APO quiz

[CheshireCat]

Advertisement (gone after registration)

Which is which ?

 

Lens A: APO Summicron-M 90

Lens B: APO Sonnar 135

Lens C: EF 85/1.2

 

All lenses shot at f/2 on the M240, subject at 1m, crops from mid right part of the image (sorry the focus point is not exactly the same).

 

Sort left to right.

Your APO thoughts are greatly appreciated.

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!

[Advertisement]

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

[Michael Geschlecht]

Hello CheshireCat,

 

It would appear that you can tell which lens is which by looking at the relative sizes of the round devices which are least effected by various lens anomalys.

 

The image with the smallest round devices are most likely taken with the shortest focal length lens. The image with the largest round devices is most likely taken with the longest lens & the middle sized devices are most likely taken with the middle focal length lens.

 

Since they were all photographed at a distance of 1 meter.

 

It might be more instructive if you reshot the picture with the same number of round devices in each frame. That is: All focused in order to produce the same image magnification.

 

You might also consider coloring the central-most round device a different color & focussing on that. Even if the comparison portions of the images compared here are not taken from the center of the entire image photographed.

 

You might also consider putting a second colored dot at the center of the portion that you will put here.

 

This would remove a number of the incedental variables which make your very interesting comparison more difficult to evaluate.

 

Best Regards,

 

Michael

[CheshireCat]

Micheal, thanks for your comments and suggestions, they make sense.

I could have done much better, but for some reasons I cannot use my tripod today.

 

The distance is 1 meter as per the lens barrel, so the error may be so high that it would be hard to tell which lens is 90 and which is 85.

 

In any case, the main purpose of this "quiz" is discussing APO properties of each lens.

 

It is not hard to identify the AS 135 (higher magnification), but can you tell which one is the APO Summicron ? It is the APO one, right ?

[gpwhite]

Micheal, thanks for your comments and suggestions, they make sense.

I could have done much better, but for some reasons I cannot use my tripod today.

 

The distance is 1 meter as per the lens barrel, so the error may be so high that it would be hard to tell which lens is 90 and which is 85.

 

In any case, the main purpose of this "quiz" is discussing APO properties of each lens.

 

It is not hard to identify the AS 135 (higher magnification), but can you tell which one is the APO Summicron ? It is the APO one, right ?

 

I am sorry to say that I believe there is something wrong with the (your) APO 50 Summicron. Presuming that the image in the middle is the APO 50 (because it has the widest FOV and richest blacks - highest contrast-), the central CA is not at all normal. I have shot both the APO 50 and the APO 135 Zeiss Sonnar ZF.2 on my M240, and my APO 50 was clearly better corrected.

 

What do you think is going on?

[CheshireCat]

I am sorry to say that I believe there is something wrong with the (your) APO 50 Summicron. Presuming that the image in the middle is the APO 50 (because it has the widest FOV and richest blacks - highest contrast-), the central CA is not at all normal. I have shot both the APO 50 and the APO 135 Zeiss Sonnar ZF.2 on my M240, and my APO 50 was clearly better corrected.

 

What do you think is going on?

 

It is important to note that it is a 90, not a 50.

Honestly, I really don't know what's going on. I think neither of the two shots on the left (one of which is the APO Summicron-M 90 ASPH) deserves the definition of APO.

[gpwhite]

Sorry, my mistake after a bit of IPA, in not catching which Summicron you used.

 

APO is an unregulated, "qualitative" designation. The Summilux-M 50 ASPH is as APO as is the APO Summicron-M 90mm ASPH, yet Leica chose not to market it that way.

 

My AA 90 does show the sort of aberration you found in your test. I find very little indeed with the APO 135mm Zeiss, which must be four times the size of the Leica 90mm (I cannot believe people shoot the Otus handheld!). The AA 50, however, is just on the money!

[adan]

Advertisement (gone after registration)

It's important to remember that "APO" refers - simply and solely - to a lens's ability to focus three wavelengths of light in the same plane. Usually r,g,b - but sometimes infrared,g,b.

 

That may not be a "regulated" definition - but it is the scientific definition for APO (short for "APOchromatic". If a lens can do that, it is apochromatic. If it can't, it isn't.

 

Apochromat - Wikipedia, the free encyclopedia

 

The key word is "focus" - what happens in the unfocused parts of the picture (color fringes, poor bokeh, or whatever) is completely irrelevant to the APO designation.

 

Note in the wiki illustration of an APO lens, that the colored light rays are NOT aligned everywhere - only at the point of focus. So one can easily get color fringing in the blur circles even with a fully compliant APO lens.

 

My 75 Summilux is a perfect example of a non-APO lens - at wide apertures it focuses green and blue light together, but the red image is clearly focused "elsewhere". Easy to see viewing the color channels separately in Photoshop. Result is a cyan stain in white areas, and a red stain in black areas.

 

http://www.l-camera-forum.com/leica-forum/digital-forum/145063-what-chromatic-aberrations-really-look-like.html

[CheshireCat]

Note in the wiki illustration of an APO lens, that the colored light rays are NOT aligned everywhere - only at the point of focus. So one can easily get color fringing in the blur circles even with a fully compliant APO lens.

 

Andy, you are certainly right. Thanks also for the 75 Lux example that clearly explains the difference between achromatic and apochromatic.

 

But still, this does not explain why the 90 APO renders like a classic achromatic lens.

 

If you check the Wiki graph here:

Apochromat - Wikipedia, the free encyclopedia

the apochromatic function has a clearly different shape than the achromatic function.

The achromatic function holds together red and blue, and this gives the magenta cast to object in front of focus and the green cast behind focus.

 

The apochromatic function is very different, and by my own experience using what I call real APO lenses, it typically produces a very faint cyan cast behind focus and very faint reddish cast in front of focus.

Nice example here (scroll down to the bottom): http://zeissimages.com/showreplies.php?qid=950

 

So the mistery remains: Why is the APO Summicron rendering like an achromatic lens ?

Or... had Leica marketing been drinking a couple more beers than usual at launch ? If so, maybe the real name of the lens should have been the IPA Summicron 90

[jaapv]

As Andy points out, APO designates the equal focus at three wavelengths in the focal plane only. What happens to all other wavelengths remains undefined and will vary with the individual design of the lens.

Thus there is no specific APO behaviour in this kind of test.

[CheshireCat]

Thus there is no specific APO behaviour in this kind of test.

 

But there is clearly an achromatic behavior in this kind of test. And an achromatic lens is not an APO.

[jaapv]

How do you define achromatic behaviour? The only way you can demonstrate that is by having a graph of the point of focus against an extended spectrum. Do the lines cross twice or three times? That is all that is needed for the definition.

It says very little about the performance of the lens

The only thing we can say is that apochromatic correction demands more of the design, so we can assume that the lens will be better corrected in other aspects as well. But it is not a given.

All three lenses show no chromatic aberrations in the plane of focus, so they are within the definition of APO by a safe margin.

 

To paraphrase a well-known saying from behavioural analysis of laboratory animals:

A lens that has been carefully corrected to project a perfect image in the plane of focus and is subjected to careful standardised tests will behave as it darn well pleases in the out of focus areas.

 

Not that it makes your observation less interesting. It is always good to know how our tools behave.

[CheshireCat]

How do you define achromatic behaviour?

 

An achromatic lens is corrected for two frequencies. The function is quadratic with two zeros on the plane of focus. The graph on Wiki clearly illustrates it.

This quadratic function usually has the minimum value on green, while red and blue have roughly the same value. This is why the bokeh is green behind focus and magenta in front.

 

An apochromatic lens has a cubic function, and if all zeros are in the visible spectrum, then it is impossible to see pure green or pure magenta bokeh. Also, the chromatic error will be much lower, hence very hard to notice.

 

All three lenses show no chromatic aberrations in the plane of focus, so they are within the definition of APO by a safe margin.

 

If this is true, then the Canon 85L is an APO lens

 

A lens that has been carefully corrected to project a perfect image in the plane of focus and is subjected to careful standardised tests will behave as it darn well pleases in the out of focus areas.

 

There is no such thing as a perfect image in the plane of focus.

If you carefully check the graph, you will see how higher order functions are used to greatly reduce the focus error all over the spectrum.

Here we are talking about an APO lens (three zeros) that has a very similar error to an achromatic lens (two zeros). This means that the third zero is outside of the visible range, as humans like us can clearly see green and magenta fringing.

 

So APO it may be, but not for humans...

 

If you compare this lens to the APO Macro Elmarit-R 100, you will have no doubt this lens is disappointing... at least at near focus. Which brings us to another interesting discussion.

[Erik Gunst Lund]

I can recommend to study this review of the Costal Optics 60/4 UV-VIS-IR APO macro

 

One of the only lenses that I have used that are real APO or as close as you can get for a specialist lens.

 

The UV Micro Nikkor 105mm is similar although not quite up there in Visible light...

 

The Leica M 50mm Summicron APO Asph is close but only in Visible light...

[John Black]

If this is true, then the Canon 85L is an APO lens

 

Some lenses are rumored to be APO's but not have APO's in their formal name - such as the 50 Lux ASPH M. "the 50 lux ASPH is an APO lens, containing an APO correction element. But, he thought the idea of an APO 50 was a bit silly so they never put it on the lens or in any marketing materials." From this interview with Peter Karbe, David Farkas Photography Blog: Photokina 2008 - Day 2 - Taking it easy and getting an education Not saying the 85L is or isn't an APO, but it may be. I don't know.

[henning]

Focussing distance matters. The 90AA performs better at long distances and without floating elements its performance starts breaking down at shorter distances, including its Apo performance.

 

A couple of decades ago, when process cameras were still in common use, some Apo lenses with correction optimized for the three process filters, some lenses were adjustable in their element spacing to allow acceptable performance at a couple of different reproduction ratios. Also, some LF lenses of the time could be ordered in two versions; one for landscape use, and another for close-up work. I used both versions of Apo-Ronars at times. If you used the wrong one, you did not get Apo performance.

[CheshireCat]

Focussing distance matters. The 90AA performs better at long distances and without floating elements its performance starts breaking down at shorter distances, including its Apo performance.

 

The problem is the 90AA has APO issues also at long distances.

 

But I definitely agree with you. I find the 90AA disappointing at short distances.

This brings up the new interesting discussion I was talking about in my previous post, regarding performance at short distances.

 

I recommend this very interesting piece from Marco Cavina (you will have to translate from Italian, but it is really worth it):

 

http://www.marcocavina.com/articoli_fotografici/articolo_spostamento_fuoco_brevi_distanze.pdf

 

Be sure to check the real MTF at 2 meters. This is an eye-opener if you were considering the 90AA as a portrait lens.