As i was trying to implement my tile based level in the game i started thinking about performance. Right now the level is set up like this:
[source lang="jscript"]map: [ [2,2,2,0,0,0,0,0,0,0,0,0,0,0,2,2,2,1,1,1,0,0,0,0,0,0,0,0,0,0,0,2,2,2], [2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2], [2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2], [0,0,0,0,0,0,0,0,0,0,0,0,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,0,0,0,0], [0,0,0,0,0,0,0,0,0,0,0,2,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,2,0,0,0,0], [0,0,2,0,0,2,0,0,0,0,0,2,2,0,0,0,0,0,0,2,0,0,2,0,0,0,0,0,2,2,0,0,0,0], [0,0,0,2,2,0,0,0,0,0,0,0,2,0,0,0,0,0,0,0,2,2,0,0,0,0,0,0,0,2,0,0,0,0], [2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2], [2,0,0,0,0,0,0,2,2,2,0,0,0,0,0,0,2,2,0,0,0,0,0,0,2,2,2,0,0,0,0,0,0,2], [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]],[/source]
The width of the screen is equal to 17 of the above 23 tiles in width. Right now, all of the above are pushed into a current level array which the drawing routine later renders to screen. This basically means that all tiles that are not 0 have their own X-Y positions and as the player moves right, all of them are updated with new X positions. I was figuring if this was a bad approach since there are so many tiles to keep track of. Half of them are rendered off-screen but still...
Would it be better to just render the first 17 columns and then have a routine that pushes the next column set into the current level array and remove the "left most" ones as the player moves right and vice versa? Would this improve performance?
The thing is that i would first have to find which column of tiles that are the next in line, pull them out, set their X-Y positions and then push them into the current level array and lastly render them out. Seems equally taxing from a performance stand point so what do you guys think?
Edited by hdnine, 29 October 2012 - 10:00 AM.