Jay968 wrote:
I think that one of the things I have seen with the 14-24s is that its field curvature very often makes it look as if the corners are not very good. I've tested it against a flat surface and found that it is indeed very sharp corner to corner. But in real world shooting field curvature does some nasty stuff to foreground corners. It's imperative to either get to know how this behaves and work with it, or stop down enough to try and avoid it.
The Photographylife review of the lens which you have provided the link to seems to contradict this.
I'm not exactly sure why, but if I photograph a flat surface, I see very little falloff in sharpness in the corners. However, if I photograph a landscape I see corners ranging from what I would call fair wide open to acceptable at F8 and pretty good at F11, but still discernible from the center of the frame. I'm not sure what those who say the corner sharpness on this lens is superb are looking at or shooting. There is always a noticeable difference in sharpness between the bottom center and bottom corners. If this is not due to field curvature, I don't know what accounts for it. No I don't expect any ultra wide to have perfect corners, but the praise I have seen for corners on this lens is something I just don't understand. I've done some decentering testing using Fred Miranda's technique and while not perfect (what ultra wide zoom is) I see no problematic evidence of decentering.
I don't know Photographylife but I do sometimes wonder if some reviews are based on published MTF curves and not on real hands on use. The MTF curves for the 14-24s are superb. Far better than what I am actually seeing in using the lens.
Most samples that I see from this lens (including your terrific images at Nikon Cafe) seem to be of landscapes which have either dark or non detailed foregrounds. I've been searching for examples which show a lot of detail in the near foreground areas, but they are few and far between. In another forum I asked anyone who owns this lens to take a few simple shots of something beyond 20 feet or so but also showing the foreground closer to the camera. I have yet to see a shot which shows real good edge to edge sharpness in that foreground unless shot at at least F8. However, as I said earlier, a brick wall test makes the lens look superb corner to corner.
If not field curvature, then what is accounting for this?
Pardon me as I’m green on this subject but if flat surfaces are sharp corner to corner and you’re disappointed with foreground sharpness with landscapes, would that mean there is a lack of field curvature?
Post some examples.
Have you seen cameralab’s test shots? Corner sharpness seems to beat out the G version and stomp the 14-30. At least with their copy. Foreground looks to be a different story.
The above links show how field curvature can indeed lead to unsharp corners.
Testing using a test chart does not take field curvature into account as it is a flat surface. Nor does focusing on and including objects only at infinity.
Jay968 wrote:
I think that one of the things I have seen with the 14-24s is that its field curvature very often makes it look as if the corners are not very good. I've tested it against a flat surface and found that it is indeed very sharp corner to corner. But in real world shooting field curvature does some nasty stuff to foreground corners. It's imperative to either get to know how this behaves and work with it, or stop down enough to try and avoid it.
The Photographylife review of the lens which you have provided the link to seems to contradict this.
I'm not exactly sure why, but if I photograph a flat surface, I see very little falloff in sharpness in the corners. However, if I photograph a landscape I see corners ranging from what I would call fair wide open to acceptable at F8 and pretty good at F11, but still discernible from the center of the frame. I'm not sure what those who say the corner sharpness on this lens is superb are looking at or shooting. There is always a noticeable difference in sharpness between the bottom center and bottom corners. If this is not due to field curvature, I don't know what accounts for it. No I don't expect any ultra wide to have perfect corners, but the praise I have seen for corners on this lens is something I just don't understand. I've done some decentering testing using Fred Miranda's technique and while not perfect (what ultra wide zoom is) I see no problematic evidence of decentering.
I don't know Photographylife but I do sometimes wonder if some reviews are based on published MTF curves and not on real hands on use. The MTF curves for the 14-24s are superb. Far better than what I am actually seeing in using the lens.
Most samples that I see from this lens (including your terrific images at Nikon Cafe) seem to be of landscapes which have either dark or non detailed foregrounds. I've been searching for examples which show a lot of detail in the near foreground areas, but they are few and far between. In another forum I asked anyone who owns this lens to take a few simple shots of something beyond 20 feet or so but also showing the foreground closer to the camera. I have yet to see a shot which shows real good edge to edge sharpness in that foreground unless shot at at least F8. However, as I said earlier, a brick wall test makes the lens look superb corner to corner.
If not field curvature, then what is accounting for this?
Pardon me as I’m green on this subject but if flat surfaces are sharp corner to corner and you’re disappointed with foreground sharpness with landscapes, would that mean there is a lack of field curvature?
Post some examples.
Have you seen cameralab’s test shots? Corner sharpness seems to beat out the G version and stomp the 14-30. At least with their copy. Foreground looks to be a different story.
The above shows how field curvature can indeed lead to unsharp corners.
Testing using a test chart does not take field curvature into account as it is a flat surface. Nor does focusing on and including objects only at infinity.
Jay968 wrote:
I think that one of the things I have seen with the 14-24s is that its field curvature very often makes it look as if the corners are not very good. I've tested it against a flat surface and found that it is indeed very sharp corner to corner. But in real world shooting field curvature does some nasty stuff to foreground corners. It's imperative to either get to know how this behaves and work with it, or stop down enough to try and avoid it.
The Photographylife review of the lens which you have provided the link to seems to contradict this.
I'm not exactly sure why, but if I photograph a flat surface, I see very little falloff in sharpness in the corners. However, if I photograph a landscape I see corners ranging from what I would call fair wide open to acceptable at F8 and pretty good at F11, but still discernible from the center of the frame. I'm not sure what those who say the corner sharpness on this lens is superb are looking at or shooting. There is always a noticeable difference in sharpness between the bottom center and bottom corners. If this is not due to field curvature, I don't know what accounts for it. No I don't expect any ultra wide to have perfect corners, but the praise I have seen for corners on this lens is something I just don't understand. I've done some decentering testing using Fred Miranda's technique and while not perfect (what ultra wide zoom is) I see no problematic evidence of decentering.
I don't know Photographylife but I do sometimes wonder if some reviews are based on published MTF curves and not on real hands on use. The MTF curves for the 14-24s are superb. Far better than what I am actually seeing in using the lens.
Most samples that I see from this lens (including your terrific images at Nikon Cafe) seem to be of landscapes which have either dark or non detailed foregrounds. I've been searching for examples which show a lot of detail in the near foreground areas, but they are few and far between. In another forum I asked anyone who owns this lens to take a few simple shots of something beyond 20 feet or so but also showing the foreground closer to the camera. I have yet to see a shot which shows real good edge to edge sharpness in that foreground unless shot at at least F8. However, as I said earlier, a brick wall test makes the lens look superb corner to corner.
If not field curvature, then what is accounting for this?
Pardon me as I’m green on this subject but if flat surfaces are sharp corner to corner and you’re disappointed with foreground sharpness with landscapes, would that mean there is a lack of field curvature?
Post some examples.
Have you seen cameralab’s test shots? Corner sharpness seems to beat out the G version and stomp the 14-30. At least with their copy. Foreground looks to be a different story.
The above shows how field curvature can indeed lead to unsharp corners.
Testing using a test chart does not take field curvature into account as it is a flat surface. Nor does focusing at infinity.
Apr 15, 2021 at 04:58 AM
Previous versions of Jay968's message #15566713 « Nikon Z Images Thread »
