p.35 #1 · Announced: RX1R II with 42MP sensor and EVF
dpreview posted their studio samples from the RX1R II but they're rather soft. The center looks sharp but everything else is a bit fuzzy. Hard to tell if it's a focus or methodology issue. Here's the RX1R II compared to the RX1R and A7rII/D8100, downsampled to equivalent resolution:
p.35 #2 · Announced: RX1R II with 42MP sensor and EVF
snapsy wrote:
dpreview posted their studio samples from the RX1R II but they're rather soft. The center looks sharp but everything else is a bit fuzzy. Hard to tell if it's a focus or methodology issue. Here's the RX1R II compared to the RX1R and A7rII/D8100, downsampled to equivalent resolution:
I'm guessing it has as much to do with the setup of their test scene as anything else. If you go back and compare the original RX1/RX1R to the original A7, the results in the corners are pretty much the same as the comparison of the RX1R II and the A7RII.
p.35 #3 · Announced: RX1R II with 42MP sensor and EVF
snapsy wrote:
dpreview posted their studio samples from the RX1R II but they're rather soft. The center looks sharp but everything else is a bit fuzzy. Hard to tell if it's a focus or methodology issue. Here's the RX1R II compared to the RX1R and A7rII/D8100, downsampled to equivalent resolution:
Most likely this represents peripheral forward curvature of field at very close distances.
This is a common characteristic of fast and good wide angles.
It's important to note that this means that the piece of the image you can see as less sharp would be super sharp *if you focussed on it*
Of course you could correct for a flat field at close distance: but why would you unless it was an enlarger lens or a reproduction lens? That correction is not cost free. Other aspects of performance would suffer. And are the kinds of things you (or at least I) photograph at short distance with a wide angle are almost always very 3d and surrounded in bokeh.
Of course a flat field at longer distances matters for landscape and the like where you have a better overall sense of the DOF, and don't want the sharpness to move noticeably in. I haven't owned an RX1 series camera (love to instead of a 35mm lens, but too expensive) but what I've seen suggests that the field is very flat at longer distance.
Of course, maybe it does have a flat field and DPR have messed up, or maybe it does have a flat field but terrible peripheral sharpness at close distances, but they both seem less likely...
(note that the ingterchangeable comparison will be shot from further away with a longer lens)
p.35 #4 · Announced: RX1R II with 42MP sensor and EVF
snapsy wrote:
dpreview posted their studio samples from the RX1R II but they're rather soft. The center looks sharp but everything else is a bit fuzzy. Hard to tell if it's a focus or methodology issue. Here's the RX1R II compared to the RX1R and A7rII/D8100, downsampled to equivalent resolution:
p.35 #5 · Announced: RX1R II with 42MP sensor and EVF
As has been noted in other threads here and other places, the DP review test scenes are shot on most cameras with lenses that are longer than 35mm. Since they fill the image with the test scene, the camera is much further from the test target with cameras shot with longer lenses than with the RX1s. As a result the sensor is relatively further from edges of the test scene vs. the center of the test scene shooting the much wider (and closer) lens. If you wanted to do a fair comparison of the an RX1 series camera to another camera at such short distances, you would have to use a 35 mm lens on that other camera, which they don't. But the whole exercise is really a waste of time because most of us really don't care much about the flat field performance of the RX1 at very short distances.
p.35 #6 · Announced: RX1R II with 42MP sensor and EVF
Jeff Kott wrote:
As has been noted in other threads here and other places, the DP review test scenes are shot on most cameras with lenses that are longer than 35mm. Since they fill the image with the test scene, the camera is much further from the test target with cameras shot with longer lenses than with the RX1s. As a result the sensor is relatively further from edges of the test scene vs. the center of the test scene shooting the much wider (and closer) lens. If you wanted to do a fair comparison of the an RX1 series camera to another camera at such short distances, you would have to use a 35 mm lens on that other camera, which they don't. But the whole exercise is really a waste of time because most of us really don't care much about the flat field performance of the RX1 at very short distances. ...Show more →
Yes, the Nikon D800 comparison was shot with a 85mm, the A7RII with a 55mm and the RX1R/II with a 35mm.....what are we really testing here!
p.35 #7 · Announced: RX1R II with 42MP sensor and EVF
Well, it is useful for anyone who shoots fine art photography. But then, anyone who does that should know that a ~50mm is a much better focal length than a 35mm for just this reason.
Fred Miranda wrote:
This test just shows me that the FE 55/1.8 (Used on the A7RII comparison) performs better at the edges at close range when shooting charts.
It's more interesting to compare it to the RX1R:
The RX1RII jpeg file has less sharpening artifacts (less sharpening applied)
There is less color artifacts on the RX1RII
The RX1RII shows slightly better resolution
Assumption based on my RX1R images: The Sonnar 35/2 lens has stronger field curvature at close range when compared to infinity at f/5.6.
I would not read too much from this test, unless you tend to shoot brick walls at close range.
p.35 #8 · Announced: RX1R II with 42MP sensor and EVF
TheEmrys wrote:
Well, it is useful for anyone who shoots fine art photography. But then, anyone who does that should know that a ~50mm is a much better focal length than a 35mm for just this reason.
Agreed, the RX1 is not the tool for fine art or close up enlargement photography. As others have pointed out, you could correct the lens to have better very close flat field performance, but making that correction would most likely have negative impacts on other positive attributes of the RX1s' lens. Based on looking at lots and lots of images from RX1 series cameras, it seems to me that whoever desgined the lens made all the right compromises.
DP review should put a disclaimer on their studio comparison tool that says you cannot compare the test scene for cameras where the shots were taken with different focal length lenses, but what do you expect.
p.35 #9 · Announced: RX1R II with 42MP sensor and EVF
Fred Miranda wrote:
This test just shows me that the FE 55/1.8 (Used on the A7RII comparison) performs better at the edges at close range when shooting charts.
It's more interesting to compare it to the RX1R:
The RX1RII jpeg file has less sharpening artifacts (less sharpening applied)
There is less color artifacts on the RX1RII
The RX1RII shows slightly better resolution
Assumption based on my RX1R images: The Sonnar 35/2 lens has stronger field curvature at close range when compared to infinity at f/5.6.
I would not read too much from this test, unless you tend to shoot brick walls at close range.
It was interesting to try the three LPF settings. It seems like "normal" (middle setting) eliminates virtually all aliasing with very little loss of sharpness.
p.35 #10 · Announced: RX1R II with 42MP sensor and EVF
Jeff Kott wrote:
Agreed, the RX1 is not the tool for fine art or close up enlargement photography. As others have pointed out, you could correct the lens to have better very close flat field performance, but making that correction would most likely have negative impacts on other positive attributes of the RX1s' lens. Based on looking at lots and lots of images from RX1 series cameras, it seems to me that whoever desgined the lens made all the right compromises.
DP review should put a disclaimer on their studio comparison tool that says you cannot compare the test scene for cameras where the shots were taken with different focal length lenses, but what do you expect. ...Show more →
In the comments, the DPR testers did admit the irrelevancy of testing different focal lengths for comparison purposes. They also said the soft corners were an expected result from shooting this lens on a flat field, and their "real world" testing proves the lens is sharp corner to corner. More published samples are promised.
p.35 #11 · Announced: RX1R II with 42MP sensor and EVF
I noticed that Dteve Huff also posted some first impressions about the camera but doesn't address two of the points that manual focus users are inquiring about. Whether a power on/off will reset the manually focused distance like the RX1 did (I have trouble believing that it'll remember your intended focus and will reset since its fly by wire) and whether peaking only works with magnified view (I have a feeling that it'll be able to use focus peaking zoomed and in regular view). All signs point to the Leica Q having a better implementation of manual focusing and quite possibly better AF as well. If the Q had a 35 lens and a tilt screen on it, I would strongly consider it but the 28 is too wide for me for general shooting and would result in some seriously small 35/50 crops and inability to crop to 70 in a pinch.
p.35 #12 · Announced: RX1R II with 42MP sensor and EVF
I had a Q briefly but switched back to the original RX1R - the Q is pretty significantly larger than the Sony particularly given it has to have some kind of grip added - to me it just felt I may as well have something from the A7 series if I wanted that size camera
p.35 #13 · Announced: RX1R II with 42MP sensor and EVF
To quote myself from another forum: Most likely this represents peripheral forward curvature of field at very close distances. This is a common characteristic of fast and good wide angles. It's important to note that this means that the piece of the image you can see as less sharp would be super sharp *if you focussed on it* Of course you could correct for a flat field at close distance: but why would you unless it was an enlarger lens or a reproduction lens? That correction is not cost free. Other aspects of performance would suffer. And are the kinds of things you (or at least I) photograph at short distance with a wide angle are almost always very 3d and surrounded in bokeh. Of course a flat field at longer distances matters for landscape and the like where you have a better overall sense of the DOF, and don't want the sharpness to move noticeably in. I haven't owned an RX1 series camera (love to instead of a 35mm lens, but too expensive) but what I've seen suggests that the field is very flat at longer distance. Of course, maybe it does have a flat field and DPR have messed up, or maybe it does have a flat field but terrible peripheral sharpness at close distances, but they both seem less likely... (note that the ingterchangeable comparison will be shot from further away with a longer lens)...Show more →
Interesting. It may also be exactly what you say, but it is also then very interesting that the Leica Q uses an even wider lens (and so is even closer to the test target), yet manages to perform flawlessly.
p.35 #14 · Announced: RX1R II with 42MP sensor and EVF
I had a Q briefly but switched back to the original RX1R - the Q is pretty significantly larger than the Sony particularly given it has to have some kind of grip added - to me it just felt I may as well have something from the A7 series if I wanted that size camera
Yes, the 28/1.8 lens is quite long. I would really like a nice flat pancake 40/2 - like the Rokkor/Leitz versions for the M. Its flatness would then make less bulky even than the RX cameras. It would also be a much more useful lens for me, much for the same reasons that TMaG82 mentions.
p.35 #15 · Announced: RX1R II with 42MP sensor and EVF
absolutely. It is endlessly frustrating to me that only Sigma have run with producing fixed lens cameras with longer focal lengths. I'd love a 40mm version of the Leica, Fuji or Sony
p.35 #16 · Announced: RX1R II with 42MP sensor and EVF
wolfloid wrote:
Interesting. It may also be exactly what you say, but it is also then very interesting that the Leica Q uses an even wider lens (and so is even closer to the test target), yet manages to perform flawlessly.
The Q lens has very strong native distortion that requires major correction versus the RX1 Sonnar lens (which does have native distortion but nothing like the Q lens). The result is that Leica can better correct some other lens issues but at the expense of poor off-axis lens performance at distance (due to the extreme digital correction required). So, the Q may perform apparently "flawlessly" at extremely close distances (where it's designed-in distortion may work to it's advantage) but it will not compare well to the RX1 Sonnar at distance outside of the center.
p.35 #17 · Announced: RX1R II with 42MP sensor and EVF
wolfloid wrote:
Interesting. It may also be exactly what you say, but it is also then very interesting that the Leica Q uses an even wider lens (and so is even closer to the test target), yet manages to perform flawlessly.
It could just be the Q lens has a different field curvature characteristic. I'm not really interested in corner performance at MFD unless it's a macro lens.
p.35 #18 · Announced: RX1R II with 42MP sensor and EVF
Tariq Gibran wrote:
The Q lens has very strong native distortion that requires major correction versus the RX1 Sonnar lens (which does have native distortion but nothing like the Q lens). The result is that Leica can better correct some other lens issues but at the expense of poor off-axis lens performance at distance (due to the extreme digital correction required). So, the Q may perform apparently "flawlessly" at extremely close distances (where it's designed-in distortion may work to it's advantage) but it will not compare well to the RX1 Sonnar at distance outside of the center.
lets not forget the rx1 lens has much smoother bokeh than the Q, i'm sure that costs something in terms of lens correction. also, the answer is no the Q does not have as good a corners as the original rx1 at infinity from the comparisons i've seen.
p.35 #19 · Announced: RX1R II with 42MP sensor and EVF
which does have native distortion but nothing like the Q lens)
Is this a known fact, or is it an assumption? Do we know there is no 35/2 RAW correction in camera? This is almost routine for smaller mirrorless cameras these days
