So, ignore my post about screen filtering + final fantasy battle menus being my major slowdown. After more experimenting, it is clear that, while screen filtering does slow things down (as is to be expected), the pad overlay slows things down more. If I disable the pad overlay, final fantasy 9 runs a bit slow, but still very much playable. Seems the real optimization work needs to be done on overlays.
Otherwise, amazing work schtruck. One question, though. Isn't fpse open source, so aren't you required to release the source for a port? Not criticizing, just would be nice to be able to contribute some code, and test custom builds and the like.