I’m currently working on a project in which the camera has a 2D, more-or-less top-down view of the world, and tracks the player. The lens is an OrthographicLens.
For certain scenes, I’m experimenting with zooming the camera out a little, and I see offhand two ways of achieving this:
- Scaling the camera.
- (Specifically, in X and Z, leaving Y unchanged, due to the camera’s rotation.)
- Altering the lens’s film-size
I’ve tried out both, and both seem to work.
What I’m stuck on, then, is the question of which is the wiser approach. So: Are there any advantages or disadvantages that either might bring?
Changing the film size will not affect the transformation of objects that are children of the camera node, and based on this we can conclude that this is more correct.
Hmm… That’s a good point.
However, I think that in my specific case it actually argues in favour of changing scale: The main thing that I have attached to my camera, I believe, is a backdrop-card that is intended to cover the whole screen.
(Although I may want to check a few other cases, in case they behave differently…)
Indeed, I’ve already observed that changing the film size causes the backdrop-card to no longer fill the screen, calling for explicit resizing, I believe.