Chromatic aberration in 35mm Summicron M APO

[Derbyshire Man]

Advertisement (gone after registration)

I bought this lens in perfect used condition last week and took it out for some landscape testing. It has great resolution but in high contrast areas it on screen at lest has the attached level of chromatic aberration. Am I expecting too much? These are screen grabs from LR. I have to say it wasn't what I was expecting and makes me wonder if it's the reason the lens looks brand new! Ie sent to Leica and returned 'within specification'. Clearly two of these are contrejour but it looks no better than my Summilux 28mm wide open!

Anyone else got one who can comment on this?

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!

[Derbyshire Man]

And this final one is at f16, the others were f4.

The other odd thing is that the embedded preview has none, it's only on switching to the develop module.

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!

[UliWer]

I think the examples don‘t show chromatical aberrations but purple fringing. Both are often mixed up even though they have totally different causes: the first results from optical faults of the lens, the second from the sensor. You may find an explanation here:

https://www.imatest.com/docs/sfr_chromatic/

Edited 4 hours ago by UliWer

[jaapv]

What camera? It can occur on any sensor but high resolution sensors are more prone.

[jaapv]

10 minutes ago, UliWer said:

I think the examples don‘t show chromatical aberrations but purple fringing. Both are often mixed up even though they have totally different causes: the first results from optical faults of the lens, the second from the sensor. You may find an explanation here:

https://www.imatest.com/docs/sfr_chromatic/

I'm sure you are right. Unfortunately these images are too small to show the final proof - lack of a yellow counterfringe. 

[Derbyshire Man]

Thanks both. Attacking it with bit of purple de-fringing sorts it. 

Feel free to delete the topic in case it becomes some sort of slur on 35mm apo's!

Edited 4 hours ago by Derbyshire Man

[padam]

Advertisement (gone after registration)

33 minutes ago, UliWer said:

I think the examples don‘t show chromatical aberrations but purple fringing. Both are often mixed up even though they have totally different causes: the first results from optical faults of the lens, the second from the sensor. You may find an explanation here:

https://www.imatest.com/docs/sfr_chromatic/

https://photographylife.com/what-is-purple-fringing

This is a much better article on the subject - that one falls into the common (not fully satisfactory) explanation.
Yes, it Is mainly caused by lens design:

"Lenses designed to correct for two wavelengths are called achromatic, while photography lenses designed to be highly corrected for red, green, and blue wavelengths are referred to as apochromatic. These lenses tend to suppress chromatic aberrations, including purple fringing.

However, even with a top lens, perfection isn’t guaranteed. After all, the “apochromatic” or “APO” designation isn’t controlled by some official body, and there are still low-level chromatic aberrations on almost all lenses. (Unfortunately, as discussed earlier, low-level chromatic aberrations that are normally invisible can be the primary cause of purple fringing.)"

So yes, even though one can (mistakenly) think that this naming implies that it is as close to perfection as it can be (some even describe it as such, simply ignoring its faults), there is only so much one can do with certain size and other limitations - although still rather disappointing, considering the price. (We can go the other way and say that some consider the Summilux as an unlabelled "APO" lens - even though there are more modern designs that can outperform it in that regard, despite not being associated with this - arguably vaguely defined - designation.)

They key word I think is 'suppress', which is not (nearly) the same as 'eliminate'.

Edited 4 hours ago by padam

[Derbyshire Man]

21 minutes ago, jaapv said:

What camera? It can occur on any sensor but high resolution sensors are more prone.

Just to reply with the full facts its a crop of an M11 image with the summicron m 35 apo

[jaapv]

28 minutes ago, padam said:

https://photographylife.com/what-is-purple-fringing

This is a much better article on the subject - that one falls into the common (not fully satisfactory) explanation.
Yes, it Is mainly caused by lens design:

"Lenses designed to correct for two wavelengths are called achromatic, while photography lenses designed to be highly corrected for red, green, and blue wavelengths are referred to as apochromatic. These lenses tend to suppress chromatic aberrations, including purple fringing.

However, even with a top lens, perfection isn’t guaranteed. After all, the “apochromatic” or “APO” designation isn’t controlled by some official body, and there are still low-level chromatic aberrations on almost all lenses. (Unfortunately, as discussed earlier, low-level chromatic aberrations that are normally invisible can be the primary cause of purple fringing.)"

So yes, even though one can (mistakenly) think that this naming implies that it is as close to perfection as it can be (some even describe it as such, simply ignoring its faults), there is only so much one can do with certain size and other limitations - although still rather disappointing, considering the price. (We can go the other way and say that some consider the Summilux as an unlabelled "APO" lens - even though there are more modern designs that can outperform it in that regard, despite not being associated with this - arguably vaguely defined - designation.)

I think that the recognized authority, Imatest, is a more reliable source. The article that you refer to does not recognise that Leica uses a much more rigid "APO" designation than other brands. Photograpy Life rightly mentions that an APO lens is only fully corrected in three points, R,G and B and that anything outside those points is anybody's guess, but Leica goes much further and corrects the full spectrum, pays attention to the OOF areas as APO is only defined in the plane of focus, and even to a certain extent IR; and it  does not even  mention the influence of high resolution sensors, reason that we are seeing more complaints of this kind in the forum recently.
Nor do they mention that chromatic aberration always is a double fringe on two sides of the contrast line: Red and green for lateral aberration, purple and yellow for longitudinal (yellow is sometimes hard to see). If you only have a purple fringe it is blooming or possibly (rarely) a demosaicing error in the software.

[padam]

1 minute ago, jaapv said:

I think that the recognized authority, Imatest, is a better source.

I don't know what this is based on, clearly both were written by individuals with their own impressions and biases.
Saying that it is caused by sensors and not lenses seems quite silly to me.
In that case two 35mm lenses on the same sensor (let's suppose they are specifically designed for that mount, so we can eliminate the factor of adapting) wouldn't show different amount of purple fringing - clearly the actual performance of the lenses are the main differentiator here.

The real point here is what APO stands for, or how it is exactly defined, how it is justifiable to label a product like that (and charge exorbitant prices because of it), which can be disputed all day long.

[jaapv]

Imatest is a major supplier of lens testing infrastructure, used by most photographic magazines and lens makers.  One may assume that they know exactly what they are talking about.

https://en.wikipedia.org/wiki/Imatest

Of course the lenses would show a different amount. The fringing is dependent on the micro contrast a lens can produce and that differs by lens. The better the lens, the stronger it will be.  These are not beliefs or opinions, but simple facts. If this were CA, where is the yellow counter fringe in these images? 

[pgk]

7 minutes ago, padam said:

Saying that it is caused by sensors and not lenses seems quite silly to me.

Doesn't it actually depend on the cause?

[Derbyshire Man]

9 minutes ago, jaapv said:

Imatest is a major supplier of lens testing infrastructure, used by most photographic magazines and lens makers.  One may assume that they know exactly what they are talking about.

https://en.wikipedia.org/wiki/Imatest

Of course the lenses would show a different amount. The fringing is dependent on the micro contrast a lens can produce and that differs by lens. The better the lens, the stronger it will be.  These are not beliefs or opinions, but simple facts. If this were CA, where is the yellow counter fringe in these images? 

Yes and it's bilateral/circumferential reddish purple fringing, no partner colour.

[padam]

2 minutes ago, pgk said:

Doesn't it actually depend on the cause?

I'm sorry, the right definition is that it is caused by digital sensors, but it can be (somewhat) counteracted by lens design.
I can't understand the Imatest article saying it has little to do the lens, when tests clearly show differences, and also not at all in line with "the better the lens, the stronger it will be".

[jaapv]

Blooming is connected to micro contrast - which is logical. Design a lens with lower microcontrast or less defined OOF rendering and the blooming will be a bit less as the light on the sensor will be more smeared out. The lens with the better micro contrast (AKA the better lens) will trigger the sensor more. In that sense it has to do with the lens, but I think Imatest left that out not to overcomplicate things. The blooming itself has indeed nothing to do with the lens, but is a sensor phenomenon. (and "little"is a concession to the microcontrast bit 😉)

[jaapv]

18 minutes ago, Derbyshire Man said:

Yes and it's bilateral/circumferential reddish purple fringing, no partner colour.

Well, final proof of Uliwer's post.