(08-03-2021, 05:14 AM)thxbb12 Wrote: Strange in which way?
The borders take a deep dive wide open, center and corners are much better.
Yes, the lens needs to be stopped down to achieve best resolution, but on the two units I used for review center and corner were very good wide open (and borders just fair). For portraiture, that's of course slightly unfortunate behaviour, because you may want to place your subject's eyes where the lens shows the weakest performance. But then, on the other hand, stopping down a bit is common with these kind of portraits anyway (unless you want a blurry nose with the sharp eyes).
(08-03-2021, 05:14 AM)thxbb12 Wrote: Strange in which way?
The borders take a deep dive wide open, center and corners are much better.
Yes, the lens needs to be stopped down to achieve best resolution, but on the two units I used for review center and corner were very good wide open (and borders just fair). For portraiture, that's of course slightly unfortunate behaviour, because you may want to place your subject's eyes where the lens shows the weakest performance. But then, on the other hand, stopping down a bit is common with these kind of portraits anyway (unless you want a blurry nose with the sharp eyes).
You forget one thing, Marcus.
If a lens is sharp over the entire plane of focus, even items just out of focus look quite sharp, certainly not disturbingly unsharp. The Canon RF 50L and RF 85L are testament to that. And with those lenses, at F/1.2 they have a DoF comparable to F/0.9 on APS-C ....
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 ....
Still, a traditional head/shoulder portrait is not something you'd typically shoot wide open with a 85L (or the XF 50/1.0), unless you want to show just how thin DOF can be (instead of producing something that pleases the client).
Both L lenses (and any other fast prime I know) show a way more uniform MTF characteristic than the XF 50/1.0. I attached the official Fuji MTFs and the beginning section of our MTF chart to give you an idea.
(08-03-2021, 10:14 AM)mst Wrote: I know, Wim... I'm a fast prime addict myself
Still, a traditional head/shoulder portrait is not something you'd typically shoot wide open with a 85L (or the XF 50/1.0), unless you want to show just how thin DOF can be (instead of producing something that pleases the client).
Both L lenses (and any other fast prime I know) show a way more uniform MTF characteristic than the XF 50/1.0. I attached the official Fuji MTFs and the beginning section of our MTF chart to give you an idea.
Well, the old EF 85L II really required about F/4 for reasonable DoF for a frame filling head-portrait, i.e., nose tip to eyes and a little bit of cheeks in focus, but funnily enough, if I am careful with both RF 50L and RF 85L I can actually get away with F/1.2, no problem, especially with shots from the side or at an angle. And that is only possible because both are so dang sharp over the whole frame, at any aperture.
And despite that sharpness they both have very pleasing roll-off into the OOF areas, both in front and behind the plane of focus. The EF 85L was really only great behind the plane of focus in that regard, still showing doubling of lines in the background occasionally. The EF 50L was good on both sides, but didn't always handle further away chaotic backgrounds very well. None of this happens with the RF versions.
That, in combination with the sharpness over the whole frame, makes them completely usable at any aperture .
BTW, the EF 135L I use regularly for F/2 and F/2.8 portraits, although I tend to use F/2.8 and F/4 for little people (aka kids, babies) because of their smaller, more strongly curved heads.
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 ....
Yikes, I think there's never been anything like that in the past reviews, and I'm a nerd that memorizes most of the unusual ones.
I've seen things remotely similar to this in charts demonstrating uncorrected field curvature, but not in standardly executed sharpness measurements.
