Jim Kasson has continued his evaluation of the Fujifilm GFX100’s capabilities at a speedy tempo, which is why we now have one other technological mega-post as we speak of his findings. You can discover the first post here and for these of you which were asking about digital shutter velocity, it’s 1/3 – 1/sixth, which you’ll be able to learn extra about under.
The Last Word – Fuji GFX 100 sharpness, LoCA, focus shift with 110/2
F/2.8 is the sharpest f-stop at this distance
f/2 and f/2.2 are a good quantity softer, however f/2.5 may be very sharp.
You don’t lose all that a lot sharpness even at f/5.6, however the candy spot of the lens is between f/2.5 and f/4.
There’s sufficient focus shift that it’s best to focus on the taking aperture.
The MTF50s obtained are about the identical as I acquired earlier with the identical lens on a GFX 50S.
That signifies that the pixel-level sharpness of the GFX 100 is lower than the GFX 50S, and the picture-level sharpness is about the identical. (This will want some extra wanting in to.)
If you don’t wish to trouble with the graphs, I’ll bottom-line it: there isn’t very a lot LoCA in any respect.
The Last Word – Fujifilm GFX 100 sharpness in comparison with GFX 50s
Now you may see that the GFX 50S has the next modulation switch operate from about 60 cy/mm on up. That appears to suggest that it has a good smaller efficient fill issue than the GFX 100, despite the GFX 100’s a lot finer pitch (assuming the GFX 100 fill issue is 100%, this can be a tough affirmation of Jack Hogan’s work on the GFX 50S microlens dimension a few years in the past). Both curves cease on the Nyquist frequency for every digital camera. You can see that the GFX 50S, with an MTF of greater than 0.6 on the Nyquist frequency, is extra liable to aliasing than the GFX 100, with an MTF of about 0.35 at Nyquist. Both these numbers could also be a bit excessive due to the high-contrast goal, however I believe the ratio must be roughly appropriate.
The Last Word – GFX 100 sharpening in Lightroom
Looking at these curves, it looks like the decreased pixel-level sharpness of the GFX 100 in comparison with the GFX 50S is just not as vital as what are often thought-about reasonable sharpness strikes in Lr or ACR.
The Last Word – Fuji GFX 100 vs 50S sharpness with 3D topic
The shade stability is totally different between the 2 pictures. I’ll put that right down to Adobe’s help of the GFX 100 at the moment being preliminary.
The GFX 100 picture doesn’t look unequivocally sharper, however it’s extra detailed and smoother. But there are different systematic variations:
There is way much less false shade within the GFX 100 picture.
The blue/purple fringing within the GFX 50R picture that seemed to be LoCA isn’t current within the GFX 100 picture, indicating that it was a false shade artifact.
The Last Word – Fuji GFX 100, 50R aliasing variations
The GFX 100 can document finer element earlier than it begins to alias (It could be wonderful if this weren’t true).
The aliasing is most placing with the GFX 50r
The false shade is worse with the GFX 50R
The Last Word –Visual comparisons of Fuji GFX 100 14 and 16 bit uncooked precision
There are some small variations within the noise sample, however they fade into utter insignificance when in comparison with the horizontal banding the comes from the OSPDAF system. If you get rid of the banding in submit manufacturing, the variations may conceivably be value worrying about, however they’re tiny.
The Last Word – Visual comparisons of Fuji GFX 100 and GFX 50R shadow noise
My take: absent postproduction work to mitigate the OSPDAF banding of the GFX 100, and with no noise discount, it has a usable dynamic vary that’s inferior to the GFX 50R. Because of the finer pitch, nonlinear noise discount must be more practical with the GFX 100. I’ll check that within the subsequent submit.
The Last Word – Fujifilm GFX 50R and 100 shadow noise with nonlinear noise discount
The GFX 100 picture is smother, however traces of the PDAF banding that plagues the shadows with that digital camera stay.
The Last Word – Sensors outresolving lenses
There’s your reply. If you’re going to make use of diffraction, lens aberrations, and defocusing to create a powerful sufficient low-pass filter that there’s no seen aliasing, your picture goes to look actually mushy on the pixel degree.
The Last Word – A visible have a look at GFX 100 focus bracketing step sizes
Say you’re doing focus bracketing. Decide how far out of focus is suitable by choosing a just-marginal picture. set the step dimension to that or one unit much less. If you’re doing stacking, determine how a lot softness the stacking program can handle, and choose that step dimension.
The Last Word – How quick is the GFX 100 digital shutter?
14-bit That means it took the shutter 8.33*19.5 = 162 ms to make the journey. That’s about 1/6 of a second.
16-bit Now there are 42 bands. 8.33*42 = 350 ms, or about 1/3 of a second.
The Last Word – Quantifying the Fuji GFX 100 focus bracket step dimension
Regardless of the f-stop, the step dimension is that which produces a few 1 um diameter CoC. The digital camera is compensating for the f-stop to make your life simpler.
The Last Word – A visible have a look at GFX 100 diffraction blur
The first picture with no hint of aliasing is the f/16 one, though the aliasing is kind of low within the f/11 picture. Aliasing is eradicated with much less total blur than within the defocused examples within the earlier submit, indicating that diffraction blur is more practical at countering aliasing than defocusing.
The Last Word – GFX 100 AF-S and AF-C accuracy with 110/2 lens
With the exception of capturing extensive open or almost so, we don’t lose a lot accuracy to get the improved velocity within the GXF 100.
The Last Word – Averaging GFX 100 pictures
The random noise has diminished loads, however the PDAF banding is much more obvious.
Note: I didn’t align the layers, and used EFCS shutter mode. There is a few digital camera movement seen. Next time I do that, I’ll use ES shutter mode.
The Last Word – PDAF banding in GFX 100 in-camera JPEGs
You can nonetheless see the banding, simply not within the darkest areas of the picture. The in-camera JPEG crushes the blacks so exhausting that the banding is much less obvious. It additionally employs heavier noise discount.