As pointed out by Matias Goldberg [1], neither model is particularly optimized, however. They produce too many sliver triangles. The max-area approach [2] was b

As pointed out by Matias Goldberg [1], neither model is particularly optimized, however. They produce too many sliver triangles. The max-area approach [2] was brought up as an alternative that can significantly improve shading performance. I'd like to see benchmarks against constrained Delaunay triangulations too, which also tend to reduce sliver triangles and have some potential advantages over the greedy max-area approach because they distribute the large areas over many big triangles instead of leaving a few slivers around the edges. (Of course, none of this would be likely to improve FPS in the game, but it's an interesting computational geometry problem nonetheless.) [1]: https://twitter.com/matiasgoldberg/status/1636504187079995392 https://twitter.com/matiasgoldberg/status/163650418707999539... [2]: https://www.humus.name/index.php?page=Comments&ID=228 https://www.humus.name/index.php?page=Comments&ID=228