I don't think I was very clear when I asked my question. I guess what I'm getting at is that


1) it *seems* that given the image of point a some specified distance on both sides of focus, one should be able to determine bokeh exactly, modulo off axis aberrations (which are, as far as I know, always considered undesirable).


2) the well corrected out of focus image of a point (not counting diffraction) is the aperture itself.


This seems to indicate that the only way to modulate bokeh is by shape of blades, or by over/under correction. I've often heard, however, that the shape of the blades isn't really that important... that the design of the lens is more so.


It would be interesting to see a comparison of the 50mm f1.2 and 50mm f/1.8 stopped down to, say, f/2.8 using an external round aperture. I wonder how different, if at all, they would be then.