Originally posted by Jonson PL Bjørn Rørslet writes more about exit pupil diameter/entrance pupil diameter and Effective aperture
Thanks for the link. I think, I am now officially confused as well
Anyway, these lenses seem to have P<1:
P=0.4: Rodenstock XR-Heligon 75 mm f/1.1
P=0.4: Rodenstock TV-Heligon 50 mm f/0.75
P=0.8: Noct-Nikkor 58 mm f/1.2
P=0.9: Nikkor-O 55 mm f/1.2 CRT lens
Then, he states that N_eff (effective aperture) = N (1 + m/P) where m is the magnification. I think that this refers to DoF which is not as shallow with P<1 as one would expect. Note, that this effect is null at infinite focus (m=0).
So, lenses with P<1 do exist.
However, the 50 mm Heligon is 80 mm long with a focal plane just 4 mm after the rear lens. Unfeasible to use at infinite focus with an SLR. So, it doesn't prove the point.
The Noct-Nikkor 58 mm f/1.2 is described here:
Nikon | Imaging Products | NIKKOR - The Thousand and One Nights, Tale 16 : Ai Noct Nikkor 58mm f/1.2. It has a hand-polished aspherical front glass element and costs about $3000+ as a used lens. It is an SLR lens.
If Bjørn Rørslet is correct that P=0.8 for this lens, then yes, a f/1.0 could be made for the PK mount. However, I found no further reference that this lens has P<1. So, I reserve a personal opinion here and leave the question open for myself. Esp. as the entire topic is further complicated by the fact that even a small exit pupil may actually be closer to the focal plane than any lens element and thus, still be subject to vignetting...