02-15-2021, 12:10 AM
(02-14-2021, 10:22 PM)Klaus Wrote: I'm puzzled ...Well, I have seen similar differences before.
Yesterday I took some real-world pictures - meant to illustrate the issue with the lens. And nothing ... all seems to be in order.
I'm wondering whether the issues that I've seen in the lab were due to some close focus problem. I will have to revisit this .... *sigh*
FWIW ... here are some snapshots: https://photozone.smugmug.com/Canon-RF-50mm-f18-STM
I reckon the testing setup really shows the worst in bokeh possible, rather than what it does for real life shooting. Nothing wrong with that, because it means you can take that into account.
One earlier example was with the EF 50 F/1.2L, where you saw some not so great bokeh. I have never ever seen this happening with my own specimen of this lens, and it actually has always shown flawless bokeh, as in, very smooth transitions to both foreground and background OOF areas, and I just loved it for that. And I doubt very much this is caused by sample variance, as this tends to be caused by overall lens design and placement of diaphragm on the optical axis, IOW, generally not bound by simple lens variance.
It may well be that there is a difference in shooting flat subjects (testing), and 3D real life objects, who knows.
Having said that, other tests I have read so far, seem to indicated that bokeh is not it strongest point, and that it is only marginally better than the latest EF version, and mostly better in the corners, while showing more flare, possibly due to the front lens element being so far forward, compared to the EF versions - IOW, a lens hood is a requirement almost.
Kind regards, WIm
Gear: Canon EOS R with 3 primes and 2 zooms, 4 EF-R adapters, Canon EOS 5 (analog), 9 Canon EF primes, a lone Canon EF zoom, 2 extenders, 2 converters, tubes; Olympus OM-D 1 Mk II & Pen F with 12 primes, 6 zooms, and 3 Metabones EF-MFT adapters ....