I'm currently blocking out levels for my game, and was wondering if there are any performance hits scaling collision objects to very large sizes vs using multiple collision objects.
Here is an example.
<img src="https://dl.dropboxusercontent.com/u/95028662/scalingQ.jpg" border="0" />
These objects are on an invisible layer and must be sprites to work with the Physics behavior. I would assume 2 Scaled Objects would be better since it reduces the number of objects and collision checks, but that's just a guess. Are there any tradeoffs to using this scaling method vs multiple objects?