Those are basically the same thing in the end. Since generating "ideas" in the abstract is trivial. The interesting part is exactly figuring out if a given idea is worth anything or not.

All of that said, this topic has been beat. to. death. here on HN. Use the search box and you'll find more discussion of generating startup ideas than you can shake a stick at.