07-09-2020, 12:11 PM
(This post was last modified: 07-09-2020, 12:22 PM by Brightcolours.)
Wim, you keep on "calculating" really off diffraction limits, I have pointed out a few times before that what you calculated is not correct. I gave various sources that showed real diffraction figures, including Nikon documentation.
You mention "sensor diffraction", which is not a thing.
Klaus is correct, for f5.6 max. resolution of a green-yellow wavelength (0.55µ) for FF is about 60mp, and for a red wavelength (0.7µ) bout 38mp. So, for white light, you can see a start of diffraction limiting at f7.1 on the R5, if the lens is sharp enough wide open.
https://luminous-landscape.com/do-sensors-out-resolve-lenses/
Lumious landscape's calculations of green light diffraction are totally in line with the Nikon literature which accompanied their professional industrial ultra micro Nikkor lens line-up (lenses which were used in the 1970/1980's semiconductor production, so I am very sure Nikon knew what they were talking about and what they were writing).
Even Photozone/Opticallimit/Lenstip's tests show that diffraction impacts lenses past f5.6 on high MP FF bodies.
Btw, the 30% losses you attribute to demosaicing and anti aliasing are also off (too high).
However, with Canon's deconvolution in DPP, some diffraction softening losses can be undone, so f7.1 will not be much of an issue using this lens on an R5.
You mention "sensor diffraction", which is not a thing.
Klaus is correct, for f5.6 max. resolution of a green-yellow wavelength (0.55µ) for FF is about 60mp, and for a red wavelength (0.7µ) bout 38mp. So, for white light, you can see a start of diffraction limiting at f7.1 on the R5, if the lens is sharp enough wide open.
https://luminous-landscape.com/do-sensors-out-resolve-lenses/
Lumious landscape's calculations of green light diffraction are totally in line with the Nikon literature which accompanied their professional industrial ultra micro Nikkor lens line-up (lenses which were used in the 1970/1980's semiconductor production, so I am very sure Nikon knew what they were talking about and what they were writing).
Even Photozone/Opticallimit/Lenstip's tests show that diffraction impacts lenses past f5.6 on high MP FF bodies.
Btw, the 30% losses you attribute to demosaicing and anti aliasing are also off (too high).
However, with Canon's deconvolution in DPP, some diffraction softening losses can be undone, so f7.1 will not be much of an issue using this lens on an R5.