From Lawrence D?Oliveiro@3:633/10 to All on Tuesday, November 17, 2026 00:58:04
Feature that was introduced back in Blender 5.1 Geometry Nodes: Matrix
Singular Value Decomposition <https://docs.blender.org/manual/en/latest/modeling/geometry_nodes/utilities/matrix/matrix_svd.html>.
But what would you use it for? Well, CGMatter has a tute that shows
you one thing it?s good for, bless Thom?s little cotton socks <https://www.youtube.com/watch?v=aNQxQfdHcik>. Think: finding the
minimum bounding box for some arbitrary geometry that is not
necessarily axis-aligned. This does require the contruction of a
covariance matrix, but he goes through that, too.