Show me your favorite lenses

[OR120]

Advertisement (gone after registration)

2 hours ago, adan said:

For me personally (and sorry to bore the old-timers who have heard this many times before):

1) I find that older lenses, at least back to the 1970s, image just as well as (and in some ways better than) the modern lenses. With a few exceptions - it is a spectrum. That has nothing to do with "character"

Here is an f/2.8 image with a 1974 90mm Tele-Elmarit-M (not always considered one of the best). Full image and detail crop (M10).

Is there any modern lens that would offer a significant improvement over the detail it records? Is there a rational reason to carry something larger and heavier - and yes, more expensive?

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!

 

2) The main reason Leitz/Leica lenses "of a certain age" image better, for me, is primarily lower global contrast (more shadow and/or highlight information) for a given exposure. Especially with color slide film (professionals did not generally use color neg film - that was for family snapshots and the lower forms of posed "grip-and-grin" wedding pictures of the past...

https://www.friartux.com/blog/post/shades-of-blue-vintage-wedding

....and now with digital.

High contrast limits dynamic range - and in fact can cause/emphasize purple fringing, as blown whites or blues (sky) bleed into surrounding colors or black. See lct's example.

If a picture needs more contrast, that is much, much, much easier to add back, than trying to recover blocked shadows or blown highlights.

The second reason is that those same lenses generally produce more neutral, accurate color, compared to the red/magenta/purple "skin tones above all else" bias of the post-1989 Leica (not Leitz) lenses. Purple bias cannot be fixed with WB alone, without distorting other colors. I prefer skin tones that look sun-tanned, not sun-burned (nor pale and washed-out).

And additonally, I like dark architectural green paint to render as green, not as red/magenta-infected brown or gray.

(1980 90mm Summicron-M, M10)

Note that the greenish 1980 lens does not de-emphasize real magentas and reds. If anything, it enhances them and makes them *pop*, as they contrast with the overall greener rendering.

See: complementary colors - an essential part of understanding art and design. https://en.wikipedia.org/wiki/Complementary_colors

Now, there is room for taste in picking color rendering - but that is my taste.

And with neg-film, higher contrast can sometimes carve a crisper image into the fuzzy gelatin.

I had a very contrasty Zeiss G 90mm Sonnar back in 2001 that did that nicely on cloudy days on B&W film, but under raw sunlight (which is the norm here in Colorado, 300+ days a year), it would overexpose and block highlights (skin), and underexpose shadows (clothes) to blank film - in the same exposure!

I switched to Leica M and 1980s lenses - much better control of tonality.

How lenses of different ages render is also not a hard line. It is a continuum, and there are fuzzy edges. I like the color of my 50 Summicron v.5, and the contrast is generally acceptable (but one must remember that - optically - it is really just a 50mm v.4 from 1980, repackaged in a new barrel). The 75 APO-Summicron and v.1 35 Summilux ASPH (now a 26-year-old design) are also generally acceptable.

But I would not use a 35 Summicron ASPH if someone paid me $3700! Too contrasty, too pink. And too heavy, compared to its predecessors.

Lower form grip and grins (not just weddings) - the life source of many young photographers from long, long ago

[pgk]

But it will allow you to see into the shadows better. Lower contrast means a lower ability to record tonality from the original scene. So a contrastier lens will actually allow more shadow tonality to be recorded. A lower contrast lens will record less. That is the whole point. As I said before, a very low contrast (old) lens produces a limited tonal range. A low contrast lens cannot produce better shadow detail. Its about both overall contrast and micro-contrast. Modulation Transfer Function charts will show this too, and to quote from Nikon "An MTF chart plots the contrast and resolution of a lens from the center to its edges against a "perfect" lens that would transmit 100% of the light that passes through it. The contrast of a lens is important as this works in correlation to lens resolution." In turn this means that a lower contrast lens is not passing 100% of the light which enters it and so shadows will lack tonality if the highlights are retained. So from this it is obvious that a lower contrast lens can only provide better shadow information if exposure is increased - but then highlight data will be deficient. 

[Advertisement]

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

[lct]

3 hours ago, adan said:

Here is an f/2.8 image with a 1974 90mm Tele-Elmarit-M (not always considered one of the best). Full image and detail crop (M10).

Is there any modern lens that would offer a significant improvement over the detail it records? Is there a rational reason to carry something larger and heavier - and yes, more expensive?

People preferring more acutance can find it in smaller or slightly taller lenses like the Macro-Elmar 90/4 or the Summarit 90/2.4. Matter of tastes IMHO.

Edited June 7, 2020 by lct

[Guest JMF]

1 hour ago, pgk said:

But it will allow you to see into the shadows better. Lower contrast means a lower ability to record tonality from the original scene. So a contrastier lens will actually allow more shadow tonality to be recorded. A lower contrast lens will record less. That is the whole point. As I said before, a very low contrast (old) lens produces a limited tonal range. A low contrast lens cannot produce better shadow detail. Its about both overall contrast and micro-contrast. Modulation Transfer Function charts will show this too, and to quote from Nikon "An MTF chart plots the contrast and resolution of a lens from the center to its edges against a "perfect" lens that would transmit 100% of the light that passes through it. The contrast of a lens is important as this works in correlation to lens resolution." In turn this means that a lower contrast lens is not passing 100% of the light which enters it and so shadows will lack tonality if the highlights are retained. So from this it is obvious that a lower contrast lens can only provide better shadow information if exposure is increased - but then highlight data will be deficient. 

True !

Initially the discussion was  about the overall rendition of the lenses. More detail on faces might not be desired at all times . So as no lens can fit all the situations, we were discussing their pros and cons and price tags.

[adan]

1 hour ago, pgk said:

For anyone who thinks it does here's the challenge - come up with a good technical explanation.

Sure.

If you assume an ideal media (film or sensor) with infinite dynamic range, then that theory is probably right.

Or if one is talking about the limited case of edge contrast, but I was very, very, very careful to specify global (overall, black to white) contrast.

Since there is no such ideal medium, theory does not apply. The medium, whether film or digital, has a limited DR.

Let us set theory aside, and do an empirical experiment. (Basic equation: 1 mg of empiricial evidence > 1000 tonnes of theory.)

Same exact scene, same exposure, shot within a minute of each other on a cloudless day. Identical postprocessing. Shot with an M10 and 1) 28 Elmarit-M ASPH, and 2) 21mm Elmarit-M (1980).

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!

 

With regard to highlight and shadow detail, texture, separation, and definition, tell me what you see, in these detail crops. Then I will give you the technical explanation.

 

[Dennis]

Reading the crops, I see more details, more DR in the photos taken with the 21. But to be honest with you, I have also to say that I prefer the "look" of the 28 crops. There is less information, yes, but a better overall appearance. But this is me. De Gustibus non est disputandum

[lct]

Advertisement (gone after registration)

The point, how i understand it, is to know which lenses allow to recover shadow details more easily. There is only one answer AFAIK: less contrasty lenses. Doesn't mean that i prefer them for other reasons.

[pgk]

No, no, no. Look, what you are seeing is marginal veiling flare spillage into the shadow areas - its spurious data. To be anything else you would have to have an MTF for the lower contrast lens which exceeds that of the higher contrast lens. You cannot recover shadow data by using a low contrast lens.

I'll add in that if you think about the MTF data, if a higher contrast lens has a high MTF figure at low spatial frequencies, then it i capable of transferring more data through it then an lens with a lower contrast lens and a lower MTF figure. You simply cannot extract more information from a lesser quantity of data.

Edited June 8, 2020 by pgk

[pgk]

I should have added that 'global' lens contrast is represented by low (coarse) spatial frequency.

[pippy]

11 hours ago, adan said:

...Let us set theory aside, and do an empirical experiment. (Basic equation: 1 mg of empiricial evidence > 1000 tonnes of theory.)...

With regard to highlight and shadow detail, texture, separation, and definition, tell me what you see, in these detail crops. Then I will give you the technical explanation...

Succinctly? I see, in the 28mm image, a snap which has been over-exposed by 1/2 stop as the highlight detail has been lost............

We, here, could always go into the whole D-Max-Rel.Log-e-toe-straight-line-shoulder discussion - I still have my copy of 'Langford' to hand - but that would be rather dull and, in any case, I'm absolutely sure the vast majority of you know it all anyway.

Incidentally, adan, I don't fully agree with you on the 'empirical evidence v's theory' calculation because we would have to assume that whichever specific 1mg of empirical evidence we choose to hold up as our example is factually correct in every circumstance - and not every 1mg of empirical evidence IS factually correct in every circumstance. It has been known, through the whole of recorded history, for many people in many areas to choose to display certain 'empirical evidence' which backs-up their own personal viewpoint as 'absolute proof'. Just ask Galileo...

Philip.

Edited June 8, 2020 by pippy

[lct]

2 hours ago, pgk said:

No, no, no. Look, what you are seeing is marginal veiling flare spillage into the shadow areas - its spurious data. To be anything else you would have to have an MTF for the lower contrast lens which exceeds that of the higher contrast lens. You cannot recover shadow data by using a low contrast lens.

I'll add in that if you think about the MTF data, if a higher contrast lens has a high MTF figure at low spatial frequencies, then it i capable of transferring more data through it then an lens with a lower contrast lens and a lower MTF figure. You simply cannot extract more information from a lesser quantity of data.

Yes, yes, yes . Sorry we won't agree on that. I'm no techie at all but i have both contrasty and less contrasty lenses and i know for fact that i can recover shadow details more easily with the latter ones. The rest is pure theory to me, which doesn't mean that it is not interesting for others of course .

 

[Keith (M)]

Going back to the original purpose of this thread...

Three M's, each with a 50mm - M240 + 5cm Summicron Collapsible, M10M + 50mm Summilux-M ASPH, M7 + 50mm C-Sonnar.  Interspersing the M's - 75mm Summarit f2.5, 90mm Summarit f2.3, 135mm APO Telyt-M.  Parading in front - 28mm Elmarit ASPH (Mk 1), 35mm Summaron f2.8, 35mm ZM Biogon f2.

Favourite(s)?  Well, the C-Sonnar is pretty much welded to the M7, the Summilux finds favour on the M10M but if I could only take one lens it would likely be the Summaron.  Superbly built, beautiful appearance and a delight to handle. Lots of character too.

(A D-Lux 7 photo)

 

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!

Edited June 8, 2020 by Keith (M)

[Mark T]

3 minutes ago, Keith (M) said:

Going back to the original purpose of this thread...

Three M's, each with a 50mm - M240 + 5cm Summicron Collapsible, M10M + 50mm Summilux-M ASPH, M7 + 50mm C-Sonnar.  Interspersing the M's - 75mm Summarit f2.5, 90mm Summarit f2.3, 135mm APO Telyt-M.  Parading in front - 28mm Elmarit ASPH (Mk 1), 35mm Summaron f2.8, 35mm ZM Biogon f2.

Favourite(s)?  Well, the C-Sonnar is pretty much welded to the M7, the Summilux finds favour on the M10M but if I could only take one lens it would likely be the Summaron.  Superbly built, beautiful appearance and a delight to handle. Lots of character too.

(A D-Lux 7 photo)

 

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!

Much enjoyed! Thanks for refocusing the discussion... much appreciated 🙂

[pippy]

1 minute ago, Keith (M) said:

...Three M's, each with a 50mm - M240 + 5cm Summicron Collapsible...

I'd be very interested to hear how, in your own experience, your Summicron holds-up in comparison to the newer designs, Keith. I've been hankering after one for a while but have always been put-off by the stories of the'chalk-soft front element'. Is this actually the case and, if so, how bad would the lens need to be marked to make it a poor performer?

Philip.

[lct]

4 minutes ago, Mark T said:

Much enjoyed! Thanks for refocusing the discussion... much appreciated 🙂

Sorry for the OT folks. I will show a couple favorites when i can figure out which ones i prefer on which camera. On the CL perhaps for a change if i'm not OT again...

[adan]

Well, I will close out with this.

Which I do not see as a digression. I explained why my favorite lenses are my favorite lenses - and someone jumped in to claim that that was an invalid reason to prefer those lenses.

Pippy, if the 28mm shot is overexposed - why does it reproduce red-painted concrete blocks as black? And why is that inaccuracy preferable?

That is actually and exactly my main point - a high-contrast lens simultaneously overexposes the highlights and underexposes the shadows. Thus destroying information about the subject.

(If I, as a photographer, make a creative choice to remove that information by tonal adjustments, that is one thing; to have it forced upon me in any and all pictures by the lens is a presumption up with which I will not put.)

We can see white speckles at the base of the dark red wall (an accurate reproduction of what was there - top large pair of examples) better with the (moderately) lower-contrast lens, while the higher-contrast lens suppresses that reality.

Detail is "what can be seen." If a lens hides it unnecessarily, either by being unsharp, or by pushing it beyond the recording ability of the medium, it is failing in its function. A microphone that cannot capture the fullest range of human hearing and deliver it to tape, and clips the highest notes and the lowest notes - is a poor microphone.

As to Galileo: it was Galileo who presented empirical observation in the face of "received wisdom" and theory from a book. Tell me again, which of us is Galileo? I stand by my "heresy," regardless of what the priesthood of contrast may claim.

 

[Keith (M)]

2 hours ago, pippy said:

I'd be very interested to hear how, in your own experience, your Summicron holds-up in comparison to the newer designs, Keith. I've been hankering after one for a while but have always been put-off by the stories of the'chalk-soft front element'. Is this actually the case and, if so, how bad would the lens need to be marked to make it a poor performer?

Philip.

I recall reading that one of the rules pertaining to this lens is "never touch or attempt to clean the front element".  Chapter & verse can probably be found in the 'Collectors & Historica' section.  My usage has been mostly film (IIIg) until my M10M arrived in early Feb and since then my tests of it have shown a very pleasing performance and given that it is described by E Puts as a "low contrast lens" I find the soft, tonal transition very pleasing to the eye.  M10M & film results can be seen in this Flickr album.

Given E Puts's description, I found it interesting to see how (on the M10M) the histogram sometimes stops short of the '0' point (a quick test shot in the kitchen, imported into LR then 'screen-grabbed' & exported as a .jpg).

And now back to your regular programme...

 

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!

[pgk]

16 minutes ago, adan said:

Which I do not see as a digression. I explained why my favorite lenses are my favorite lenses - and someone jumped in to claim that that was an invalid reason to prefer those lenses.

Low contrast is not an invalid reason. My point is that the idea that low contrast lenses allow greater shadow detail to be captured is wrong. I prefer some lower contrast lenses to new ones myself but I try to understand why they work as they do because lenses interest me - the ability to shape glass to form such exquisitely intricate and detailed images is an extraordinary triumph of technology. I will photograph my favourite set soon and post them - they are a mix of current and older.

[ianman]

28 minutes ago, adan said:

Pippy, if the 28mm shot is overexposed

There's no "if" about it. The details in the white bricks are completely lost.

28 minutes ago, adan said:

That is actually and exactly my main point - a high-contrast lens simultaneously overexposes the highlights and underexposes the shadows. Thus destroying information about the subject.

Clearly, in this example, you are right.

Edited June 8, 2020 by ianman

[Advertisement]

Hallo,

Look here: Leica M

[pippy]

1 hour ago, adan said:

...Pippy, if the 28mm shot is overexposed - why does it reproduce red-painted concrete blocks as black? And why is that inaccuracy preferable?

That is actually and exactly my main point - a high-contrast lens simultaneously overexposes the highlights and underexposes the shadows. Thus destroying information about the subject...

As to Galileo: it was Galileo who presented empirical observation in the face of "received wisdom" and theory from a book. Tell me again, which of us is Galileo? I stand by my "heresy," regardless of what the priesthood of contrast may claim.

As to the first part we both know the answer is because the dynamic range of the image-area is greater than the capability of the sensor to record, simultaneously, information at both the tip of the toe and at the end of the shoulder in the rel.log-e curve for that particular set of circumstances.

I was going to write a load of stuff about the "0.5-256, 1-10 Zone System" but, as we both know it anyhow, I will shorten it to this; if - just by way of example - you live in an area where a Log-E of "0.5-2048" (or higher) is the order of the day then the Mandler-era lens is understandably your best choice. If I live in an area where "0.5-256" (or lower) is 'the norm' then the more modern higher-contrast lenses will be the clear winner over the older designs. Such are our respective choices; no one is in the wrong as far as that is concerned.

And Galileo v's 'The Firm' was a two-way street as both parties utterly believed that they had the necessary, unarguable empirical evidence to back up their own claim - hence my earlier comment about "certain empirical evidence".

And, of course, there is no accusation of Heresy............just an acceptance of a belief in slightly different faiths!

ANYHOW...back on topic!

If I'm allowed one more 'favourite' to add to my earlier three I'd choose this; my 1954 f3.5 Summaron;

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!

Philip.

Edited June 8, 2020 by pippy