The lack of 3D information in modern corrected lenses?

[Redseele]

Just saw this article about how modern lenses (highly corrected with many elements) flatten images, and how older ones (with low count of elements) are able to reproduce 3d better precisely because of the lack of such corrections (which make modern lenses lose visual information).

His examples (some of them at least) show his point very well, but I was wondering what you guys think. I do think that older lenses (like my former Summaron 3.5 or my Summicron 50 collapsible) have a very nice "rendering" that makes things appear more "real" than, say, a Voigtlander Nokton classic 40mm (which I used to own but for some reason I never really liked it that much) or very modern lenses for SLRs (some of which produce really sharp but "inert" images in my opinion).

What do you guys think? I don't know very much about optics theory, but I'm very curious to hear what some people who know a lot more have to say on this topic. Here's the link to the original article.

http://yannickkhong.com/blog/2015/10/4/the-flattening-of-modern-lenses-or-the-death-of-3d-pop

 

[mfogiel]

That is only part of the story. The other part regards the coatings. In fact, Zeiss non aspherical lenses tend to be among the most 3D also because Zeiss has been a pioneer in modern coatings. Modern Tessar type 4 element lenses are probably among the obvious examples of 3d rendering:
Zeiss Contax 45/2.8 Tessar
Nikkor 45/2.8 P
Elmar M 50/2.8 v2

The shot below was made with the Elmar.

201211121 by marek fogiel, on Flickr

 

[Bille]

What do you guys think?

Plain nonsense.

 

[BernardL]

His examples (some of them at least) show his point very well

Until proof of the contrary, I think this guy is self-deluding. Comparing different lenses with different scenes (except the Mickey Mouse)! What can you conclude?

I would be convinced if the "true believers" would pick the correct type of lens 80% of the time from randomized pairs of pics of the same scene taken with both lenses. All the rest is just an experiment in psychology.

Make an experiment: print the two Mickey pics. Trim both in the same way, to remove text at top and bottom. Shuffle them, or, better, have someone shuffle them while you look away. NOW, can you pick out the one that has that obvious 3D look?

And the pseudo-scientific "explanation"!

 

[CMur12]

There was some recent contentious discussion about this, but the Angry Photographer withdrew his posts, so the thread now has conspicuous gaps.

Though the "scientific" evidence presented strikes me as incomplete or not entirely convincing, the idea resonates with me on intuitive and personal aesthetic levels. I have always had the thought in the back of my head that the simpler the lens that can do the job, the "purer" the image. (Please note that I do not present this thought as fact.)

It was with this idea in mind that I bought a Yashica D TLR with the three-element Yashikor lens, feeling that a slightly softer lens that vignettes more than others would be a good starting point for portraits. I'm also fond of Tessar and Tessar-type lenses in medium format.

- Murray

 

[grouchos_tash]

I don't think I could tell what lens took what picture but I know what I like. I can see a big difference with my Elmar-m compared to any other lens I've used, but that could just be because it's the best quality lens I have? 🙂

 

[sevo]

Two out of three examples are utterly besides the point - different subjects in different lighting - and the trailing ones don't even attempt a comparison. The Mickey Mouse one is the only example which might perhaps be relevant - but I can't tell whether its obvious focus difference is due to a slight front focus on the Sigma or whether the field curvature of the (simpler) Nikkor turns into an advantage on this particular (convex) subject. But even the latter, while a lens artefact, is not "3D", and would turn into a disadvantage on a flat or concave subject.

And besides being unable to demonstrate the proposed phenomenon, the presented theory is painfully, ridiculously wrong. "Light adopts a spiral behavior that spins into the lens. That’s where manufacturers use special compound glass elements (Nikon ED glass per example) to homogenize the light faster in order to save on size." is beyond embarrassing. Whoever wrote that is utterly ignorant regarding optics. Welcome to the temple of esoteric photography...

 

[Michael Markey]

Stopped reading when I saw who the author was.

 

[icebear]

Rofl :d 12345

 

[JP Owens]

Over the last 50 years or so, I've shot just about every variety of lens out there, from cheap to very expensive. Apart from CA issues with some and SA issues with others, I've not noticed much difference among them in the final photos. There are just too many other factors that can alter "rendering" between the lens and a finished photo.

I've never fretted about the "rendering" of a lens.

 

[Griffin]

Wow I never knew that different colored lights travel at different speeds...

 

[Ezzie]

Nor did I. C as far as I remember is constant, however frequency differs with changes in wavelength.

 

[Corran]

Total BS - it's all confirmation bias.

The most important part for the "3D" look to a photograph is the aperture and the focus distance, but bokeh and sharpness also play into it. If you get the right aperture and focus distance, the whole subject will be in focus and the background smoothly melts away, giving that "cutout" look. Sharpness helps, to make the focus plane stand out more, and bokeh is also important for that smooth background - busy bokeh makes it less obvious and detracts from the 3D "pop."

Here's an example - this one from a Sonnar lens:

And another, this one from a modern lens, the 50mm f/1.1 Nokton:

These both exhibit it I think - one modern, one not, which therefore makes the theory pretty much null and void.

 

[willie_901]

At this point in time all photography media are two-dimensional (stereo cameras excepted). Lenses can not alter this.

Lenses with modern coatings and highly corrected optical designs do render differently than lenses based on older technologies. Human visual perception expects 3D, so it is possible to create an illusion of another dimension. The 3D look is a complicated combination of optical rendering, post-production rendering and human perception. I suggest a scene's lighting is also an important factor.

Lens rendering is a highly subjective topic. Preferring the rendering of older lenses is common and expected. It's wonderful to have choices.

 

[willie_901]

To date all speed of light measurements indicate light speed is frequency independent.

However some aspects quantum electrodynamic theory and quantum gravity theory predict otherwise. Perhaps those theories are wrong or we just haven't performed the right experiments yet.

 

[photomoof]

If one really wants to learn about 3D images, start by watching a these two 3D movies in 2D and see how the cinematographers create 3D.

First the classic Hitchcock's "Dial M for Murder," which was released in 3D in 1954, and now is mostly seen on the small screen in 2D.

Vim Wenders' "Everything Will Be Fine" (2015) was another intimate movie which was shot in 3D, and is available online in 2D.

When you watch these films, there are numerous obvious examples of what causes a 3D effect, and many more subtle ones.

Position of objects,
Sharp delineations of in focus, and out of focus, large objects,
Light and Shadow,
Intense side lighting,
Etc

One can quickly learn why these 3D movies retain most of their 3D quality, even when watched in 2D!

Obviously "Dial M for Murder" uses much older lenses than Wenders did in 2015, but the films are surprisingly similar. Although Wenders uses zoom to create 3D, obviously not available in still images.

The examples by "Corran" on this thread really are pretty clear examples of what we perceive as 3D.

Watch the films, especially "Dial M for Murder." Hitchcock set up one shot after another, to make the best use of 3D.

One can learn from Robert Burks and Benoît Debie, or from ...

However some aspects quantum electrodynamic theory and quantum gravity theory predict otherwise. Perhaps those theories are wrong or we just haven't performed the right experiments yet.

Or you can just put a lens on your camera, and take some photos. That's what I do. 🙂

 

[brbo]

Stopped reading when I saw who the author was.

My first thought, without even opening the link, was "AngryPhotographer".

And then I open the link and the first name I see is... Ken Wheeler.

(Will RFF ever recover from the embarrassment of letting this joker run it's own subforum here?)

 

[mfogiel]

Joker or no joker, but the 3d effect is our there. Anyone who has used a Hasselblad SWC will know what I mean. And it has nothing to do with DOF.

DROPLETS ON CABBAGE, MANOIR DE LA BRUNIE by marek fogiel, on Flickr

 

[brbo]

Joker or no joker, but the 3d effect is our there. Anyone who has used a Hasselblad SWC will know what I mean. And it has nothing to do with DOF.

Then just insert a quote from the OP's "article". There's a marvelous "explanation" of it (and it even doesn't have anything to do with DOF). Really, I dare you. Quote it here. With a straight face.

😀

 

[photomoof]

Joker or no joker, but the 3d effect is our there. Anyone who has used a Hasselblad SWC will know what I mean. And it has nothing to do with DOF.

And yet you show as an example, an image which depends totally on light, shadow, and a soft focus background for its 3D effect? 🙂