This is a question for more advanced users, or at least for the users who strongly use pathfinding behavior.
I intend to create a huge layout and use pathfinding on it. Object A will try to chase object B in a complicated maze. It will use pathfinding behavior to do so - any other solutions of navigating and moving would be too complicated to implement, hence the PATHFINDING. Now - the pathfinding solution of route from object A to object B should be fast and done twice a second (it is sufficient for the purpose) because object A and object B are constantly moving.
Now for the proper question: despite the huge layout and obstacle map in the memory, object A will chase object B only when both are visible on the screen. How does it relate to CPU performance? The question behind is - how algorythm is performing calculations of pathfinding: does limiting distance from A to B gives any benefits? My intuition tells me yes, but I'm curious.