Open w23 opened 7 years ago
Reversed-z would be ideal. However, OpenGL is crap (for its -1 to 1 mapping which destroys precision for no good reason), and OpenGL < 4.5 (which also implies ES) is double crap, because it doesn't have a way around that (glClipControl is either ext or gl 4.5).
So, according to this: https://developer.nvidia.com/content/depth-precision-visualized my best option seems to be to try separating projection matrix from model-view.
Separating projection matrix from model-view matrix doesn't lead to image improvement.
I don't mean coplanar surfaces from different maps. Should either tweak z-buffer params, or completely change z buffer approach to something less naive and more contemporary.