I’m afraid this doesn’t make sense. You should apply a translation to the camera object (e.g. via camera.position.set( 1, 0, 1 )) but not to the projection matrix.
Please ensure that you actually understand the difference between the view and projection matrix and how both matrices are derived.
Yes. As i said before, you should translate the camera object and thus the respective model matrix.
I’m sorry but i still don’t understand what kind of result you try to achieve. Can you share your current progress with a live demo? Or maybe show some pictures or references of the desired result?
What I want to achieve is the instead of setting camera position I want to change projection of camera. This way it looks like camera is orthographic when user pans it.
I’ve read this thread for several times and still have no idea what you want to achieve.
So far, what I can imagine is that you, possibly, want object’s behaviour whilst panning like sprite objects do.
From what I understand, you just want to change the Field Of View (FOV) of the camera, so that the objects become bigger/smaller but nothing in the scene actually translates?
camera.fov = 75; //change the value to be what you want, the lower the value, the bigger the scene will be
camera.updateProjectionMatrix();
so I think I understand what you want, @s333, and the bad news switching over from perspective to ortho is nontrivial. Your best bet is to create a ortho camera and switch.