Leica 35mm lenses and focus shift at infinity

[indergaard]

Advertisement (gone after registration)

I currently own the 35mm Summicron 35 asph (v1) and a Summilux 35mm asph fle.

Both was recently serviced in Wetzlar, along with the camera as well (M10).

 

All my other lenses was serviced at the same time.

 

I am experiencing focus shift at infinity with both of my 35mm lenses. This happened before servicing, and also now after servicing, with both lenses, on the same camera body. I don't experience any focus shift at infinity with my other lenses (Lux 50 ASPH, Elmar-M 24/3.8 ASPH, Macro-Elmar-M 90/4).

 

Summilux 35 ASPH FLE @ infinity position on the lens:

f/1.4-f/2: no focus shift

f/2.8-f/8: focus shift

 

Summicron 35 ASPH @ infinity position on the lens:

f/2-f/2.8: no focus shift

f/4-f/8: focus shift

 

Both lenses align perfectly with the rangefinder at infinity, but after doing some testing by focus stacking, and also using magnification and focus peaking with the EVF, I can easily detect that optimum sharpness at infinity is achieved well before the infinity stop on the lens. The exception is wide open (f/2 on summicron, f/1.4 on summilux), which focuses correctly at the infinity stop.

 

I achieve maximum sharpness on a distant landscape scene with the affected apertures (f/2.8-f/8) at the following settings on the distance scale of the lens (see attachment of the lenses):

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Is this normal behavior on the 35mm Leica's? Or should I send the lenses back again?

Both lenses focus correctly at other distances from what I can see. It is only at infinity that I am experiencing this issue. And if viewing an image at 100% the difference is easily seen.

 

 

[TomB_tx]

Out of curiosity I've checked many fast lenses of various makes, and all I've tried had some focus shift (plane of focus moving to the rear) when stopping down. The amount varies, and on most I wouldn't have noticed if I wasn't looking for it, as pictures were "acceptably sharp" as DOF largely covers the shift. Many RF lens makers (Leica and Zeiss included) used to set lenses with ideal focus at 1 to 1-1/2 stops down, so that there would be slight shift to the front wide open and a lesser shift to the rear at normal shooting apertures. And because film has some thickness to the image, the effect wasn't as noticeable in film days.

Today it has become more fashionable to shoot wide open, and to examine details at 100%, so people want lenses sharpest wide open. As a result people find the shift more when they test to look for it.

I've used a Nokton 35 f1.4 for several years, and got very nice pictures with it before reading about its shift and testing for it. Suddenly I wanted something better.

I'll likely be happy with it once again when senility sets in and I forget about the issue.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[indergaard]

I found TIm Ashley's review and in-depth examination of the 35 FLE, and basically found that I am experiencing the same issue as he is. I have only been looking at the centre of the image, where the 35 FLE is softer than the 35 Cron at f/8 focused at the infinity hard stop on the lens. I did not check the sides and edges of the images however.

Tim Ashley explains it a lot better than I can:

 

 

What I found today was that this is pretty much exactly true. But there were other useful lessons: (all regarding the F4 only shots so far pending review of all 60 frames)

1. Focussing wide open with the RF or EVF gives much better results across the frame. Pretty much all of the frame is between very acceptable and excellent, especially viewed at print resolution or 50% view
2. Refocussing at F4 with the EVF gives, if anything, sharper results on centre but a repeat of the 'wandering in and out of focus across the rest of the frame' seen in my original shot
3. Reaching stopped-down EVF focus from 'far to first shimmer' gives equally good results to the RF focussed shot across the frame
4. Reaching stopped-down EVF focus from 'near to first shimmer' gives equally good results to the RF focussed shot on centre but very poor results across large parts of the frame

From this I conclude provisionally that my fist instincts were right: there is an uneven field of focus, probably wave shaped, and that when stopped down to F4 it is of a shape such that the EVF provides too much ambiguity to focus stopped down, because the POF needs to be focussed towards the back of the DOF in order to get as much of the planar target as possible within the field of focus. 

This will be somewhat counter-intuitive to some users who, assuming that because the lens has a slight tendency to shift focus further away as one stops down, will feel that the safest way to focus is 'stopped down EVF'. In fact, because the actual point of focus remains (laudably) within the DOF as you stop down, despite this focus shift, one needs to be more concerned about focussing according to the 'shape' of the field of focus and this requires one to bias focus towards the rear of the field. 

When you focus with the very accurate RF, or with the EVF wide open, this process then happens naturally: you stop down without changing focus and the lens's inherent focus shift moves so as to bias the POF rearwards within the field, thus compensating really nicely for the shape of the field.

But when you focus with the EVF stopped down, it is a crapshoot biased toward failure: coming at focus from FAR to 'first shimmer' gives a very good result but from NEAR to first shimmer, gives a good centre and a very poor average across the frame. 

So: using the RF, the EVF wide open or the EVF stopped down but with a far bias are the best ways to get good results across the frame.

 

One final note: the frame that was very sharpest on centre (by a tiny margin) did not have the best average performance across the frame. There is, as far as I can see, no possible focus setting that would achieve that, which is exactly what I would expect and is just not going to matter in practice!

 

 

Link: https://tashley1.zenfolio.com/blog/2013/4/leica-m-240-with-35mm-f1-4-fle---some-observations

 

The focus shift, or rather, uneven focus at infinity, is more noticeable on the 35 FLE than on the 35 Cron ASPH, which is something I did not expect however. The Cron has some mild focus shift between f/2.8-f/4 at mid distances though, which the 35 FLE does not.

 

My conclusion is that Leica's 35mm lenses are tricky. And that the Cron 35 ASPH is a better lens for planar and distant subjects. Noticeably so, even at f/8.

Edited August 18, 2018 by indergaard

[indergaard]

Out of curiosity I've checked many fast lenses of various makes, and all I've tried had some focus shift (plane of focus moving to the rear) when stopping down. The amount varies, and on most I wouldn't have noticed if I wasn't looking for it, as pictures were "acceptably sharp" as DOF largely covers the shift. Many RF lens makers (Leica and Zeiss included) used to set lenses with ideal focus at 1 to 1-1/2 stops down, so that there would be slight shift to the front wide open and a lesser shift to the rear at normal shooting apertures. And because film has some thickness to the image, the effect wasn't as noticeable in film days.

Today it has become more fashionable to shoot wide open, and to examine details at 100%, so people want lenses sharpest wide open. As a result people find the shift more when they test to look for it.

I've used a Nokton 35 f1.4 for several years, and got very nice pictures with it before reading about its shift and testing for it. Suddenly I wanted something better.

I'll likely be happy with it once again when senility sets in and I forget about the issue.

 

 

The reason I noticed it was because I was comparing the two 35's to each other for landscape use. To my surprise, the Cron was consistently sharper for this use stopped down. It is very easy to see. The only exception was wide open, where the Lux was sharper. It seems like the Lux is a compromise at the infinity setting, where the center will be noticeably less sharp to achieve an acceptable sharpness across the frame. I would maybe not have thought much about it, unless I had compared it with the much older and more affordable Cron, which was noticeably sharper in the center, and just as sharp across the frame, with the exception of the extreme edges, where the Lux is sharper...

 

The Cron 35 ASPH also suffers some minor focus shift or wavy plane of focus at infinity though, but it is less noticeable than on the Lux FLE. But, focusing the Cron slightly before the infinity hard stop gives me a slightly sharper result than at the infinity hard stop of the lens.

 

Is this normal on the 35 Cron?

Edited August 18, 2018 by indergaard

[IkarusJohn]

Yes, as I recall - I had the previous version of 35 Cron ASPH, and that is my recollection.

 

I assume the new 35 Summicron ASPH is an improvement - not much information about it here, sadly.

[Jeff S]

My 35 Summicron ASPH (1st version) shifts focus mildly at f4-f5.6, even before infinity focus. Sean Reid (I haven’t subscribed in years) reported exactly the same with his sample. No problem, though, as it’s easily accommodated and presents no problem in my prints.

 

I don’t know about the 2nd iteration, but haven’t seen any information specifically addressing the issue. The 35 Summarit (both versions), however, exhibit no focus shift, and further, minimize flare much better than the Summicron ASPH, at least the first version. If I were buying today, I’d give the f2.4 strong consideration. I’m likely to pick up the equally superb 75 Summarit sometime soon.

 

Jeff

[darylgo]

Advertisement (gone after registration)

I sent two summilux' back and eventually purchased the Zeiss Biogon. Landscape was a challenge with the summilux that I could not resolve. Since, Zeiss has released their F1.4 that is reported superb.

[indergaard]

Yes, as I recall - I had the previous version of 35 Cron ASPH, and that is my recollection.

 

I assume the new 35 Summicron ASPH is an improvement - not much information about it here, sadly.

 

Ok thanks. That's nice to know, but also not expected. I have owned my Summicron 35 ASPH for a while, and it has always performed great, but I have never tested it with LV/EVF and focus stacking. The difference I am experiencing for landscape use is minimal, and I wouldn't notice it unless I had done these tests.

 

The center resolution of the Summilux 35 ASPH FLE however was very noticeable for landscape use, even at f/8, and I'm not sure if I will keep it. If the 35 ASPH II is improved, then I might consider selling the Lux FLE and Cron ASPH v1 and purchase a new Cron ASPH II. But I can't find a lot of information on the Cron 35 ASPH II. I found this test/comparison however, and it seems like the out-of-focus rendering of the newer version is slightly smoother, and the newer version also seems to retain contrast much better in strong backlight: http://www.streetsilhouettes.com/home/2017/3/30/leica-35mm-lenses-5-summicron-versions

 

 

My 35 Summicron ASPH (1st version) shifts focus mildly at f4-f5.6, even before infinity focus. Sean Reid (I haven’t subscribed in years) reported exactly the same with his sample. No problem, though, as it’s easily accommodated and presents no problem in my prints.

 

I don’t know about the 2nd iteration, but haven’t seen any information specifically addressing the issue. The 35 Summarit (both versions), however, exhibit no focus shift, and further, minimize flare much better than the Summicron ASPH, at least the first version. If I were buying today, I’d give the f2.4 strong consideration. I’m likely to pick up the equally superb 75 Summarit sometime soon.

 

Jeff

 

Mine does to, and the shift is very minimal. But what I am experiencing is also a very mild shift or wavy plane of focus at infinity, even at f/8. The difference is very minimal, but if you enable focus peaking in LV/EVF, and target a distant landscape scene, set the lens at the infinity hard stop and slightly nudge it back (as seen in the pictures I posted), you can see that the red speckles caused by the focus peaking slightly increases as you go slightly back from the infinity hard stop. Make exposures at each setting (tripod, timer, no LV) and the result from the exposure that is focussed slightly before the infinity hard stop will be slightly sharper. 

 

I sent two summilux' back and eventually purchased the Zeiss Biogon. Landscape was a challenge with the summilux that I could not resolve. Since, Zeiss has released their F1.4 that is reported superb.

 

 

My wife has a Biogon 35/2 ZM, and it shows none of these problems as you say. Zeiss's 35/1.4 ZM is too big for my liking. Even the Summilux 35 ASPH FLE is too big and heavy for my liking, but I could live with the size considering it's maximum aperture, although my preferred size is the Summicron 35 ASPH, which balances perfectly on the M10 and MP (analog), and is just a joy to use.

Edited August 19, 2018 by indergaard

[indergaard]

I quickly re-did the test of the 35 Lux FLE, and what Tim Ashley found is very easy to find.

 

At the infinity hard stop, the center resolution isn't very good, even at f/8. The outer zone resolution is good. The difference is very noticeable. When focussing slightly before the infinity hard stop (as shown in my attached picture of the lens), the center resolution is extremely good, but the resolution in the outer zones of the image completely fall apart. Again, very noticeable.

 

I am experiencing the same with the Cron 35 ASPH v1, but to a much smaller degree, and it is barely noticeable by comparison.

 

The 35 Lux FLE is not a good landscape lens. I guess the solution is to focus stack and blend in photoshop to get an exposure that is really sharp in the center and in the outer zones. But that presents a lot of other issues (motion blur, changing light, changing scenery/landscape, moving trees, etc).

Edited August 19, 2018 by indergaard

[Gregm61]

I think I’ll play the “ignorance is bliss” game with this lens. At $5K per copy we all have to use some rationale or can choose to test the thing to end of time. I’m going to go with Ming Thein’s review and continue to enjoy the images I’m seeing from mine out of the MP240 and M262.

Edited August 19, 2018 by Gregm61

[Jeff S]

I would be careful to judge based on a couple of lens samples (and bodies). Even though my 35 Summicron ASPH exhibited the same focus shift behavior as Sean Reid’s, some here have reported no focus shift at all. Likewise, Tim Ashley’s Summilux findings, along with one similar experience, shouldn’t lead one to conclude that all such lenses are “not good for landscapes”. And even if this is a common characteristic, I’m sure a good photographer could produce some fantastic landscape work using that lens in an overall print workflow. People have done that with far less.

 

Jeff

[semi-ambivalent]

I’m going to go with Ming Thein’s review and continue to enjoy the images I’m seeing from mine out of the MP240 and M262.

Remembering, of course, if you like the images you're seeing from the 35 Summilux FLE, Thein's opinion is empty air, with no need of mention. If you have paying clients it is, almost always, a different game. Or not.

 

Best,

s-a

[Gregm61]

i would venture a guess, most posting here don’t have any paying clients and only our own sanity/vanity to be concerned about.

 

Isn’t opinion-filled empty air a pretty good description of the internet as a whole?

Edited August 20, 2018 by Gregm61