Discussion in 'Open Discussion' started by Ranger Rick, Mar 12, 2016.
Writing headline made me feel like I was on the Leica blog
The Problem with Modern Optics
Although I dont get all the technical subtleties I do find it interesting that according to the diagram, two lenses that are very often quoted for having a very special rendering i.e. the pana 20mm and the pana 25 1.4 they both rank in the lenses that record life as it is. MAybe this is why a lot of people find these 2 lenses so intriguing? (Maybe I should take another look at the 17 2.8
Is this the same stupid blog recommending older lenses with less glass elements? because adding more glass elements removes the 3d effect from lenses?
It's pure bullshit, you know that right?
Believe that is the one- care to elaborate on why it is bullshit - not saying it isnt just would like to understand why
Because it's unscientific nonsense that keeps getting regurgitated that doesn't have a basis in fact. It's incredibly hard to tell lenses of different design apart even when you know which one is which, make it a blind test and it's just guessing.
The element count and design makes very little difference on final rendering, coatings are the main factor in older lenses which has an effect on their rendering (this is why you can buy modern glass with single coating (Hawk would be the best known), or uncoated (Nikon/Zeiss will do it as a special order item if you ask nicely)). The reason modern lenses moved away from older coatings is their transmission was much lower (an f1.4 lens with a t2.0 is the reality of many older lenses). Thus the main usage for wide aperture (low light) is not forfilled by older lens coatings, that's why the 'rendering' changed... not some bullshit to do with element count.
I'm all for using lenses which produce pictures that talk to you, just don't justify it with pseudoscience. Likewise don't pretend that you can tell modern lenses apart in a blind test (you really can't).
I've gotta say, everytime I hear about "3d rendering" I think about audiophiles waving their hands talking about sound warmth and other nonsense. The pictures in that post are utterly unconvincing since they are different people in different light at different distances...crikey. I realize objective tests of this are difficult but at least make an attempt.
I am surprised Phoblographer republished this. I am not a lens designer, and I haven't taken the time to test or refute the conclusion, but the evidence presented is...insanity. The mad triangular charts, the "look at how flat/3D" captions that (a) don't appear to accurate describe anything, and (b) ignore the difference in subject and lighting between shots. There is nothing here but a theory and a bunch of confirmation bias to support it. And I suspect the author did not test all of the lenses charted for "realism" and just charted them based on lens design specs.
I didn't think much when this stuff was published on the guy's personal blog, but I am surprised to see it rebroadcasted (and seemingly more insane with the diagrams).
Personally, I'd like to have a lens that gave nice 3d effects to portraits. I use a CV 25 for most of my portraits, but haven't noticed a 3D effect. Does anyone have a mft lens that gives a 3D look?
The photos in the article shows bad examples of portraits - but I get similar effects when I use a wide angle lens up close, and you will get distortion.
There's the old trick of shifting the focus point, shoot at around f4-11 (larger aperture at increasing distance to subject, you want to balance the DoF so that the background will be sharp but not so much that the subject comes completely into focus) using a longer lens (45mm), have the subject at a reasonable distance (3-4m) -> focus on the subject -> move the focus ring a small amount to push the focus point backwards to behind the subject.
It will give a picture where everything appears to be in focus (subject, background, but not foreground) but the subject will be slightly softened due to you moving the focus point backwards to give smoother skin (depth of field is shallower in front of focus point than behind). The subject won't be ruined since you stopped down significantly and the background should appear sharp (even at 100%).
This will make the subject 'pop' out of the background as they're the only thing which isn't completely sharp, your eye will catch it as wrong (it's out of focus) however it won't be out of focus so much that you can actually pick it at closer inspection. It only works stopped down significantly as with shallow DoF you will be able to pick the plane of focus, only works on subjects where you wish to keep the background, If you go too far the subject will go out of focus too much and ruin the shot (once you can pick that it's out of focus, it's too far), YMMV.
Personally I'm not a huge fan of shallow depth of field and prefer to use other tools if I can.
The article is pseudo science nonsense. Ignore it.
Can you share them? I've tried a really shallow dof with the eyes being in focus, but it's hard to nail when the dof is narrow. If there's other ways to make a portrait pop, I'd give it a try
The article seemed to make sense until I looked at the photographs chosen to illustrate this thesis.
The flat nose picture was taken in a flat diffused light whilst the 3D noses were taken under sunny contrast conditions.
If one has to cherry pick the illustrations in such a manner then something becomes a little bit fishy about the reasoning.
I'm pretty skeptical about the claims, but I am interested in ways to make a portrait pop. Eteless provided a technique I didn't know about, and perhaps others can share examples and techniques they may have to refute this article - using mft cameras of course.
I really can't tell much from what he says is 3D vs. flat. I do like my 17 2.8 however.
In a side by side comparison, i saw some differences. Hard to say which i liked better.
He said some reasonable things, like the in my eyes unnecessary focus on shapness in the edges and corners.
I see some photos (especially portraits in the oly 75mm 1.8 section) that i find too sharp. There i see people rendered in a way i dont recognize in the real world. But, does that mean that those photos are flat? I never noticed that.
Maybe i dont look the way i should. I shoot with both old SMC Pentaxes as a sigma 19mm and an oly 12-40 and i dont perceive any difference in image quality.
But then again i dont find all those parameters not THAT important. For me a photo has succeeded if it represents the scene as i saw it.
I dont pretend to be an artist. Just a guy who finds pleasure in going out and take some shots.
The idea that nothing interesting is happening "most of the time" at the edges of the frame is stupid. If you take a full body portrait picture, you need sharpness across the whole picture.
If you are a hack who only takes "artsy" headshots with super bokeh then yes I guess it doesn't matter how sharp the edges are
Those triangle diagrams remind me of the pseudo-scientific babble we sometimes get presented with at work, on the latest initiative in how to run a business. Invariably someone (we're all engineers) will put their hand up and say something like "what's the scale on that diagram?" "How have you defined that quantity?" etc. To which there's no answer cos it's all made up nonsense...
Sort of silly to flatly claim there's nothing important that needs to be sharp at the edges or in the corners. Think of a straight on shot of a building, wall or other vertical plane, and you'll most likely want sharpness everywhere.
"3d" rendering of faces and figures is more a relationship of the subject to camera distance, subject to background distance, and especially the lighting. As well as the angle of the face and nose to the camera. It's really not in the optics folks.
Unnecessary sharpness is necessary when your shooting landscapes. It looks to me he only shoots portraits. Frankly I think its nonesence.
I must agree with this. I am an audiophile but not the sort that thinks modern=soulless, old=charismatic. And that is exactly what this guy does.
I understand the sentiment, but I actually think this paradigm is related to one's personal value set. Most people don't think about their value set as something that strongly influences what they think is true and untrue, good and bad. They just think "this is true, that is bad, etc" and then proceed to build arguments, often very rational arguments but nevertheless built on a starting point that isn't neutral, and that important point is invisible to the prosecutor of the argument.
There is a very substantially held and common human value set where people are not comfortable with the modern world. Its complexity can be overwhelming and can make one feel stupid. Its complex products can have the same effect. In the end, simplicity itself becomes a value, and espoused as a philosophy. When you read things like 'less lens elements make for a livelier look, more elements make it flat and lifeless', the proponent of that view is literally rebelling against modern technology getting between his eye and his experience of the world.
Separate names with a comma.