Dear S2 Users:
For those of you that own and shoot the 35mm FL lens, would you comment on the amount of chromatic aberration seen in, and at the extremes of the images produced? I have been unable to locate any Imatest data on the lens.
Better yet, some corner examples at a few significant F-Stops would be real keen, if you have a moment! 🙂
Most grateful,
Pete Myers
Pete Myers;526 wrote: Dear S2 Users:
For those of you that own and shoot the 35mm FL lens, would you comment on the amount of chromatic aberration seen in, and at the extremes of the images produced? I have been unable to locate any Imatest data on the lens.
Better yet, some corner examples at a few significant F-Stops would be real keen, if you have a moment! 🙂
Most grateful,
Pete Myers
Hi Pete,
First, welcome to Red Dot Forum.
When the 35mm lens first came out, I did a brief write-up:
Taking the Leica Summarit-S 35mm f/2.5 ASPH and the S2 for a spin around the neighborhood
Here’s what I wrote regarding CA:
[COLOR=#333333][FONT=Trebuchet MS]
[/FONT][/COLOR][INDENT][INDENT][COLOR=#333333][FONT=Trebuchet MS] If I had any nit to pick regarding image quality it is that on very high contrast subjects (sunlit white building against deep blue sky, for instance), there is a trace of what appears to be magenta-red fringing. My first impression was that what I was seeing was lateral chromatic aberration, but there was no corresponding cyan-green edge on the opposite side. The effect appears most readily against blue skies (lots of UV, which can exacerbate fringing) and is most likely a result of either a small amount of sensor blooming or imprecision in LR’s demosaicing algorithm. Either way, the fringe is quickly fixed with a -15 red/cyan correction in Lightroom’s Lens Correction palette, along with Leica’s recommended setting of Defringe to “All Edges.” I’d expect the forthcoming profiles from Adobe and Leica to include this correction automatically as a lens profile. [/FONT][/COLOR][/INDENT][/INDENT]There are numerous 100% crops including corner crops.
I’m sure other S2 users will chime in with their experiences as well. For me, the 35 has been an outstanding overall performer.
David Farkas
Red Dot Forum
Leica Store MiamiHi David:
I certainly read your review, and enjoyed it very much—beautiful images!
But after reading your review, I took a look at some full size images from this lens over here:
http://www.lenstip.com/1835-news-Leica_Summarit-S_35_mm_f_2.5_ASPH._-_sample_images.htmlI found it astonishing how much CA was present in the images presented. I am not sure how to account for it, so I thought I would post, not just lurk.
Grateful for your response, better yet your wonderful article and images.
Pete
Pete Myers;529 wrote: … I took a look at some full size images from this lens over here:
http://www.lenstip.com/1835-news-Leica_Summarit-S_35_mm_f_2.5_ASPH._-_sample_images.htmlI found it astonishing how much chromatic aberration was present in the images presented. I am not sure how to account for it …
Well—I don’t have an S2, much less any Summarit-S lenses … but I do own several good lenses including some Leica M lenses, and I’d consider the amount of chromatic aberration seen in the test images from the link given above as normal, even for an expensive high-end lens. Even the best lenses still aren’t perfect, and when pixel-peeping then you will always find a minor fault or two.
I downloaded one of the test images from lenstip.com and tried to remove the colour fringes in Camera Raw. I arrived at red/cyan -9 and blue/yellow +4. That’s pretty low figures, almost worthy of an apo lens. One of my shots taken with the Summilux-M 35 mm Asph at medium distance can use R/C -3, B/Y +9; a Summilux-M 50 mm Asph (which allegedly has apo correction) image gets near-perfect with R/C -8, B/Y +2 (YMMV). So with regard to the magnitude of the chromatic aberration, they all are in the same ballpark. I’d say for a non-apo wide-angle lens, that’s very good performance and nothing to worry about.
Furthermore, for Adobe Camera Raw and Photoshop Lightroom there’s a very elaborated lens profile for the Leica Summarit-S 35 mm Asph which includes corrections for all combinations of apertures and distances. However you need to shoot in raw format in order to take advantage of it. I wish such profiles were available for the Leica M lenses, too …
It would be nice to see whether those Adobe profiles are working as planned. Perhaps an end-user of this lens can provide a demonstration of the lens correction in ACR on and off, and how it well it works?
It should be interesting to see if it works well. My experience with ACR lens correction has been mixed at best, but that has been with manual focus Zeiss ZF lenses. Perhaps with depth information, the correction profiles work better.
Pete
Pete,
Ask and you shall receive. I went out today with the S2 and 35mm Summarit-S around noon when the South Florida light is at its most harsh and direct. I headed for a polished stainless steel sculpture in nearby Greenbelt park as I figured it to be the toughest test of CA within driving distance.
The files here were processed in LR 3.4.1 with my standard S2 preset applied (which includes auto lens profile correction). I then unchecked the lens profile so that the shots are identically processed otherwise. I made no additional CA correction or moved the sliders in the lens profile tab from their default positions (100 for distortion, 100 for CA). For each set, the image with no lens profile is first and the one with lens profile applied is second.
The full shots when clicked will bring up 1800×1200 images sized down from the original. The crops when clicked are non resized crops and are pretty large. The thumbnails generated by the forum software don’t show much difference, so please make sure to click on each example.
Here’s the results of the tests:
First, the full shot with no lens profile applied
S2 with 35mm Summarit-S, 1/750th @ f/8, ISO 160
Full shot with the lens profile applied
100% crop from upper-left corner with no lens profile
100% crop with lens profile
Another full shot with no lens profile applied
S2 with 35mm Summarit-S, 1/750th @ f/5.6 ISO 160
Full shot with lens profile applied
100% crop from upper-left corner with no lens profile
100% crop with lens profile
All in all, a very strong showing from both the 35mm Summarit-S and the LR lens profile. What little CA is present in these extreme examples (at the frame edge I might add) is almost completely eliminated with one click.
Hopefully, this was the kind of test you were looking for.
David Farkas
Red Dot Forum
Leica Store MiamiThank you for taking the time to do this test. I am very impressed with the lens performance without the profile, but your test goes to show how much improved the LightRoom processing has become since the S2 came out.
David,
Thank you for doing the test and damn you – now I lust for the 35 😡 Very impressive performance even without the corrections. Overall, the entire stable of S lenses have far exceeded my expectations.-Al
Atanabe;548 wrote: David,
Thank you for doing the test and damn you – now I lust for the 35 😡 Very impressive performance even without the corrections. Overall, the entire stable of S lenses have far exceeded my expectations.-Al
You’re welcome on both counts, Al. 😀
The S lenses are really extraordinary. Peter Karbe told me, “The S lenses are the best that we [Leica] have made… for any system… ever.” Strong words from an understated genius…. and true.
David Farkas
Red Dot Forum
Leica Store MiamiDavid,
Peter Karbe is truely a genius! His “affordable” Summarit line for the M series sparkle. It is interesting to note that there is a no compromise approach to design and execution of the S lenses. While Hasselblad is content to offer software corrections for their D series lenses, Leica refuses to do so and instead produce the best glass bar none.Hi David:
I am grateful to you for the wonderful tests! Simply stunning in all regards. That is the kind of performance one is hoping to see in the S2 with the 35mm. Now, there will be no way for me to sleep! 🙂
Thank you for you kindness in the tests!
Pete Myers
You must be logged in to reply to this topic.