The 'flattening of modern lenses'

[kuuan]

some of my older lenses, in a 'brickwall - sharpness" kind of scenario, show strongly diminishing sharpness towards the corners, however if focused there they may, even at wide apertures, produce decently sharp corners.
In other words they show strong "field curverture". Modern lenses hardly will have that but sometimes I get the impression that strong field curverture, in certain scenarios, can enhance the impression of 3D.

 

[RichC]

As others have said, the two linked articles are pseudoscientific drivel - a statement made wearing my BSc science and MA photography hats! The explanations are incorrect and full of opinion and errors, and the differences between compared images clearly owe more to varying lighting, subject distance, field of view and the like (when the subjects aren't different that is - which immediately negates everything!).

Modern lenses undoubtedly produce a different look than older lenses - which is down to current lenses being technically superior, using better optical designs and materials. I've heard it said that today's general preference for photographs that are very sharp and high contrast (the "white highlights, black shadows" look) began with the influx of Japanese cameras to the West in the 1960s, with Japan preferring that "technical" (some would say "clinical") look in photos. This may well be true: in contrast to the typical Western preference for tonality in B&W photographs, there is a long tradition of extreme contrast and abstraction in Japanese photography, as in Fukase's famous work "The Solitude of Ravens" (http://www.michaelhoppengallery.com/exhibitions/127/overview/#/artworks_standalone/10108).

Older lenses have more aberrations, which is not necessarily a bad thing. For example, I like 1970s Leica lenses, especially those designed by Mandler: these lenses are sharp but tend to suffer from flare more than modern lenses. Leica seems to have worked with this flaw, so rather than it producing ugly pictures, the flare tends to gently lighten shadows and suffuse highlights and creates images with character (presumably the so-called Leica "glow" some people mention). The more these lenses are stopped down, the more closely they perform to modern lenses.

Kuuan above mentions that soft corners can give photographs more of a "3D impression". Another technical flaw - one that isolates the subject of the picture and so may actually give an appearance that is preferred.

At the end of the day, it's all down to personal preference and the look you want in photographs. There are no such things as good and bad lenses... just ones you dislike!

 

[kuuan]

..
Kuuan above mentions that soft corners can give photographs more of a "3D impression". Another technical flaw - one that isolates the subject of the picture and so may actually give an appearance that is preferred..

right, though enhanced 3D impression through soft corners is not entirely what I had meant.
It also could be enhanced when various contents, though being on different focal planes, are sharp and that only because of strong field curverture of the lens.

 

[flavio81]

The cited article is full of pseudoscience and silly claims, BUT there is truth at the bottom:

In the past (1950s, 1960s), it was hard to correct for excellent wide-open sharpness with good contrast and low distortion and low chromatic abberration and low focus shift and low curvature of field.

So compromises had to be made and those designers knew they could take the flaws the lens was going to have anyways (i.e. residual spherical aberration, residual vignetting, etc) and "balance" them to provide nice bokeh, nice rendering overall.

This you can find in many many 50s and early 60s lenses.

Now fast forward to 2016 and everybody is obsessed with megapixels and DxO benchmarks and the ultimate sharpness that can be put into the sensor. Thus, manufacturers design their lenses accordingly. And the lens that puts ultimate cripness into the sensor may not necessarily be the one with the most likeable or "rounded" or "3D" picture rendering.

It depends on what is the goal. Fortunately, with the trend of fitting old lenses into modern cameras becoming more and more popular, perhaps manufacturers will start making lenses that are more optimized for the actual "look" than for achieving high ratings on sharpness tests. Perhaps this has already started, I think the modern Nikon 58/1.4 was engineered for nice rendering rather than sharpness.

So those 1950s and 1960s lenses have often very nice qualities. By the way, I believe that once Olympus brought its damn OM-1 machine with its dinky little lenses, lens design went to the trash -- because all the other manufacturers started focusing on reducing the lenses' size as priority #1, regardless of the impact on the ultimate performance on the lenses.

But that's another topic.

 

[johnnyrod]

... also the statement that a larger front element is used to collect more light to compensate for the light loss of the additional elements...

That'll be the comparison in pictures made between a f1.8 and a f1.4, so maybe gives me hope that my old f2.8s can be made to work a few stops faster, after all they're really getting on a bit now and have only three groups! I also enjoyed the bit about the light spinning into the glass.

 

[Ronald M]

Check out The Angry Photographer on YouTube. He calls people done with new glass iron faces as they look ironed down.

 

[giganova]

The combination of modern lenses + ultra-high res sensor cameras often produces images that IMHO have too much detail and distraction. Do I need to see every pimple and nose hair in a portrait? 😱

 

[flavio81]

Canon 85mm f/1.8 9 elements, 7 groups, 8 blades. Sony 85mm f1.4 11 elements 8 groups, 11 blades, rounded. Not so sure groups/elements is a factor. Her face is narrower in the Canon shot to my eyes. But have to see two different lenses on same camera same subject same lighting - the video link at the top does attempt this.

What's old, becomes "New Again" 🙂

It's because number of aperture blades is not a factor.
Neither is number of elements or groups.

"Good" blur or bokeh characteristics depends on the decisions of the lens designer. He/she may, for example, favor better sharpness over the whole picture area in detriment of bokeh and viceversa. Or may choose to go with a more expensive/complex design to do less compromises.

On the comparison of the Canon and Sony 85mm lenses you can see how different can be the character. The 85/1.8 Canon EF is a wonderful lens (i own it too.) But wide open is not sharp, just ok.

Usually modern buyers are overly obsessed with MTF curves and DxO tests, so lenses are designed accordingly. But at the same time there are lenses designed intentionally for nice rendering.

 

[michaelwj]

The combination of modern lenses + ultra-high res sensor cameras often produces images that IMHO have too much detail and distraction. Do I need to see every pimple and nose hair in a portrait? 😱

I remember some reviews of "The Hobbit", shot in 4k, were similarly unimpressed by the detail provided. Many said the amount of detail left nothing to the imagination, and it made it impossible to see that it wasn't shot on a set.

While I admit that sometimes more detail is better, I generally find that the photos I keep coming back to don't rely on detail and imaging "perfection".

 

[DanskDynamit]

Which basically surmises that modern, complex lens designs, whilst making technically better lenses also have the effect of 'flattening' the resulting images in perceived depth and micro contrast. He lists more complex designs, aspherical and ED elements as devices which can cause an unnatural 'plane-like' sharpness in photographs, whilst less complex optical formulas using plain glass tend to deliver better micro contrast and depth.

I personally have had a hunch regarding this theory for a while now, tending to have a strong preference for my photographs shot on older, simpler lenses as opposed to highly corrected modern ones - for example I have several personal favourite photographs shot on olympus stylus epics (4 elements in 4 groups), the Voigtlander 35mm f2.5 color-skopar, contax G 45mm f2, the old canon EF 35mm f2 which far outnumber favourite photographs taken with more expensive or better corrected lenses like my stint with m4/3 or canon L lenses.

The Oly Stylus Epic uses 4 elements in 4 groups with an aspherical element added.

 

[valdas]

I remember some reviews of "The Hobbit", shot in 4k, were similarly unimpressed by the detail provided. Many said the amount of detail left nothing to the imagination, and it made it impossible to see that it wasn't shot on a set.

When I went to see Hobbit I did not know it was shot in 4k. While watching it I was wondering all the time - what's wrong with it? Why do I see this "soap opera effect" that I always hated when TV screens appeared with all kind of motion smoothing features... Then, after the movie, I read it was 4k. That was soooo terrible 🙂

 

[Pioneer]

Interesting and entertaining. Particularly when I tapped into the Angry Photog stuff.

There has always been fringe groups around every endeavor. They used to stand on a street corner in New York City and entertain Helen. Now the government has given them computers and modems so you can find them on YouTube and other interesting websites, trying to entertain the entire world.

And great fun was had by all. Thanks for this entertaining interlude.

 

[nikonhswebmaster]

chips anybody?

Salsa or Guacamole?

 

[Jdi]

This is really complicating matters. Now I have to learn pseudoscience as well as dealing with all those mountains of scientific testing that never led anywhere? I think I'll continue my practice of selecting lenses based on the draw I prefer. Though I'm sure all of you are onto something.

 

[Addy101]

When I went to see Hobbit I did not know it was shot in 4k. While watching it I was wondering all the time - what's wrong with it? Why do I see this "soap opera effect" that I always hated when TV screens appeared with all kind of motion smoothing features... Then, after the movie, I read it was 4k. That was soooo terrible 🙂

It isn't the 4K bit, it is the 48 fps it was shot in. Films were 25 fps, cheap soaps were at 50/60 fps (depending on system) - The Hobbit was shot in the same way as those cheap soaps. :bang:

Thankfully I'm not into the Tolkien universe 😀

 

[valdas]

It isn't the 4K bit, it is the 48 fps it was shot in. Films were 25 fps, cheap soaps were at 50/60 fps (depending on system) - The Hobbit was shot in the same way as those cheap soaps. :bang:

Thankfully I'm not into the Tolkien universe 😀

Yes, you are absolutely right - it's fps stuff which does this soap opera trick... I don't know for what reason I just refereed to a previous post which mentioned 4k without even thinking...

 

[Alberti]

I think it is quite correct. Some remarks (slows the light) look silly to the average photographer but I think it is quite aptly said (it is the reason for bending). And plastic lens compound disperses more than glass, it is the large molecules, and that unevenly over the spectrum also.
Anyway, Marco Cavino remarks that the 'secret' of the sonnar is that the large last element deliberately has one over-corrected and one under-corrected element in one group.
Mandler has similar doublets in his last group in some designs.

I have had a Jupiter-3 and -8 and -9. All fantastic even sharp lenses with a nice "deep field" touch to them, and the pictures indeed get more WOW acclaim. But they are a pain on my M-camera's. To focus. So only really useful with EVF. It is with a bleeding heart I let go of these.

so yes that is where I sit, waiting for some lenses that behave properly.

 

[Rob-F]

I think it may be possible that light travels slower through glass than in free space. Radio waves, another form of electromagnetic radiation, travel about ten percent slower in transmission lines and antennas (wires) than in free space. Perhaps glass is the optical analog of that. But I can't imagine that glass can be the optical analog of a capacitor, which actually stores energy. Possibly a watch dial that glows in the dark (storing and re-releasing energy) after exposure to light is more arguably an "optical capacitor."

 

[Benjamin Marks]

Based on this thread. I tried flattening my lenses, but they then didn't work so good. I am now trying to reinflate them, but the pieces won't stay together. Particularly difficult are the Mandler-designed lenses that seem to take the impact of the "flattener" (actually, a Portugese-assembled aftermarket copy of a Midland designed sledgehammer) better than the modern glass that actually turns more "powdery" after the second hit. Oh, RFC, how could you do me so wrong?

 

[joe bosak]

I think it is quite correct. Some remarks (slows the light) look silly to the average photographer but I think it is quite aptly said (it is the reason for bending).

Yes that certainly used to be the standard physics explanation for what happens when light is refracted - how lenses used to work, in fact.