in a while, I start a side project whose ideal format is a website. This article is about how I eventually settled on F# and Fable.