Fairy tales about 'diffraction'

[agorabasta]

Greg,

I actually avoided the term 'Airy disk'. (http://en.wikipedia.org/wiki/Airy_disk) That's because that specific diffraction pattern happens only for the images of point-like sources, like stars in the night skies. And such point sources are not so rare.
But diffraction patterns fairly related to the Airy disks easily do happen for the extended objects like thin wires against the bright skies, or around every sharp edge in the picture. The only scenario for diffraction to never show at all is in an image with absolutely no sharp detail in it. That's hardly an image you normally intend to take.

The important feature distinguishing the diffraction blur from, say, the Gaussian blur is that the diffraction pattern commands a totally zeroed value at a certain distance from the peak. That zero value has no other chance than to intermodulate (produce coloured beat patterns much like moire but of more complex shapes) with the regular ordered digital sensor arrays.

The fact of life is that there are hardly many a lens allowing that to happen, though there are very few high-quality primes that do. So most of the lenses available have some other blurring 'mechanisms' that are yet stronger than diffraction.

[agorabasta]

Telecentricity means the exit pupil at infinity, by definition. Exit pupil at infinity means the output flux is always collimated, regardless if lens is focused or not.

Sure. But first—telecentric lenses usually are not found in the average hobbyist photographer's closet.

Rubbish.
Image-side telecentricity is a pretty common thing among lenses designed for digital cameras. The sensor nature simply requires that.
(Actually, every cheap supezoom in existence becomes very much telecentric at its long end.)
That's also why you should put the light source at the front rather than at the back element - because object-side telecentric lenses are as rare as hen's teeth.
http://en.wikipedia.org/wiki/Telecentri ... ric_lenses

And second—they're not required anyway, because nothing is keeping you from properly adjusting the distance between light source and collimator.

Rubbish.
Getting a collimated flux from a non-telecentric lens requires focusing the lens on the light source, and you immediately get the image of the light source over the target slide, thus having to use darn good diffusers to avoid problems.

The rest of your post is mere regurgitation of the all the same old rubbish, so... you know.

[01af]

Image-side telecentricity is a pretty common thing among lenses designed for digital cameras. The sensor nature simply requires that.
(Actually, every cheap superzoom in existence becomes very much telecentric at its long end.)

Yeah, sure ...

If you are a professional physicist then I am Peter Pan.

[agorabasta]

... I am Peter Pan.

Thanks for the clarification. I always suspected I was talking to a kid...