I am curious about how VFP handles things behind the scenes when you're using OOP. When you instantiate an object, where are all the methods and properties and properties stored? I would think, since VFP tends to construct most of its higher level operations on tables, that it wouldn't be necessary to load all these elements into memory, that object instantiation would simply involve setting pointers to the existing VCX table.

This is an important issue when you evaluate the advantages of using COM objects vs. native objects with VFP. In Mike Myer's article in FPA (Oct 2000), he makes a good point of how important it is to keep COM objects slim, to keep your overhead low (and having to load everything into RAM)...but I would tend to that VFP's native objects would have an advantage here because of the lower overhead.

It would be helpful to have an expert opinion here.