New ef 50mm 1.2 annoucement coming?

[rfdesigner]

The size of the rear element has nothing to do with the lens maximum aperture. It's a relationship between the focal legnth, the iris size and objective lens size. It's quite possible to make an f1.2 lens with the rear element the same size as the aperture / Iris hole.

The f-number gives you the the apparent size of the aperture as seen from the sensor, i.e. the angle under which the aperture appears from a given point of the sensor. So the rear element has to have at least the diameter d/N, where d is the distance of the rear element from the focal plane (sensor) and N is the f-number.
I think smaller diameters would violate Liouville's theorem.

To quote from Wikipedia:
An f stop is the focal length divided by the size of the entrance pupil.

No mention of the rear element, which it may be assumed to the the same size or larger than the pupil...it's not mentioned anywhere.

I have previously discussed this misunderstanding at great lengths. The reality is that the f-number is both of these relationships: f-number = focal length / entrance pupil diameter = exit pupil to image plane distance / exit pupil diameter. The quantities involved are not independent.

I have also pointed to documentation, written by an actual lens designer at Zeiss, if nobody cares to believe what I say, since on the internet, anyone can claim anything:

http://www.zeiss.com/content/dam/Photography/new/pdf/en/cln_archiv/cln35_en_web_special_bokeh.pdf

Basic and elementary reasoning should force one to arrive at the conclusion that if what the sensor sees of the exit pupil is only a tiny aperture of light, the f-number cannot possibly be, say, f/1.2 or f/1.0. You cannot, by doing something with the optical path, "concentrate" the intensity of photons passing through that tiny aperture somehow: you can only lose photons through losses in transmission, for a fixed number of photons reflected off the scene and reaching the lens.

Furthermore, to address the earlier comment about Nikon's maximum relative aperture of the F mount,

http://www.pierretoscani.com/echo_shortpres.html#shortpres04

This should again put to rest various nebulous claims and speculations. Thorough research and presentation of evidence and a basic understanding of optics should be more than sufficient.

+1

light cone approaching the sensor will be f1.2 on a f1.2 lens, Canon sensor to mount distance is ~44mm.

to get f1.2 you need at least 36mm diameter rear element.. and that will only get you f1.2 in the centre of the field.

In the corners of a full field lens/camera combination you won't beat f2.7 (sigma 85mm f1.4)

That's one thing I like about my 100f2.0 it achieves f3.4 in the corners.. even if I sign a blank cheque to canon I can't get much more than half a stop improvement in the corners.

If you want a lens that's faster in ther corners then you need a bigger mount.

 

[YuengLinger]

to all those who have called the Canon 1.2 soft and begun using replacements ... which ones do have AF (sorry, Zeiss) and are sharp? Is the Sigma worth checking out? Does it have AF?

I'm getting ready to ebay my Canon off.

I have the Sigma 50 Art is wonderfully crisp even when used in F1.4. In some bodies Canon, Sigma Art has trouble focusing through the viewfinder (phase-detection), but works well with Live View, and optimally well with Dual Pixel AF.

I know Bokeh is subjective, but for me the quality of the bokeh Sigma 50 Art, is better than the Canon 50L. Yes, the 50L is capable of a larger background blur, but Sigma Art has a smoothness on the boundary of the "balls" Bokeh, which pleases me very much.

Most of us longing for a better 50mm 1.2 have already given the Sigma Art a try. Or several tries. Abject failure with 5DIII AF for me. But the 35mm Art is fine.

Live View? Are you serious?

Anecdotal claims of the Art being more reliable with the 7DII don't convince or impress.

Personally, I found the 50mm Art tended to be a bit dark, harsh...Call it grungy. But adequate--when it hit.