Method 1: One tilemap of an image with the solids behavior serving as the floor, and one tilemap of the same image without the solids behavior and serves as the background.
Method 2: One tilemap of an image that is just the background, and an invisible tilemap that is only a 16 x 16 image with the solids behavior and is tiled wherever it must be solid.
Here's a link to a capx showing my example
I'd like to know which method is better performance wise. Thanks!
Ashley, you might know this answer better than anyone though
the efficient one is the graphics with solid... and you can also edit a poligon collision for every tile....tilemap is the best method to make a large level, and save a lot of memory...
but you can also conbine a tilemap graphics and tilemap solid (only one square) like your example, with this, you can make a player go inside to some holes/make some parallax effect etc...etc...