Ok.. so taking a look at this FAKE screenshot.
My original intention was to do a "tile system" ala Dragon Warrior or Final Fantasy for NES (where a tile is say "30x34" pixels as well as the hero's sprite is 30x34 and takes up an ENTIRE tile... etc). This would make collision detection as well as triggered events SUPER easy to do. (and I'm sure this is how it was done in the late 80's early 90's).
But I realized today that isn't going to be enough for me, not even for a "proof of concept" release. I've looked into the idea of using an array for each pixel (suggested by a fellow RPG coder making a zelda game on his TI-83 heh) that has what color the pixel should be, the collision boolean etc. But this seems sort of impractical from a coding standpoint.
So back to the "screenshot". Ideally being able to do something like that would work, but what do you guys think is the most logical way to do it? Should the backdrop be loaded seperately as well as a predefined array for collision? That seems unlikely since it would take weeks just to program a 6or so backdropped "island" area.
Option two (which I think I'm going with as of now..) : load a green "grass" backdrop, then load the tree sprites on top of that, using their starting location and their width I can set a boundry dynamically something like:
void addSprite(int x, int y, int width, int height, int collision) where "collision" would be a 1 or 0 saying whether or not that sprite was crossable... (could be more complex as time goes on..)
Is there a BETTER way to do it that I'm not seeing? How was a game like Zelda: LTTP (which the sprites obviously are from) done?
Any input is appreciated...