Thanks Gerald! Not trying to argue with you, but "typing out loud". I'm going to have to think about this more, I'm not getting it. I think Ralph's optimum aperture for contrast was slightly smaller than Rayleigh's optimum for resolution, and maybe the contrast is increased by the dark part of the first diffraction ring? When we're this close to diffraction does the shape of the hole really matter that much? ( In microwaves, which I know more about, it does not matter very much and what matters is that the longest dimension can admit the wavelength and polarization. Elliptical or circular or rectangular slot antennas all function well, with the main difference being polarization. ) In most physical camera shapes, as soon as you are the slightest bit off axis, there is squint in the pinhole so effectively it is an ellipse at all points but one on a flat film plane anyway. And that ellipse has a different orientation if the displacement from center is horizontal or vertical. I wonder if the polarization of the light would matter, like it does for a microwave slot antenna? I never took an optics course in college, so I'm completely ignorant here. Some "light summer reading" is in order now...

This is making me want to build one to see for myself!