Open
Description
In #9725, I started adding a few different variables for position in different coordinate systems, positionMC
for model space, positionEC
for eye space.
But the tricky bit is handling global coordinates. there's a couple things to determine:
- Precision. The easier thing to implement is to only use 32-bit precision positions. Though this will cause precision issues when working on small scales. The more robust thing would be to split positions into high and low parts (see the article Precisions, Precisions), though this requires more memory and calculations in the shader.
- Cartographic coordinates will require a bit of shader code for the ECEF -> cartographic coordinate converstion. Though the implementation depends on the answer to the first question
This goes beyond the scope of #9725 so I'm opening this issue for later.