bhatnathan
commented
4 years ago What we decided on is that we want the camera to smooth lerp to a point that is the average of a pivot point that is placed in each "room" and the player's current position. This will either limit the size of each room; if we have a constant camera size we will have a max distance of player and pivot be the size of the screen (smaller actually since you always want a bit of visual space around the player); or the camera will have to zoom to fit the room (not always great with pixel games but it works).
I personally prefer to have the size of the rooms be constrained as it gives a more metroidvania feel.
We should also have a "buffer area" between every pair of rooms. Say the player is in room A and is walking into room B, then the player would first cross a trigger that switches the pivot to room A (nothing happens because the camera is already set to that pivot point) and then hits the trigger to switch to room B's pivot point. So if a player walks back and forth quickly on the edge of the room the camera won't switch back and forth between room pivots (due to the distance between triggers).
Some sort of system for the camera so that it follows the player. Take a look at this for inspiration:
https://www.gamasutra.com/blogs/ItayKeren/20150511/243083/Scroll_Back_The_Theory_and_Practice_of_Cameras_in_SideScrollers.php
How the camera will work needs to be discussed and more decisions on this need to be made.