24-90 vs SL50

[cpclee]

Advertisement (gone after registration)

I also don’t quite follow your argument about Mandler vs Karbe designs. I’m well familiar with their characteristics and differences. But the 50 Summilux ASPH that you hold in such high esteem (as do I) is a Karbe design. In fact, it was the first of his “new paradigm” designs that broke away from the Mandler Planar / Double Gauss tradition of standard lenses. The 75AA and the 50AA that followed and now the S and SL lenses are the logical conclusion of what that lens started.

 

Fair point but I didn't ignore it. I was acknowledging that in one scenario you have both one of the best 50mm lenses ever made (albeit manual focus) and a reportedly excellent AF zoom with OIS for a marginal cost of EUR700 and in the other you just have an AF 50mm. I would think that having spent a lot of time manually focusing lenses, continuing to use one would not be a hinderance if the OP also had access to an AF lens.

 

 

I'm not an optician, I'm an aesthete and a psychologist so I have no idea and it may well be that what I call micro-contrast isn't what you call it or what optical engineers call it. But I can see the difference and someone on the Leica SL FB forum has explained it to me in really excellent terms. It's the compromise between having subtlty in tonal reproduction and a flatter contrast and sharpness and high global contrast, which this person explains as being the delineation between Mandler and Karbe designed lenses. It makes sense to me from an aesthetic point of view as that's precisely what I see as the weakness of the 50SL. But I understand that for some/a lot of people, the thing they care most about is how sharp a lens is.

 

There are really good discussion on it here:

 

http://yannickkhong.com/blog/2016/2/8/micro-contrast-the-biggest-optical-luxury-of-the-world

 

 

Again you can call what you want to call whatever you like. There is a world of difference in how one lens renders a scene versus another and that difference needs a name. In this instance I think it's micro contrast but it could also be called 'tonal palete' or 'tonal range' on 'tonal subtelty', i.e. the ability to represent a smoother and more gradual change in tonal contrast, this being critical to offering a more natural rendering of a scene.

[geetee1972]

 

Microcontrast is about the ability of a lens to render spatially nearby details as distinct. It has to do with resolution and fine structures. It is not the same as the ability to render tonally nearby hues as distinct. Tonal acuity actually has more to do with macro contrast. A lens that lets through as much light as it can without loss or compression will necessarily have high macro contrast and a wide palette of tonal values. (Macro contrast is the difference between the lightest and darkest values.)

 

Well that makes sense and is consistent with some of the arguments I've read elsewhere that state the problem with modern lenses is that they have high element counts in order to compensate for things like CA and distortion and provide a very 'sharp' image. So a lens with fewer elements will indeed let through more light and therefore have higher 'macro-contrast'.

 

I'm very happy to be told I've mixed up the terms; I'm learning this stuff all the time and while I'm comfortable in recognising what I see and what I like, I honestly don't know what the rest of the optical community uses as terminology. The article I linked to does seem to be consistent in some respects to what you're saying here and also entirely at odds with it.

 

 

I also don’t quite follow your argument about Mandler vs Karbe designs. I’m well familiar with their characteristics and differences. But the 50 Summilux ASPH that you hold in such high esteem (as do I) is a Karbe design. I

 

Yes, I asked the same question of the individual who was posting on the Leica SL Facebook page (he seemed to know the details of the design characteristics between Mandler and Karbe very well so I'm not in a position to question him).

 

His answer was that the 50mm M Summilux, being the first Karbe designed lens, was an attempt to get the best balance between the Mandler characteristic of tonal range and contrast and the new demand for high resolution, low CA optics. It was 'best compromise' result. That explanation is consistent with my experience as well.

[Advertisement]

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

[cpclee]

Modern multi coatings are very effective. Each lens-air interface lets through more than 99.9% of light. Which is to say that the light loss difference due to having 20 elements vs 5 elements is less than 0.5% and hence very negligible. On the other hand, techniques like lining the inside of lenses with light absorptive materials (velvety stuff) to minimise internal reflections makes is on the order of ten times that difference. This Leica started doing in recent years.

 

What’s prohibitive with using more lens elements is cost and size/weight, not loss of contrast.

 

Well that makes sense and is consistent with some of the arguments I've read elsewhere that state the problem with modern lenses is that they have high element counts in order to compensate for things like CA and distortion and provide a very 'sharp' image. So a lens with fewer elements will indeed let through more light and therefore have higher 'macro-contrast'.

 

I'm very happy to be told I've mixed up the terms; I'm learning this stuff all the time and while I'm comfortable in recognising what I see and what I like, I honestly don't know what the rest of the optical community uses as terminology. The article I linked to does seem to be consistent in some respects to what you're saying here and also entirely at odds with it.

 

 

Yes, I asked the same question of the individual who was posting on the Leica SL Facebook page (he seemed to know the details of the design characteristics between Mandler and Karbe very well so I'm not in a position to question him).

 

His answer was that the 50mm M Summilux, being the first Karbe designed lens, was an attempt to get the best balance between the Mandler characteristic of tonal range and contrast and the new demand for high resolution, low CA optics. It was 'best compromise' result. That explanation is consistent with my experience as well.

[Jeff S]

Karbe explains his thoughts regarding the 50 Summilux- M ASPH in this article (see both pages)...

https://www.shutterbug.com/content/leica-lens-saga-interview-peter-karbe

 

His comments regarding the glass required are especially interesting.

 

Jeff

[thighslapper]

Karbe explains his thoughts regarding the 50 Summilux- M ASPH in this article (see both pages)...

https://www.shutterbug.com/content/leica-lens-saga-interview-peter-karbe

His comments regarding the glass required are especially interesting.

Jeff

Should be mandatory reading for anyone on the forum moaning about the size/weight/cost of Leica lenses. The end result is always a well thought out compromise with image quality as the bottom line.

[geetee1972]

Modern multi coatings are very effective. Each lens-air interface lets through more than 99.9% of light. Which is to say that the light loss difference due to having 20 elements vs 5 elements is less than 0.5% and hence very negligible. On the other hand, techniques like lining the inside of lenses with light absorptive materials (velvety stuff) to minimise internal reflections makes is on the order of ten times that difference. This Leica started doing in recent years.

 

What’s prohibitive with using more lens elements is cost and size/weight, not loss of contrast.

 

 

One of my other passions was/is (i've had to step back lately due to family commitments) music and enjoying music through high end systems. I believe if you looked at photographers and 'audiophiles' as a group, you would find considerable overlap between them. I suspect it is because it offers the chance to combine aesthetics/culture with technology, but this is a different point.

 

It is well recognised (though not remotely understood) that while the audio spectrum above 20KHz cannot b e heard by the human ear, it's presence nevertheless significanty changes how the rest of the audible spectrum sounds. To put it another way, something that isn't even perceptible by the human ear, change what the human ear hears.

 

Modern lenses may well transmit 99.5% of the light entering the front element (or it may not, I honestly don't know but we do at least agree that it's not 100%. How important the light that doesn't make it through is would be an interesting question to explore.

[Daedalus2000]

Advertisement (gone after registration)

Modern multi coatings are very effective. Each lens-air interface lets through more than 99.9% of light. Which is to say that the light loss difference due to having 20 elements vs 5 elements is less than 0.5% and hence very negligible. On the other hand, techniques like lining the inside of lenses with light absorptive materials (velvety stuff) to minimise internal reflections makes is on the order of ten times that difference. This Leica started doing in recent years.

 

What’s prohibitive with using more lens elements is cost and size/weight, not loss of contrast.

 

 

I believe that the 90-280 lens is considered a lens of very high quality and performance. It has 23 elements in 17 groups... Is this is not a good example that the number of elements (at least in modern lenses) does not restrict the quality of the lens?

[cpclee]

The 90-280 is examplary as is the 24-90 (18 elements). And that’s exactly my point. These lenses achive higher everything than lenses that came before them, like the 35-70/2.8 ASPH or the 70-180/2.8APO, regardless of whether we are talking about contrast, micro contrast, consistency across apertures / distances, etc.

 

The increase number of elements and volume were instrumental in making this possible. They added degrees of freedom to the design process to allow the designer to avoid performance compromises.

 

To me Leica has been on the same path towards optical perfection for at least two decades. I don’t think their design or aesthetical goals have changed. Only refined and empowered by ever improving design tools, advances in techniques, tigher manufacturing tolerances etc. That plus the fact that the sensors today can reveal that much. If we are still shooting film these advances would not have been visible in the final product that is the image.

 

I believe that the 90-280 lens is considered a lens of very high quality and performance. It has 23 elements in 17 groups... Is this is not a good example that the number of elements (at least in modern lenses) does not restrict the quality of the lens?

[cpclee]

I’ve been shooting Leica long enough to remember. Back in the mid 1990s when Leica got started with moderning their M lens lineup (with the ASPH designs), which by that time was long in the tooth, there were raging debates about how the newer lenses produced sterile, clinical, emotionless, what-have-you images. For the initial several years the last pre-ASPH designs were fetching higher prices than the newer lenses on the used market.

 

Then came digital. It made the advantage of the newer lenses visible and controlled comparisons easy. The debates were gone. By now, everyone and their mother agrees the latest M lenses are the best they have ever been.

 

The current situation with the SL lenses isn’t all that different. The differences I’m seeing in these new designs are on the same order of magnitude improvement like what the ASPH designs brought. And it’s not stopping here. Karbe has said there is nothing like [the new designs still to come]. David Farkas has reported that with these new lenses Leica started measuring the 80lp/mm curves that previously were negligible. (That’s micro micro contrast for you, geetee...) It’s a new era. And the debate will be gone by the time 50mp SL bodies arrive because the advantage over the M lenses will be apparent and undeniable.

 

There will always be people wanting to argue that optical perfection does not equate with emotive images. I agree. But neither do optical flaws. All you can say is that awesome images can be made with whatever equipmnent.

[Leicaiste]

The 90-280 is examplary as is the 24-90 (18 elements). And that’s exactly my point. These lenses achive higher everything than lenses that came before them, like the 35-70/2.8 ASPH or the 70-180/2.8APO, regardless of whether we are talking about contrast, micro contrast, consistency across apertures / distances, etc.

 

I would love to see a fair comparison between the SL 24-90 and the R 35-70/2,8 on the SL.

[Waterden]

All you can say is that awesome images can be made with whatever equipmnent.

Quite! My move to the SL has not been driven by IQ, although the sensor is great. It is because I enjoy using it. The ergonomics are excellent and well thought out, particularly the dual-use buttons. The heft, for me, is perfect and the body is not heavy. The EVF is excellent, almost there. And I love the ability to use Leica lenses, mainly R, on a body with a sensor which has been designed with them in mind. I am sure the two native zooms are fabulous but I don't need them. For me, the SL is a camera to use sparingly and with thought. I can walk around with a pocket camera - all of the "enthusiast" ones have excellent IQ - so don't need a walk-around zoom on the SL. My R lenses cover a range from 21mm to 360mm (with extender) and all are light. The EVF means I can focus manually with ease so don't need AF. If you are a BIF or sports photographer you do need AF and the SL is not ideal for you. But it is for me.

[tom0511]

I would love to see a fair comparison between the SL 24-90 and the R 35-70/2,8 on the SL.

 

The 35-70/2.8 is a theoretical solution IMO. It is hard to find and even if you find one it is exorbital expensive. I would consider it more as a collectors item, even though someone has too much money or allready owns the lens from earlier times.

The 24-90 has a much more usefull range, lower price, IS and for sure it is good enough for me.

[Leicaiste]

The 35-70/2.8 is a theoretical solution IMO. It is hard to find and even if you find one it is exorbital expensive. I would consider it more as a collectors item, even though someone has too much money or allready owns the lens from earlier times.

The 24-90 has a much more usefull range, lower price, IS and for sure it is good enough for me.

All this is true, still I would love to see a fair comparison between the SL 24-90 and the R 35-70/2,8 on the SL.

 

The R 35-70/2,8 was supposed to be the pinnacle of the zoom lenses at the time of its release almost 20 years ago.

 

I really wonder how it stands today, assuming it is perfectly adjusted.

Edited October 8, 2017 by Leicaiste

[caissa]

The question is how long will it stay perfectly adjusted.

I would rather have a second (smaller) SL zoom (28-70 or 24-50) and see a comparison.

Edited October 9, 2017 by caissa

[robgo2]

Fair point but I didn't ignore it. I was acknowledging that in one scenario you have both one of the best 50mm lenses ever made (albeit manual focus) and a reportedly excellent AF zoom with OIS for a marginal cost of EUR700 and in the other you just have an AF 50mm. I would think that having spent a lot of time manually focusing lenses, continuing to use one would not be a hinderance if the OP also had access to an AF lens.

 

 

I'm not an optician, I'm an aesthete and a psychologist so I have no idea and it may well be that what I call micro-contrast isn't what you call it or what optical engineers call it. But I can see the difference and someone on the Leica SL FB forum has explained it to me in really excellent terms. It's the compromise between having subtlty in tonal reproduction and a flatter contrast and sharpness and high global contrast, which this person explains as being the delineation between Mandler and Karbe designed lenses. It makes sense to me from an aesthetic point of view as that's precisely what I see as the weakness of the 50SL. But I understand that for some/a lot of people, the thing they care most about is how sharp a lens is.

 

There are really good discussion on it here:

 

http://yannickkhong.com/blog/2016/2/8/micro-contrast-the-biggest-optical-luxury-of-the-world

 

 

Again you can call what you want to call whatever you like. There is a world of difference in how one lens renders a scene versus another and that difference needs a name. In this instance I think it's micro contrast but it could also be called 'tonal palete' or 'tonal range' on 'tonal subtelty', i.e. the ability to represent a smoother and more gradual change in tonal contrast, this being critical to offering a more natural rendering of a scene.

 

 

I have been following Yannick Khong's blog for a while, and I think that he uses incorrect terminology. What Khong calls "microcontrast" might better be called "microtonality," which is to say the separation of very subtle differences in tone. However, I agree with his conclusion that this fine tonal discrimination is what lends depth and fullness to an image. Some lenses may be very sharp and detailed, but they render perceptually flat images, because they lack this characteristic. As for myself, I like depth rendering.

[thighslapper]

I have been following Yannick Khong's blog for a while, and I think that he uses incorrect terminology. What Khong calls "microcontrast" might better be called "microtonality," which is to say the separation of very subtle differences in tone. However, I agree with his conclusion that this fine tonal discrimination is what lends depth and fullness to an image. Some lenses may be very sharp and detailed, but they render perceptually flat images, because they lack this characteristic. As for myself, I like depth rendering.

 

Having read this as well and a selection of other articles on this subject I remain rather sceptical. 

 

If you cannot accurately measure it and attach a number to it I am inclined to doubt some of the statements made. Particularly the dogmatic 'Sigma Art lenses have no (sic) micro-contrast' which does make me doubt the rigour of Mr Khong's analysis ..... 

 

Much of this is wooly perceptionism with all the bias that entails. MTF's I can understand ... but 'micro-contrast' which has no apparent units of measurement is another matter .....

 

Where does it exactly fit in between Macro, Pico and Nano Contrast ????

Edited October 9, 2017 by thighslapper

[dgktkr]

I have been following Yannick Khong's blog for a while, and I think that he uses incorrect terminology. What Khong calls "microcontrast" might better be called "microtonality," which is to say the separation of very subtle differences in tone. However, I agree with his conclusion that this fine tonal discrimination is what lends depth and fullness to an image. Some lenses may be very sharp and detailed, but they render perceptually flat images, because they lack this characteristic. As for myself, I like depth rendering.

 

"microcontrast" (?), "microtonatlity (?) What about just plain old "contrast"? Isn't it contrast that separates differences in tone? In digital photography there is a lower limit to the size of the tone differences that can be recorded (effective bit depth), but that is a property of the sensor and associated processing, not the lens.

 

In his article, Yannick shows two examples of a transition between light and dark. One happens more abruptly than the other, but I don't discern a difference in the number of gray levels that are displayed. On my screen I can clearly see that the grays are not continuous, indicating that there is not enough bit depth coming through on my monitor to make the transitions from one gray level to the next look smooth.

 

dgktkr

[robgo2]

"microcontrast" (?), "microtonatlity (?) What about just plain old "contrast"? Isn't it contrast that separates differences in tone? In digital photography there is a lower limit to the size of the tone differences that can be recorded (effective bit depth), but that is a property of the sensor and associated processing, not the lens.

 

In his article, Yannick shows two examples of a transition between light and dark. One happens more abruptly than the other, but I don't discern a difference in the number of gray levels that are displayed. On my screen I can clearly see that the grays are not continuous, indicating that there is not enough bit depth coming through on my monitor to make the transitions from one gray level to the next look smooth.

 

dgktkr

 

 

Perhaps I am inventing a new term, but I believe that there is a genuine difference between global contrast and microtonality. You have to look closely for the latter, but it is there (or isn't), and it makes a significant difference in the sense of depth conveyed by an image. In my experience, microtonality is not only a function of the lens but also of the raw converter. Some converters excel at separating close tones, while others do not.

Edited October 10, 2017 by robgo2

[dgktkr]

Perhaps I am inventing a new term, but I believe that there is a genuine difference between global contrast and microtonality. You have to look closely for the latter, but it is there (or isn't), and it makes a significant difference in the sense of depth conveyed by an image. In my experience, microtonality is not only a function of the lens but also of the raw converter. Some converters excel at separating close tones, while others do not.

 

With a scientific background, which includes optics, I'm of the opinion that global contrast refers to low spatial frequency MTF response and microtonality refers to high spatial frequency MTF response. And I agree that both of those depend on the lens, sensor and digital processing. 

 

What I would like to know is what Yannick is seeing when he describes "micro-contrast". He claims it is not measurable. OK, I accept that as a possibility, but can he describe it well enough and are there examples of images where I can see it for myself?

 

In some of his other, related, articles, it seems to me that the differences he is demonstrating can be explained by differences in setting the black level and in the differences lenses have with regard to veiling flare. Those can affect shadow contrast and color saturation especially in shadows. All lenses have veiling flare to some degree. Some writers ascribe some of the reputation of Leica lenses to be due to lower amounts of veiling flare as a result of more careful design and more effort in baffling and blackening the inside of their lenses and also by using more effective antireflection coatings that cut down on diffuse light bouncing around inside the lens.

 

Yannick seems to be asserting that micro-contrast is not the contrast of fine details in a scene (i.e. sharpness). In fact, he shows two black to white gradients, one of which is more abrupt than the other. He claims the more gentle one has greater micro-contrast even though neither of them appears to have fine details. Except for, perhaps, the visible discrete steps in the shades of gray. On my PC and to my eye, the one with the fast transition from dark to light has more distinct steps in gray values. The one with the more gradual transition still seems to have steps in gray values, but they definitely are more subtle. If that is what he calls micro-contrast, then I would say that that doesn't depend on the lens. The gradients of light intensity imaged by a lens are continuous; they aren't the cause of the discrete steps of gray in Yannick's gradient images.

 

He further seems to be asserting that he can sense whether a lens has good micro-contrast even if an image made with it is unsharp or out of focus! Could he be suggesting that what he calls micro-contrast involves bokeh? 

 

dgktkr

Edited October 10, 2017 by dgktkr

[Advertisement]

Hallo,

Look here: Leica L-Mount

[robgo2]

With a scientific background, which includes optics, I'm of the opinion that global contrast refers to low spatial frequency MTF response and microtonality refers to high spatial frequency MTF response. And I agree that both of those depend on the lens, sensor and digital processing. 

 

What I would like to know is what Yannick is seeing when he describes "micro-contrast". He claims it is not measurable. OK, I accept that as a possibility, but can he describe it well enough and are there examples of images where I can see it for myself?

 

In some of his other, related, articles, it seems to me that the differences he is demonstrating can be explained by differences in setting the black level and in the differences lenses have with regard to veiling flare. Those can affect shadow contrast and color saturation especially in shadows. All lenses have veiling flare to some degree. Some writers ascribe some of the reputation of Leica lenses to be due to lower amounts of veiling flare as a result of more careful design and more effort in baffling and blackening the inside of their lenses and also by using more effective antireflection coatings that cut down on diffuse light bouncing around inside the lens.

 

Yannick seems to be asserting that micro-contrast is not the contrast of fine details in a scene (i.e. sharpness). In fact, he shows two black to white gradients, one of which is more abrupt than the other. He claims the more gentle one has greater micro-contrast even though neither of them appears to have fine details. Except for, perhaps, the visible discrete steps in the shades of gray. On my PC and to my eye, the one with the fast transition from dark to light has more distinct steps in gray values. The one with the more gradual transition still seems to have steps in gray values, but they definitely are more subtle. If that is what he calls micro-contrast, then I would say that that doesn't depend on the lens. The gradients of light intensity imaged by a lens are continuous; they aren't the cause of the discrete steps of gray in Yannick's gradient images.

 

He further seems to be asserting that he can sense whether a lens has good micro-contrast even if an image made with it is unsharp or out of focus! Could he be suggesting that what he calls micro-contrast involves bokeh? 

 

dgktkr

 

 

Well, I have no background in optics, but I am fairly experienced photographer who has won some trivial awards and had some images and one major project published in reputable places. And I have a story to tell about depth rendering. About 2 years ago, when I was using a Sony A7ii as my main camera, I had the opportunity to perform a direct comparison of the Zeiss Batis 85 f/1.8 and the CY 85 f2.8. Both lenses were shot at f/2.8 to take identical portraits of my friend, who happened to be the owner of the Batis. Well, as expected, the Batis images were absolutely pin sharp and amazingly detailed, but they were also as flat as a pancake. The nose, the cheeks and the eyes all appeared to be in the same plane. Meanwhile, the CY images were very sharp, and they also showed tremendous depth. The facial features appeared to be rounded and separated in different planes. Upon close inspection, it was obvious that the CY photos had much richer tonality, by which I mean many more subtones within a given tonal range. There was nothing that I could do in post to bring the Batis photos to that level of "microtonality," or whatever it might be called. My friend agreed, and I have no idea if he has kept the Batis or sold it.

 

Microtonality is an important, but almost always overlooked, feature of photographic images. Yannick Khong most certainly misuses the term "microcontrast," but I am convinced that the point he is trying to make is valid. None of which, BTW, has any bearing on the main subject of this thread, which has to do with the SL 50 vs. the SL 24-90.

Edited October 10, 2017 by robgo2