Imagine a 3D CGI animated scene (not a stereoscopic scene where you need the glasses but a regular 3D CGI scene).
In the scene there is a main area (let us call this the background element) and the foreground area or foreground element. The background element in the scene has a car driving on the road in the countryside. The foreground element consists of trees and shrubs growing on the trees. The shot is for 4 seconds and the camera move is a simple camera tilt upwards, where the the camera is located up at the crown of a tree and behind that tree and the camera tilts from down to up as it looks at the car driving in the background.
It is recommended to render elements (background, foreground etc.) in separate passes and composite them in post-production.
The question is:
Why is it advisable to render background and foreground elements in separate passes in a 3D CGI animated scene? In the shot example above, why not render the background and foreground together in the same pass?
To boost this argument, I read somewhere online from an excerpt from a book that said the below:
"Moreover, it is also common to render components of 3D animated scenes, such
as the background and foreground elements, in separate passes to be composited
Why is it common to render the components separately in the animated scene?
If you have separate passes, you can choose to manipulate certain items in the compositing application, rather then rerendering everything in 3D, which can take a long time. For example, you could choose to blur the background in the compositing application, or you could choose to brighten or darken it. Similarly, when rendering passes of the scene, you can have---say---a pass just for the highlights, so you can decide to blend the highlights/glints differently in the compositing application, to bring them out more or subdue them. And so on.
Thank you Todd; that makes so much sense! Now I see why separate renders can help great deal.