M11 + Apo 90 Chroma and fringing

[S Maclean]

Advertisement (gone after registration)

I love the camera, I love the lens, but this is really bad.  Purple and green fringing, together and all by themselves, on a caputre that should be fairly straight forward.  I am amazed.

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!

[adan]

Which 90mm? At which aperture?

I'm sorry you're seeing this, but it does seem to be a feature of the M11's high resolution (see other threads on the subject).

The high resolution is effectively a higher-power magnifying glass, which magnifies and reveals more subject detail, but also can magnify and reveal various imaging flaws, including lens chromatic aberration, that were not visible on film, or on previous sensors.

It may be the case that the M11 sensor has an additional tendency to add CA (beyond magnifying that produced by the lens), but I don't know that that has been confirmed yet.

Edited June 21, 2022 by adan
added link

[Advertisement]

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

[S Maclean]

Sorry, that's an autocorrect on the title.  It's the APO 90 at 2

 

Moderators if you will please amend title. I don't think I can edit on my end.

Edited June 21, 2022 by S Maclean
Asking for title fix

[Studienkamera]

There was a lengthy discussion about similar findings using this lens:

 

[adan]

57 minutes ago, S Maclean said:

It's the APO 90 at 2

Thanks for that.

It is the case that the 90 APO-Summicron-M-ASPH (ASMA), at larger apertures, can produce purple fringes on/around very high contrast edges (like skin so overexposed that it turns from dark to near-white, or dark tree branches against blown-out sky).

It's also the case that the 90 ASMA is a very high-contrast lens that increases the odds that high-contrast scenes will have blown highlights, which also make the purple fringe more extensive and obvious.

It's also the case that it produces bi-colored purple-green out-of-focus edges (as in the background sculpture). See the link studienkamera just posted (Thanks! I was about to post that myself).

The 90 ASMA earned the APO designation 24 years ago, on film, and so far as I know has not been updated since then. So part of the problem is with its normal characteristics themselves, as revealed in all their "glory" on silicon. I shot with one recently (not on the M11) and it will throw out the occasional purple/magenta rim to "hot" highlights at f/2.0.

It's also the case that most digital sensors are more sensitive to such fringing. It has been a common complaint with other high-spec M lenses on other M digitals in scenes with strong contrast (specular sunlight reflections in particular) that get overexposed: 35 f/1.4 ASPH shot at f/1.4 and 50mm Noctilux ASPH f/0.95 being other examples.

However, as I said, the 60 Mpixels of the M11 do reveal it more (make it bigger, spread across more pixels) than lower-Mpixel sensors, or the films of the era in which it was designed.

And it is possible Leica is working on a firmware fix. And perhaps (in secret) an updated 90 f/2.0 (as my Leica rep used to say with a wink, "You just never know!")

In the meantime, as you may know, the face fringe (at least) can be repaired in most post-processing software - purple fringing is not unique to Leica lenses and cameras.

Edited June 21, 2022 by adan

[opera207]

1 hour ago, S Maclean said:

Sorry, that's an autocorrect on the title.  It's the APO 90 at 2

 

Moderators if you will please amend title. I don't think I can edit on my end.

That CA is really an "apologize"😂

[lct]

Advertisement (gone after registration)

Not sure what lenses you are comparing the Summicron 90/2 apo to folks? Or are you comparing it to 90/2.8 or 90/4 lenses? If you compare apples to apples i.e. 90/2 lenses, the 90/2 apo is better CA wise than the 90/2 v3. Is is beaten by the 90/2 v2 (left) but it is a more bulky lens.  For significantly less CA, go for the 90/4 macro (right), the Elmar-C 90/4 or the old Elmar 90/4 collapsible. The Elmarit 90/2.8 v1 does well too.

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!

 

 

[tom0511]

I also feel to see more CA with some other lenses on the M11 compared to the m10r. For example with the 28/1.4.

[S Maclean]

The resolution size may be a factor.  If adjusted to a 18 or 40 mp it may not be that far off.

im test on g a VL 15 mm 4.5 and it seemS to behave well. 
 

Also the example I posted here, in hindsight, may be a little overly extreme as the areas affected are not in focus ( on purpose) and are hit by very strong direct late day sun.

the purple is easy to manage but the green requires more work because the de fringing tool, if used in general setting, then affects the edges against grass etc., even though they don’t have visual aberration, but share the same exact tonality.  This is specially visible if then converted to monochrome.

Edited June 23, 2022 by S Maclean
Typo

[tom0511]

Its visible at same resolution. There is some difference, but it corrects easily in LR.

[jaapv]

This is a recurring question. There are two main causes for purple fringing and on a digital camera they are interrelated. First there is the lens: Axial chromatic aberration can cause purple fringing in high contrast situations. An Apo lens is corrected for chromatic aberration but only fully in three spots of the spectrum: R,G and B (and in the plane of focus only). So not necessarily completely for purple or magenta. Then there is the sensor: blooming causes purple (or blue) fringing  on high contrast edges. There is anti-blooming on the sensor but it cannot be 100% effective, certainly not on modern hi-res sensors. So there you see what is happening: A lens corrected to a high standard is not completely protected against purple fringing and still produces high-contrast edges that the sensor cannot handle completely. Result: the better the gear, the more this can happen.