Orthographic cameras are specified by setting the `orthographic` attribute of the `<camera>` element. The `fovy` attribute is still used to specify the field-of-view, but its semantic is different for orthographic cameras. For orthographic cameras, the field-of-view is expressed in units of length, rather than degrees.
Other related changes:
* Fix bug in the ordering of `cam_xxx` elements in `mjModel`.
* Make camera visualization translucent only when the frustum is visualized.
* Added a button to `simulate` to toggle between perspective and orthographic free cameras.
https://youtu.be/ZXBTEIDWHhs
PiperOrigin-RevId: 642293435
Change-Id: Id090a421ad88ab404b5b27ddbbd6bfc81ad49bc5
If GL_ARB_clip_control not available then [znear, zfar] -> [1, -1] instead of [1, 0] in normalized device coordinates
Thus, Z is still reversed, but not shifted, clipping should still work properly, and conversion to window coordinates does not cause accuracy regression compared to non-reversed Z rendering.
Combined, both options greatly improve camera depth accuracy compared to
OpenGL default Z mapping and int24 depth buffer.
Requires GL_ARB_clip_control and ARB_depth_buffer_float extensions
- Before this change, line width was hardcoded to 3 pixels and the width argument of mjv_makeConnector was ignored.
- After this change, the width argument is respected and interpreted in units of pixel. The non-standard width unit is ok since lines can only be drawn programmatically.
PiperOrigin-RevId: 516529265
Change-Id: I27bffabba69cde415d2ada67263a887c9240efbf
Tom Erez