This topic is in the 2015 archive. You can read it and like its posts here. It is read only.
Archive 2015Comparison of Zeiss 25/2 ZE and Canon L lenses
p.2 #1Comparison of Zeiss 25/2 ZE and Canon L lenses
I do not understand these two statements which directly contradict each other. Presumably "get" should be "at" and FC means field curvature and WO means wide open.
wayne seltzer wrote:
Your lens test shots do not expose the ZE 25/2's weakness of significantly blurring get [sic] the extreme corners of the image due to some extreme FC.
[...]
The ZE 25/2's main strength is its great sharpness across the frame WO
wayne seltzer wrote:
Your lens test shots do not expose the ZE 25/2's weakness of significantly blurring get [sic] the extreme corners of the image due to some extreme FC.
[...]
The ZE 25/2's main strength is its great sharpness across the frame WO
Dec 19, 2015 at 01:42 PM
p.2 #2Comparison of Zeiss 25/2 ZE and Canon L lenses
Great work Jim. I would have to say, I think your 24-105 is an absolutely spectacular example of that lens! The one I tested briefly, was not anywhere near the level of your copy of that lens.
Dec 19, 2015 at 02:00 PM
p.2 #3Comparison of Zeiss 25/2 ZE and Canon L lenses
Thanks for putting in all the effort on the nice comparison. Glad to see that my 16-35 performs well. One side comment about your scoring system. I think if two lenses are equal, the third best should get a score of 3. Why should the score that the third lens receives benefit from the first two being equal? It is still the third best. In other words, it could get a different score while having identical performance and still being the third best...only because the two better than it are not noticeably different. Not sure that makes sense in my mind.
Dec 19, 2015 at 02:18 PM
p.2 #4Comparison of Zeiss 25/2 ZE and Canon L lenses
The test tells me:1. I am glad I have never owned a 24-105 and 2. I don't need a 25mm Zeiss if I ever get a 16-35mm F 4.0
Dec 19, 2015 at 02:46 PM
p.2 #5Comparison of Zeiss 25/2 ZE and Canon L lenses
For me the 16-35 looks to have the most resolving power and best contrast. Zeiss has the same quality but because of the zoom being a zoom, I think that trumps the prime lens. And this coming from someone who only owns prime lenses...................
Dec 19, 2015 at 03:04 PM
p.2 #6Comparison of Zeiss 25/2 ZE and Canon L lenses
wayne seltzer wrote:
Your lens test shots do not expose the ZE 25/2's weakness of significantly blurring get the extreme corners of the image due to some extreme FC...
Hi Wayne,
I'm not particularly concerned about the extreme corners. I wanted to determine if the edges reached 'excellent' resolution, as shown in PhotoZone tests. I'm happy with the results. Also, the extreme corners aren't bad, just not as good. This lens is extremely well suited for the uses that I plan for it.
The lighting for this particular test was perfect for the resolution comparisons that I wanted to do. I'm well aware of the other characteristics that Zeiss lenses bring to the light table. That's why most of my 'second string' lenses are Zeiss Contax and ZE. Images of mossy rocks in a stream would probably be an interesting study, but that's not what's on the table today.
Your lens test shots do not expose the ZE 25/2's weakness of significantly blurring get the extreme corners of the image due to some extreme FC...
Hi Wayne,
I'm not particularly concerned about the extreme corners. I wanted to determine if the edges reached 'excellent' resolution, as shown in PhotoZone tests. I'm happy with the results. Also, the extreme corners aren't bad, just not as good. This lens is extremely well suited for the uses that I plan for it.
The lighting for this particular test was perfect for the resolution comparisons that I wanted to do. I'm well aware of the other characteristics that Zeiss lenses bring to the light table. That's why most of my 'second string' lenses are Zeiss Contax and ZE. Images of mossy rocks in a stream would probably be an interesting study, but that's not what's on the table today.
Dec 19, 2015 at 03:17 PM
p.2 #7Comparison of Zeiss 25/2 ZE and Canon L lenses
Don Clary wrote:
Great work Jim. I would have to say, I think your 24-105 is an absolutely spectacular example of that lens! The one I tested briefly, was not anywhere near the level of your copy of that lens.
Thanks, Don. I've always been pretty happy with the results from my 24-105/4L IS, and I've used it a lot over the last eight years. As Norm Abram would say, '"it's not the sharpest knife in the drawer", but it's a fine choice for when you don't know what to expect. I use it and the 100-400L IS II for my "car kit".
Great work Jim. I would have to say, I think your 24-105 is an absolutely spectacular example of that lens! The one I tested briefly, was not anywhere near the level of your copy of that lens.
Thanks, Don. I've always been pretty happy with the results from my 24-105/4L IS, and I've used it a lot over the last eight years. As Norm Abram would say, '"it's not the sharpest knife in the drawer", but it's a fine choice for when you don't know what to expect. I use it and the 100-400L IS II for my "car kit".
Dec 19, 2015 at 03:22 PM
p.2 #8Comparison of Zeiss 25/2 ZE and Canon L lenses
drewmey wrote:
Thanks for putting in all the effort on the nice comparison. Glad to see that my 16-35 performs well. One side comment about your scoring system. I think if two lenses are equal, the third best should get a score of 3.
I sometimes skip a level if the difference is large (and that didn't happen in this case), but when they're close, I think it's appropriate. After all, if you're in an Olympic event where two people tie for gold, the next person still gets silver.
Thanks for putting in all the effort on the nice comparison. Glad to see that my 16-35 performs well. One side comment about your scoring system. I think if two lenses are equal, the third best should get a score of 3.
I sometimes skip a level if the difference is large (and that didn't happen in this case), but when they're close, I think it's appropriate. After all, if you're in an Olympic event where two people tie for gold, the next person still gets silver.
Dec 19, 2015 at 03:25 PM
p.2 #9Comparison of Zeiss 25/2 ZE and Canon L lenses
Monito wrote:
Thank you for doing a thorough test of an interesting set of lenses, Jim.
It would be interesting to see the same set of comparison images, after the best post-processing has been applied -- after distortion, aberration, vignetting correction, and some modest degree of sharpening. Interesting unless I have to do the work
thanks again.
First, I should have also offered my thanks to Jim for conducting this test and sharing the results. This isn't easy stuff to do, and it is difficult to do it well.
I, too, have often wanted to see some sort of test in which a few photographers who are skillful and talented in every aspect of the workflow (from capture to print, as they say) take the gear under examination and use it the way they would in the real world — making decision at the time of exposure based on the strengths/weaknesses of the gear and then using a post-processing workflow designed to optimize the final output quality of a print or screen-based output.
There are a few aspects of such an admittedly subjective approach that could prove interesting:
- To what extent would the results mirror or not the results of objective measurements of the gear, and what might that imply?
- What is the magnitude of the differences in the final output — are things that can be measured on the test bench visible "in the flesh?"
- What are the elements of the varied approaches that photographers use to maximize the quality of the final output?
To cite one potential test case... sometimes the primary observable image quality difference between two lenses might be something like CA. If normal post-processing equalizes performance in that regard, who is to say which lens is better in real-world performance. Or how about a circumstance where lens A has more CA but a larger focal length range, while lens B seems to have less CA but half the focal length range? If after post processing, all else being equal, both exhibit the same amount of CA, which would we prefer?
It could be argued — accurately! — that this would not be a very objective test and that it would not only fail to minimize variables but actually embrace them. On the other hand, when we imagine that, say, starting with a raw file from different cameras eliminates variables, it turns out that this isn't quite accurate either — as we know, raw files are not precisely raw files any more, and they often hold data that have been processed and altered in-camera in various ways that are not consistent from case to case. And, in the end, while test results are useful and interesting, the most important thing is the best results that talented photographers can obtain using various kinds of equipment. Of course, if one doesn't post-process (and shoots raw?) then the SOOC quality is the thing, I guess. ;-)
Seems like it would be worth it for someone to give this a shot!
Monito wrote:
However, in favour of doing post-proc analysis we must note that for most photographers including me, any serious image that I print large has gone through post-proc correction for distortion, aberration, vignetting, and sharpening. If I understand, I think this is true for Dan's pictures too, hence his conception of this kind of test result presentation and analysis.
Absolutely true. I use different post-processing procedures for different lenses, different cameras, different subjects and different interpretations.
Dan
Thank you for doing a thorough test of an interesting set of lenses, Jim.
It would be interesting to see the same set of comparison images, after the best post-processing has been applied -- after distortion, aberration, vignetting correction, and some modest degree of sharpening. Interesting unless I have to do the work
thanks again.First, I should have also offered my thanks to Jim for conducting this test and sharing the results. This isn't easy stuff to do, and it is difficult to do it well.
I, too, have often wanted to see some sort of test in which a few photographers who are skillful and talented in every aspect of the workflow (from capture to print, as they say) take the gear under examination and use it the way they would in the real world — making decision at the time of exposure based on the strengths/weaknesses of the gear and then using a post-processing workflow designed to optimize the final output quality of a print or screen-based output.
There are a few aspects of such an admittedly subjective approach that could prove interesting:
- To what extent would the results mirror or not the results of objective measurements of the gear, and what might that imply?
- What is the magnitude of the differences in the final output — are things that can be measured on the test bench visible "in the flesh?"
- What are the elements of the varied approaches that photographers use to maximize the quality of the final output?
To cite one potential test case... sometimes the primary observable image quality difference between two lenses might be something like CA. If normal post-processing equalizes performance in that regard, who is to say which lens is better in real-world performance. Or how about a circumstance where lens A has more CA but a larger focal length range, while lens B seems to have less CA but half the focal length range? If after post processing, all else being equal, both exhibit the same amount of CA, which would we prefer?
It could be argued — accurately! — that this would not be a very objective test and that it would not only fail to minimize variables but actually embrace them. On the other hand, when we imagine that, say, starting with a raw file from different cameras eliminates variables, it turns out that this isn't quite accurate either — as we know, raw files are not precisely raw files any more, and they often hold data that have been processed and altered in-camera in various ways that are not consistent from case to case. And, in the end, while test results are useful and interesting, the most important thing is the best results that talented photographers can obtain using various kinds of equipment. Of course, if one doesn't post-process (and shoots raw?) then the SOOC quality is the thing, I guess. ;-)
Seems like it would be worth it for someone to give this a shot!
Monito wrote:
However, in favour of doing post-proc analysis we must note that for most photographers including me, any serious image that I print large has gone through post-proc correction for distortion, aberration, vignetting, and sharpening. If I understand, I think this is true for Dan's pictures too, hence his conception of this kind of test result presentation and analysis.
Absolutely true. I use different post-processing procedures for different lenses, different cameras, different subjects and different interpretations.
Dan
Dec 19, 2015 at 03:27 PM
p.2 #10Comparison of Zeiss 25/2 ZE and Canon L lenses
My best ~24mm lens for landscape use stopped down is Distagon ZF 2.8/25. Canon 2.8/24 IS is better at apertures up to f5.6, but it doesn't get much better beyond and has more CA at small apertures.
Recently I purchased Distagon ZE 2/25, to replace my ZE 2/28 and ZF 2.8/25. It's incredible WO, sadly the extreme corners never reach the quality of ZF 2.8/25 at f8.
24-105L IS is very usable on my 1DsIII with help of DPPs DLO.
I didn't tested 16-35L IS yet.
Recently I purchased Distagon ZE 2/25, to replace my ZE 2/28 and ZF 2.8/25. It's incredible WO, sadly the extreme corners never reach the quality of ZF 2.8/25 at f8.
24-105L IS is very usable on my 1DsIII with help of DPPs DLO.
I didn't tested 16-35L IS yet.
Dec 22, 2015 at 09:49 AM
p.2 #11Comparison of Zeiss 25/2 ZE and Canon L lenses
vpPhoto wrote:
My best ~24mm lens for landscape use stopped down is Distagon ZF 2.8/25. Canon 2.8/24 IS is better at apertures up to f5.6, but it doesn't get much better beyond and has more CA at small apertures.
Recently I purchased Distagon ZE 2/25, to replace my ZE 2/28 and ZF 2.8/25. It's incredible WO, sadly the extreme corners never reach the quality of ZF 2.8/25 at f8.
24-105L IS is very usable on my 1DsIII with help of DPPs DLO.
I didn't tested 16-35L IS yet.
Thanks for this info. A question...You say you had the 28mm f2. How does the 28mm compare with the 25mm f2? I'm considering one of these and any comparisons of these is appreciated.
My best ~24mm lens for landscape use stopped down is Distagon ZF 2.8/25. Canon 2.8/24 IS is better at apertures up to f5.6, but it doesn't get much better beyond and has more CA at small apertures.
Recently I purchased Distagon ZE 2/25, to replace my ZE 2/28 and ZF 2.8/25. It's incredible WO, sadly the extreme corners never reach the quality of ZF 2.8/25 at f8.
24-105L IS is very usable on my 1DsIII with help of DPPs DLO.
I didn't tested 16-35L IS yet.
Thanks for this info. A question...You say you had the 28mm f2. How does the 28mm compare with the 25mm f2? I'm considering one of these and any comparisons of these is appreciated.
Dec 22, 2015 at 10:02 AM
p.2 #12Comparison of Zeiss 25/2 ZE and Canon L lenses
If Canon releases a new 16-35/2.8 with the sharpness of the f4 version then I am all set. Without AF Zeiss is just not in the running no matter what perceived optical qualities it has.
Dec 22, 2015 at 10:27 AM
p.2 #13Comparison of Zeiss 25/2 ZE and Canon L lenses
Monito wrote:
I do not understand these two statements which directly contradict each other. Presumably "get" should be "at" and FC means field curvature and WO means wide open.
[...]
The ZE 25/2's main strength is its great sharpness across the frame WO
Sorry, new phone system's auto-correct is a little too crazy.
Should have been no "get" in that sentence.
Lens is blurry in the extreme corners but fine most of the way across the frame. It bothered me to much in landscape shots. Others might not notice or care. This lens is more optimized for shooting wide open and gets only moderately sharper stopped down but not any sharper than ZE 21 even though it is a less wide wide angle lens.
Jim, did you do any non- infinity distance comparison shots, say like the side of a building?
I do not understand these two statements which directly contradict each other. Presumably "get" should be "at" and FC means field curvature and WO means wide open.
wayne seltzer wrote
Your lens test shots do not expose the ZE 25/2's weakness of significantly blurring get [sic] the extreme corners of the image due to some extreme FC.[...]
The ZE 25/2's main strength is its great sharpness across the frame WO
Sorry, new phone system's auto-correct is a little too crazy.
Should have been no "get" in that sentence.
Lens is blurry in the extreme corners but fine most of the way across the frame. It bothered me to much in landscape shots. Others might not notice or care. This lens is more optimized for shooting wide open and gets only moderately sharper stopped down but not any sharper than ZE 21 even though it is a less wide wide angle lens.
Jim, did you do any non- infinity distance comparison shots, say like the side of a building?
Dec 22, 2015 at 11:39 AM
p.2 #14Comparison of Zeiss 25/2 ZE and Canon L lenses
wayne seltzer wrote:
...Jim, did you do any non- infinity distance comparison shots, say like the side of a building?
Hi Wayne,
Nope. I might do that later, but the "nearly infinity" condition was the one of highest importance for me to decide if I would keep it.
...Jim, did you do any non- infinity distance comparison shots, say like the side of a building?
Hi Wayne,
Nope. I might do that later, but the "nearly infinity" condition was the one of highest importance for me to decide if I would keep it.
Dec 22, 2015 at 02:40 PM
p.2 #15Comparison of Zeiss 25/2 ZE and Canon L lenses
I have the ts24, zeiss 25/2, 24-105, and 24-70-2.8-v1.
I always carry my zeiss 25/2 (plus ts17, 24-105) because:
1) Its small
2) f2 with quality
3) It complements the 24-105 which is not great at 24 but very good beyond 30mm
4) I don't find I need the shift as much at 24mm (Ts24) as 17mm plus in a pinch I can use my 1.4x with TS17 which is the lens I base my system on. Or 7dii with TS17.
5) f2 and hardstop at infinity and 25mm make it the perfect lens for me nightscape lens with northern lights/ milky way. Canon has faster but with lessor quality at f2.
6) I can't carry them all
The fact that if sits in my bag always with ts17 is my vote for it.
Thank you for all the work and confirming through analysis my gut feel - zeiss 25/2 is the best at this mm range. Incidently the zeiss 25/2 is sharpest at f5.6.
I always carry my zeiss 25/2 (plus ts17, 24-105) because:
1) Its small
2) f2 with quality
3) It complements the 24-105 which is not great at 24 but very good beyond 30mm
4) I don't find I need the shift as much at 24mm (Ts24) as 17mm plus in a pinch I can use my 1.4x with TS17 which is the lens I base my system on. Or 7dii with TS17.
5) f2 and hardstop at infinity and 25mm make it the perfect lens for me nightscape lens with northern lights/ milky way. Canon has faster but with lessor quality at f2.
6) I can't carry them all
The fact that if sits in my bag always with ts17 is my vote for it.
Thank you for all the work and confirming through analysis my gut feel - zeiss 25/2 is the best at this mm range. Incidently the zeiss 25/2 is sharpest at f5.6.
Dec 22, 2015 at 03:56 PM
p.2 #16Comparison of Zeiss 25/2 ZE and Canon L lenses
Scott Stoness wrote:
4) I don't find I need the shift as much at 24mm (Ts24) as 17mm ...
Me too, although I often use shift on the TS-E 24/3.5L II to make two- and three-image panos.
Scott Stoness wrote:
5) ... and hardstop at infinity ...
Interestingly, the 25/2 ZE wasn't quite at the infinity stop on this test, while most manual focus Alt lenses are for this test site, up to focal lengths of about 35mm. I haven't yet had a chance to use it at "real" infinity, but I will...
Scott Stoness wrote:
6) I can't carry them all.
Tell me about it.
Scott Stoness wrote:
...Incidently the zeiss 25/2 is sharpest at f5.6.
I agree. I used f/8 because I was there...
4) I don't find I need the shift as much at 24mm (Ts24) as 17mm ...
Me too, although I often use shift on the TS-E 24/3.5L II to make two- and three-image panos.
Scott Stoness wrote:
5) ... and hardstop at infinity ...
Interestingly, the 25/2 ZE wasn't quite at the infinity stop on this test, while most manual focus Alt lenses are for this test site, up to focal lengths of about 35mm. I haven't yet had a chance to use it at "real" infinity, but I will...
Scott Stoness wrote:
6) I can't carry them all.
Tell me about it.
Scott Stoness wrote:
...Incidently the zeiss 25/2 is sharpest at f5.6.
I agree. I used f/8 because I was there...
Dec 22, 2015 at 04:11 PM
p.2 #17Comparison of Zeiss 25/2 ZE and Canon L lenses
Where in the scene did you focus?
In theory - depending on whether the lens field of focus is convex, concave, or flat at infinity - you would focus on a different spot. eg centre on concave and edge on convex - to maximize depth at infinity.
Thus really you should test focus on centre/check edge and focus on edge/check centre - and pick better - in determining which lens is better.
I will have to try his sometime with my ts24 and zeiss 25/2.
In theory - depending on whether the lens field of focus is convex, concave, or flat at infinity - you would focus on a different spot. eg centre on concave and edge on convex - to maximize depth at infinity.
Thus really you should test focus on centre/check edge and focus on edge/check centre - and pick better - in determining which lens is better.
I will have to try his sometime with my ts24 and zeiss 25/2.
Dec 22, 2015 at 04:23 PM
p.2 #18Comparison of Zeiss 25/2 ZE and Canon L lenses
Thanks, this is very helpful!
The Zeiss together with the 24-105 looks like a great combo.
Also, I wonder if this test was done on a 5Dsr, if the results would be about the same or if the difference between the primes and zooms would be even bigger.
The Zeiss together with the 24-105 looks like a great combo.
Also, I wonder if this test was done on a 5Dsr, if the results would be about the same or if the difference between the primes and zooms would be even bigger.
Dec 22, 2015 at 04:49 PM
p.2 #19Comparison of Zeiss 25/2 ZE and Canon L lenses
Scott Stoness wrote:
Where in the scene did you focus?
Dead center.
In particular, within the three regions shown in the red rectangles, below.
Center area focus regions
Where in the scene did you focus?
Dead center.
In particular, within the three regions shown in the red rectangles, below.
© jcolwell 2015
Center area focus regions
Dec 22, 2015 at 05:21 PM
p.2 #20Comparison of Zeiss 25/2 ZE and Canon L lenses
Scott Stoness wrote:
In theory - depending on whether the lens field of focus is convex, concave, or flat at infinity - you would focus on a different spot. eg centre on concave and edge on convex - to maximize depth at infinity.
(dead center)... same as a previous test including the TS-E 17/4L, where Fred commented about its curvaceous proclivities:
Lens tests, 17mm/18mm, https://www.fredmiranda.com/forum/topic/1267345
Fred's comments on focusing TS-E 17, https://www.fredmiranda.com/forum/topic/1267345/2#12075780
Scott Stoness wrote:
Thus really you should test focus on centre/check edge and focus on edge/check centre - and pick better - in determining which lens is better.
This raises at least two interesting issues, such as: (i) what is "best" (and how do you use it); and (ii) what is best vs. what is practical, in the given circumstances.
For this test, "practical" was the governing constraint. I was able to quickly take a series of about thirty scenic images with sufficient detail for a useful comparison of a new candidate lens (the ZE) with my "usual suspects", at or about 24mm. I also took a series of comparison photos for lenses at 400mm and 450mm (hint: if somebody opens their trench coat and says, "Pssst! Wanna great deal on a rare, mint Vivitar Series 1 450mm f/4.5 "solid cat", just move on). I also planned to perform a similar series of tests at 35mm, but I ran out of time (and weather).
OK, back to point (i). What is "best" depends on a variety of considerations, including post-processing. For example, if the centre is sharpest when focused there, but the edges are sharper when focused there (using apparent sharpness in magnified LiveView), then you might get the best photo from a two-image composite; one focused for the centre and one focused for the edges (plus, maybe a third focused for the corners). So, should you determine the best compromise (i.e. focus at some intermediary point), or should you spend the time in PP to do a focus stack to produce the final image? Further, depending on the image content and your intent for the image, it might be better to have less-sharp edges and/or corners, to help guide the viewer's gaze into the photo, rather than producing an image that's super-sharp, with fantastic micro contrast and 'boom in-your-face' 3D, from stem to stern.
Don't get me wrong - I plan to spend more time getting to know the Zeiss Distagon T* 25mm f/2 ZE. This test was simply to determine if it was worth the effort.
In theory - depending on whether the lens field of focus is convex, concave, or flat at infinity - you would focus on a different spot. eg centre on concave and edge on convex - to maximize depth at infinity.
(dead center)... same as a previous test including the TS-E 17/4L, where Fred commented about its curvaceous proclivities:
Lens tests, 17mm/18mm, https://www.fredmiranda.com/forum/topic/1267345
Fred's comments on focusing TS-E 17, https://www.fredmiranda.com/forum/topic/1267345/2#12075780
Scott Stoness wrote:
Thus really you should test focus on centre/check edge and focus on edge/check centre - and pick better - in determining which lens is better.
This raises at least two interesting issues, such as: (i) what is "best" (and how do you use it); and (ii) what is best vs. what is practical, in the given circumstances.
For this test, "practical" was the governing constraint. I was able to quickly take a series of about thirty scenic images with sufficient detail for a useful comparison of a new candidate lens (the ZE) with my "usual suspects", at or about 24mm. I also took a series of comparison photos for lenses at 400mm and 450mm (hint: if somebody opens their trench coat and says, "Pssst! Wanna great deal on a rare, mint Vivitar Series 1 450mm f/4.5 "solid cat", just move on). I also planned to perform a similar series of tests at 35mm, but I ran out of time (and weather).
OK, back to point (i). What is "best" depends on a variety of considerations, including post-processing. For example, if the centre is sharpest when focused there, but the edges are sharper when focused there (using apparent sharpness in magnified LiveView), then you might get the best photo from a two-image composite; one focused for the centre and one focused for the edges (plus, maybe a third focused for the corners). So, should you determine the best compromise (i.e. focus at some intermediary point), or should you spend the time in PP to do a focus stack to produce the final image? Further, depending on the image content and your intent for the image, it might be better to have less-sharp edges and/or corners, to help guide the viewer's gaze into the photo, rather than producing an image that's super-sharp, with fantastic micro contrast and 'boom in-your-face' 3D, from stem to stern.
Don't get me wrong - I plan to spend more time getting to know the Zeiss Distagon T* 25mm f/2 ZE. This test was simply to determine if it was worth the effort.
Dec 22, 2015 at 05:21 PM