K. S. Ernest (iFire) Lee
K. S. Ernest (iFire) Lee
Here is the diff https://github.com/godotengine/godot/compare/5035d24d6f24d26d5c1d936deec8e56d7caf2628..7706099a14b2b52e17f04470a320d8e9bd99d59e Note the part with the index.
@lyuma I can't see any of the changes you're referring to in the current code.
[godot-csg-mesh-repro.zip](https://github.com/godotengine/godot/files/15053328/godot-csg-mesh-repro.zip)
@Calinou can send me a test case where this errors. I am not sure if it exists. As far as I know this snapping value was because of a bug...
@Calinou I made a video of the double precision test with https://github.com/V-Sekai/godot-double-precision-test. How did you expect it to break because it doesn't seem to be breaking on a CSG box...
> [OpenSCAD](https://openscad.org/), [IFCjs](https://ifcjs.github.io/info/), [Grid.Space](https://grid.space/), and [OCADml](https://github.com/OCADml/OManifold) have all integrated our Manifold geometry kernel! Why? Because its reliability is guaranteed and it's 1,000 times faster than other libraries. See our https://github.com/elalish/manifold/discussions/340...
It might be precision. Clay mentioned that it could be related to one of the prs for "mirrors" support with the modified projection, but trying to narrow down the problem.
Do you handle basis value tracks?
I think the isfinite function is useful for checking for nan and infinities that cause corruption.
I'm curious if this happens with a gltf file too, I suspect it might be a general issue.