Re: GFX 23mm lens vs adapted Canon 24/3.5 ii tilt-shift
D_Sh wrote:
Probably you would not even see that part on 35 mm full frame camera. So no surprise that Canon is weaker at the edges (lens is designed for 36mms wide sensor and not for 44mm).
Just intuitively, a ts24 is likely to have more field curvature when viewed at medium format vs full frame because the cropped version (ff) will not extend as far and avoiding some of the field curvature in the picture. However the distortion is quite low on the ts24v2 so this is not likely to be dramatic. Thus using more of the image circle (medium format) will reveal more field curvature if it exists.
The other thing to keep in mind is that the ts lens are very sensitive to off centre tilt. Eg even if its set to no tilt it can be slightly tilted and this results in poor corners. And sometimes the itlt is on when marked at no tilt.
But the most important thing to consider is that the field curvature of a lens can be concave or convex or flat - or even worse because you can work with convex or concave by choosing your focus spot - moustache shaped. What looks flat close in (focussed well edges and centre) really implies convex field curvature because what works for close will tend to be convex (eg the lens to edge is further than the lens to centre) . And what works well for infinity will tend to be flatter (edge to side is similar to edge to centre depending on the scene). Thus a good close in lens is not necessarily a good infinity lens.
I note that further someone has posted TS24 at a further distance and it looks much better on edges.
So the posted samples may not be definitive - maybe the ts24 works better than 23 at infinity or maybe ts24 copy was not set well or maybe the 23 was just better.
This is a critical factor for me in being interested in fuji 100s because I shoot mountains at infinity with nearby trees and the 23mm will result in bent trees - no acceptable.
Re: GFX 23mm lens vs adapted Canon 24/3.5 ii tilt-shift
D_Sh wrote:
Probably you would not even see that part on 35 mm full frame camera. So no surprise that Canon is weaker at the edges (lens is designed for 36mms wide sensor and not for 44mm).
Just intuitively, a ts24 is likely to have more field curvature when viewed at medium format vs full frame because the cropped version (ff) will not extend as far and avoiding some of the field curvature in the picture. However the distortion is quite low on the ts24v2 so this is not likely to be dramatic. Thus using more of the image circle (medium format) will reveal more field curvature if it exists.
The other thing to keep in mind is that the ts lens are very sensitive to off centre tilt. Eg even if its set to no tilt it can be slightly tilted and this results in poor corners. And sometimes the itlt is on when marked at no tilt.
But the most important thing to consider is that the field curvature of a lens can be concave or convex or flat. What looks flat close in (focussed well edges and centre) really implies convex field curvature because what works for close will tend to be convex (eg the lens to edge is further than the lens to centre) . And what works well for infinity will tend to be flatter (edge to side is similar to edge to centre depending on the scene). Thus a good close in lens is not necessarily a good infinity lens.
I note that further someone has posted TS24 at a further distance and it looks much better on edges.
So the posted samples may not be definitive - maybe the ts24 works better than 23 at infinity or maybe ts24 copy was not set well or maybe the 23 was just better.
This is a critical factor for me in being interested in fuji 100s because I shoot mountains at infinity with nearby trees and the 23mm will result in bent trees - no acceptable.
Re: GFX 23mm lens vs adapted Canon 24/3.5 ii tilt-shift
D_Sh wrote:
Probably you would not even see that part on 35 mm full frame camera. So no surprise that Canon is weaker at the edges (lens is designed for 36mms wide sensor and not for 44mm).
Just intuitively, a ts24 is likely to have more field curvature when viewed at medium format vs full frame because the cropped version (ff) will not extend as far and avoiding some of the field curvature in the picture. However the distortion is quite low on the ts24v2 so this is not likely to be dramatic. Thus using more of the image circle (medium format) will reveal more field curvature if it exists.
The other thing to keep in mind is that the ts lens are very sensitive to off centre tilt. Eg even if its set to no tilt it can be slightly tilted and this results in poor corners. And sometimes the itlt is on when marked at no tilt.
But the most important thing to consider is that the field curvature of a lens can be concave or convex or flat. What looks flat close in (focussed well edges and centre) really implies convex field curvature because what works for close will tend to be convex (eg the lens to edge is further than the lens to centre) . And what works well for infinity will tend to be flatter (edge to side is similar to edge to centre depending on the scene). Thus a good close in lens is not necessarily a good infinity lens.
I note that further someone has posted TS24 at a further distance and it looks much better on edges.
So the posted samples may not be definitive - maybe the ts24 works better than 23 at infinity or maybe ts24 copy was not set well or maybe the 23 was just better.
This is a critical factor for me in being interested in fuji 100s because I shoot mountains at infinity with nearby trees and the 23mm will result in bent trees - no acceptable.
Re: GFX 23mm lens vs adapted Canon 24/3.5 ii tilt-shift
D_Sh wrote:
Probably you would not even see that part on 35 mm full frame camera. So no surprise that Canon is weaker at the edges (lens is designed for 36mms wide sensor and not for 44mm).
Just intuitively, a ts24 is likely to have more field curvature when viewed at medium format vs full frame because the cropped version (ff) will not extend as far and avoiding some of the field curvature in the picture. However the distortion is quite low on the ts24v2 so this is not likely to be dramatic. Thus using more of the image circle (medium format) will reveal more field curvature if it exists.
The other thing to keep in mind is that the ts lens are very sensitive to off centre tilt. Eg even if its set to no tilt it can be slightly tilted and this results in poor corners. And sometimes the itlt is on when marked at no tilt.
But the most important thing to consider is that the field curvature of a lens can be concave or convex or flat. What looks flat close in (focussed well edges and centre) really implies convex field curvature because what works for close will tend to be convex (eg the lens to edge is further than the lens to centre) . And what works well for infinity will tend to be flatter (edge to side is similar to edge to centre depending on the scene). Thus a good close in lens is not necessarily a good infinity lens.
I note that further someone has posted TS24 at a further distance and it looks much better on edges.
So the posted samples may not be definitive - maybe the ts24 works better than 23 at infinity or maybe ts24 copy was not set well or maybe the 23 was just better.
This is a critical factor for me in being interested in fuji 100s because I shoot mountains at infinity with nearby trees and the 23mm will result in bent trees - no acceptable.
Feb 12, 2021 at 08:56 PM
Previous versions of Scott Stoness's message #15499758 « GFX 23mm lens vs adapted Canon 24/3.5 ii tilt-shift »
