My only request for a Construct 2, is more optimization of the software itself. Takes a while to load back into the layout editor after previewing in full-screen. Sometimes with a larger project, it will hang forever ... and that is on an 8GB RAM development PC.
I am sure just the simple act of rewriting the program will help already, though.
1) Re-usable code. That would be simply fantastic.
2) The ability to export to Flash would be cool, though probably unlikely. Still, since there seem to be no decent free tools for Flash development, and Contruct is the most awe-inspiringly easy game development tool I've ever used, Flash and Construct would make for a pretty lethal combination.
Transverse collision detection - when object was moving too fast in frame 1 and it skips over an object in frame 2 collision can be backtracked to check the previous frame positions and detect the collision doing the call.
Inheritance tree - parent-child objects to speed up cpu time among other things.
Well ok, but that can be done anyways. I was hoping to keep this off of objects, behaviors, etc, and keep it to things we want in the actual program. Like runtime editing of imagepoints, or an actual z index(0 based? ).
I mean this with all due respect but I can't believe C1 doesn't have sprite indexing. There are alot of cool features, but its basically a MMF2 remake with less extensions until it truly delivers power. I know a list of about a dozen programmers that would easily use C1 (for fast completion) if there were only more power features like this. I also believe that going zero-based was something that should have been done from the start; there's actually no difference in how hard it is to teach, because non-programmers will have to learn either way.
Same as Python's sprite[index]. Would be nice if we could do that for families as well. Perhaps something could be done in the family editor where the user could arrange objects to the order they wanted them.