Camera collision: why not cube?


Is there a reason you chose to set an ellipse for camera collision detection instead of a potentially faster cube ?
I understand precision is a factor but could we have this possibility in the future?


You could probably implement it outside of the camera code by managing it on the side as a workaround.

I would say it could be the great start of creating a box collider instead of our current ellipse one.

Would be actually a great contrib for the engine ?