Low Element Count Lenses, 3D pop, and Micro-contrast -- My Theory

oldwino said:
And you do realize that a photograph is not real life?
Click to expand...

Hmm the photographs in my life seem pretty real. Sorry if that still sounds a bit condescending. I'm in the habit of trying to provoke people to elaborate on their reasoning by poking questions and pointing out flaws, which often doesn't go down well, understandably...

But I think my concern is valid, if we use the same eyes with their optical flaws to look at pictures, why would it help to incorporate these flaws in the pictures, even if the premise that they make things look nicer holds true? Because more of them is better or?
 
Last edited:
retinax said:
Hmm the photographs in my life seem pretty real. Sorry if that still sounds a bit condescending. I'm in the habit of trying to provoke people to elaborate on their reasoning by poking questions and pointing out flaws, which often doesn't go down well, understandably...

But I think my concern is valid, if we use the same eyes with their optical flaws to look at pictures, why would it help to incorporate these flaws in the pictures, even if the premise that they make things look nicer holds true? Because more of them is better or?
Click to expand...

Because, I think, that our flawed eyes “see” a representational view of the world, and certain lenses seem to “see” the world in the same way. So the resulting photo, a 2D version of our 3D world, is somehow more “lifelike”.

I’m not a huge Garry Winogrand fan, but his assertion that he photographs something to so how it looks like in a photograph is right on.
 
This thread is veering dangerously close to "what is art?".
I am still waiting for a near-consensus on which lens (that will fit my M2) (that I can afford) will POP.
 
Rather than a debate about what is art, this is veering more towards what is reality. What we see is not really what is there in front of us, it's simply how we view reality. With two eyes that see the external world at a different angle than the other, our brain has to put the two images together to come up with what we see. Fortunately, most of us see reality pretty much the same as someone else sees it, but not always. And this certainly doesn't mean that we see things as they are. Another type of being like a cat, dog, etc sees the world differently than we do. Which is more real than the other?

If we saw things as they really are, then objects would not get bigger and bigger as we get closer. That is impossible. The size remains constant, but in order to give us depth perception, our brain creates the illusion that things are getting bigger as we move closer to them. That doesn't really happen. There are many visual illusions that we take as fact because that's the way our brains have evolved. If they had evolved in a different manner, then what we see would look different than it does now.

NONE of what we see is actually there in the manner that we see it, so to think it's reality is a mistake. It's simply what we have agreed by consensus to call reality. Our human ears cannot hear certain sound frequencies, but that doesn't mean the sound isn't there. It's the same with our eyes. There are lots of things that are beyond the range of our optical capability, but that doesn't mean those things aren't there. And optical illusions are the clear perception of things that are NOT there. If you take a pencil, hold it in the middle with two fingers, and shake it up and down, it clearly appears to bend. That doesn't mean it really bends. If you photograph a spinning top with a high speed shutter it no longer blurs but appears to be clearly defined, while our eyes absolutely see a blurry, spinning image. Which is "real"? None of it!

Anyway, it's obvious to my eyes that many of the lenses I've owned that had lots of elements occasionally made photos that had lots of "pop" and a decided 3-D look if used in a manner that would give that effect. It just is what it is. Or from a different perspective, it clearly isn't what it is, it's simply what I see.
 
davidnewtonguitars said:
This thread is veering dangerously close to "what is art?".
I am still waiting for a near-consensus on which lens (that will fit my M2) (that I can afford) will POP.
Click to expand...

Where is no such thing. Cheap and with POP.
Where are some lenses with not flat rendering. No cheapies, but not ebay Leica prices.
Ultron 35 1.7, you could find white LTM under 500 and VM under 1000.
Biogon C 35. Well under 1000, white is available.
Summarit-M 35 2.5 slightly above 1000.
I have Summarit and very often after looking at prints from negatives taken with this lens I'm saying YES, which is in some languages POP-POP-POP.
 
Steve M. said:
Rather than a debate about what is art, this is veering more towards what is reality. What we see is not really what is there in front of us, it's simply how we view reality. With two eyes that see the external world at a different angle than the other, our brain has to put the two images together to come up with what we see. Fortunately, most of us see reality pretty much the same as someone else sees it, but not always. And this certainly doesn't mean that we see things as they are. Another type of being like a cat, dog, etc sees the world differently than we do. Which is more real than the other?

If we saw things as they really are, then objects would not get bigger and bigger as we get closer. That is impossible. The size remains constant, but in order to give us depth perception, our brain creates the illusion that things are getting bigger as we move closer to them. That doesn't really happen. There are many visual illusions that we take as fact because that's the way our brains have evolved. If they had evolved in a different manner, then what we see would look different than it does now.

NONE of what we see is actually there in the manner that we see it, so to think it's reality is a mistake. It's simply what we have agreed by consensus to call reality. Our human ears cannot hear certain sound frequencies, but that doesn't mean the sound isn't there. It's the same with our eyes. There are lots of things that are beyond the range of our optical capability, but that doesn't mean those things aren't there. And optical illusions are the clear perception of things that are NOT there. If you take a pencil, hold it in the middle with two fingers, and shake it up and down, it clearly appears to bend. That doesn't mean it really bends. If you photograph a spinning top with a high speed shutter it no longer blurs but appears to be clearly defined, while our eyes absolutely see a blurry, spinning image. Which is "real"? None of it!

Anyway, it's obvious to my eyes that many of the lenses I've owned that had lots of elements occasionally made photos that had lots of "pop" and a decided 3-D look if used in a manner that would give that effect. It just is what it is. Or from a different perspective, it clearly isn't what it is, it's simply what I see.
Click to expand...

Not going into the philosophical discussion deeper for now, although it's interesting, but the issue of perspective I think you got wrong. Think of the camera obscura, it too represents further objects smaller. That's not in our brains. It's a consequence of projecting three dimensions onto a plane. I don't think there's an alternative.
Edit: P.S. I nearly got my brain in a knot trying to figure this out when I was a kid. If you know the excellent German children's book Jim Button, you know why.
 
Sticking with the Nikkor 50mm f/1.4 AF-D. Classic 7 element, 6 group 50/1/4. Nothing more (or is there?). Nothing less. Light, well-made. Auto focuses. Nikon coating. Current production lens for 23 years now, introduced in 1995. There's a reason for that. Incidentally, its DXO score is the same as the newest Nikon G 50/1.4.

Let's go over to DXO and compare it to the similarly spec'd Zeiss Milvus:

Milvus
Price: $1200
Score: 35
Sharpness: 20 P-Mp
Distortion: .4%
Vignetting: -1.6
CA: .8 um
TRANSMISSION: 1.7

Nikon
Price $386 (but less than $200 minty used)
Score: 32
Sharpness: 16
Distortion: .4%
Vignetting: -1.4
CA: .9 um
TRANSMISSION 1.4

The newish $1200 Zeiss is sharper (at 1.4 mainly, and in the corners and borders nobody cares about, I'm sure) and vignettes less (does this even matter s'much in the digital era, really?)

However, the Zeiss has a t-stop value of 1.7. Like the Sigma Art 1.4, this is a horrid transmission score, while Nikon's is perfect 1.4 in, 1.4 out. No loss of light -- and the information that light is carrying, via I/O processing. The Milvus is another overcorrected 10 element lens that only transmits 82% of the light it captures to the sensor, "losing" 18% of the information along the way. Another lens to appease the DPReview MTF chart gods that sacrifices overall fidelity to win the sharpness wars, which is the new megapixel wars for lenses.

Plus none of the Zeiss lenses even autofocuses. (Perish the thought, right?) Welp I like autofocus. It's faster and more convenient. Must be nice for the Zeiss (and Leica) engineers not having to worry about such things as cramming gears, motors, electronics and firmware into a lens and make a MODERN fast/accurate objective that actually can autofocus. Imagine what the "Milvus" would cost if they did! But they better watch out. Samyang is putting out some pretty nice manual focus lenses these days for 1/2 the Zeiss price.

Now let's compare the lowly mass-produced (for 23 years) Nikon to the Leica Summilux M. DXO doesn't have a rating for that lens, so we head over to Photodo:

Nikon Photodo MTF score: 4.2
Leica Summilux M MTF score: 4.2

A tie! Did I mention the Nikon has autofocus? (But the Leica does have that "buttery smooth" manual focus ring, an engineering marvel! So there's that...)

Finally, fwiw, take it with a grain, enter the search term "lenses with best microcontrast" and you'll get a box straight-away, from which this is cut-n-pasted:

Nikkor AF 50mm 1.4D.
Nikkor AF 35mm 2D.
Voigtlander SLII 58mm 1.4 Nokton.
Zeiss ZF2 35mm 2.0 Distagon.
Nikkor Ai-S 50mm 1.4.
Voigtlander SLII 58mm 1.4 Nokton.
Nikkor Series E 135mm 2.8.

Obviously others are seeing what I am seeing.

Oh -- and nobody really cares about "bokeh" or how it looks. Bokeh is to be ignored -- that's the whole point of it, and I've never observed bokeh, even the allegedly "worst" based on internet bokeh raters to be "distracting" or "bad". Any cheap telephoto lens shot zoomed out with the subject close in will produce "amazing" bokeh better than any Zeiss or Leica 50. Shoot with a cheap Tamron zoom if that's your thing.
 
NickTrop said:
However, the Zeiss has a t-stop value of 1.7. Like the Sigma Art 1.4, this is a horrid transmission score, while Nikon's is perfect 1.4 in, 1.4 out. No loss of light
Click to expand...

You are claiming the 50mm f1.4d is the best based on one specification and then using a spec that is clearly wrong for the 1.4d.

What you continue to claim about the light transmission in the Nikon lens is not possible, it reflects light. And you continue to ignore that. As well as the tested data that shows the 50 f1.4d has about 91% light transmission. Not that that spec means anything like what you think it does.

If you want to use data to try and 'prove' your hypothesis you can't pick and choose to ignore data that disagrees with your hypothesis.

There are plenty of comparisons between Nikon 50mm available online. And they do much more than base this on a couple of specs. For example:

https://medium.com/vantage/comparing-nikon-s-cheap-50mm-prime-lenses-653bb2b46c2c

https://nikonrumors.com/2011/11/03/...r-nikon-f-mount-compared-by-cary-jordan.aspx/

Shawn
 
shawn said:
You are claiming the 50mm f1.4d is the best based on one specification and then using a spec that is clearly wrong for the 1.4d.
Click to expand...

Is it wrong? The new Nikon G series 50/1.4 is rated at 1.5 t-stops with one more glass element (8) for the light to pass through. Perhaps this is a spec Nikon focuses on, while the others are trying to outdo each other in meaningless (beyond a certain point) resolution tests? Did DXO screw up twice? Or is there confirmation bias and cognitive dissonance happening here? I say the D-series 1.4 has a very high transmission score and that 1.4 = 1.4 due to rounding. Yes, there are surface-to-air regions but it likely loses a little less than 0.10, hence the 1.4 spec. Now that "Thom" guy doesn't like this particular Nikon s' much, complaining about a center hot spot "especially shooting IR film" at f16 that I've never seen and never shoot at and haven't shot IR film in 15 years. But he's a zoom guy anyway. What's he know from primes?

Do note that I do not cite quantitative scores as "be all and end all" but am trying to reduce subjectivity and make it as apples-to-apples as possible.

Basically I regard Zeiss and Leica glass as classic Veblen goods:

"Veblen goods are types of luxury goods for which the quantity demanded increases as the price increases, an apparent contradiction of the law of demand. Consumers actually prefer more of the good as its price rises, and the result is an upward sloping demand curve." -- Wikipedia entry

If Veblen goods are your thing? Go for it!
 
F-stop is a rounded off marketing tool. So a lens with an advertised f-stop of 1.4 might actually be 1.5 or 1.6, at which point a t-stop of 1.7 is not that much different.
All of this is also dependent on how much of the frame is used in the t-stop measurements. Or, to put it another way, do you know how much vignetting is taken into the t-stop measure? One stop of vignetting could easily lead to a half stop reduction in transmission over the whole frame.
When you stop the Nikkor down to f/2 the t-stop value drops to around 2.0! That’s right, you lose half of all that pop information carried by the light. Even worse is the 70-200/2.8 Nikkor wide open doesn’t even have a t-stop value of 2.0! That’s why it sucks. :D
 
A picture is worth a thousand words. Not many pictures here.

When intellectually discussing "something" you have to define what you are discussing. No one is saying what "Pop" is. Saying "It looks more 3D...." isn't saying anything.

Here is what I'd like. Someone set up a STILL LIFE, and show me how one lens exhibits more "pop" than another. Both with the exact same subject, lighting, focal length. Then we can dissect the proverbial "pop", if there really is any with any lens.

I watched the Cooke vs Leica cinema vid, and I still can't see it. When the speaker says "see how more rounded the subject looks....see how the wall in the back is flatter...blah" What? Where? Am I the only one that wants to say The King Has no Clothes!?
 
michaelwj said:
F-stop is a rounded off marketing tool.
Click to expand...

No. It's not. It's a standard unit of measure that describes the light-gathering ability of a lens for subjects focused at infinity, though the "infinity" part of the equation is largely ignored today. It's no more a "marketing tool" than shutter speed. It was defined in 1867 by Sutton and Dawson. It is calculated thusly:

N= F/D

Where F = focal length, and D is the diameter of the aperture at its widest.

I bet it's rounded because if this division results in an irrational number, the size of lens barrels would become quite unwieldy or the print so small so as to be illegible. By the way, your eye has an f-stop range of 2.1 to 8.3.
 
goamules said:
A picture is worth a thousand words. Not many pictures here.

When intellectually discussing "something" you have to define what you are discussing. No one is saying what "Pop" is. Saying "It looks more 3D...." isn't saying anything.

Here is what I'd like. Someone set up a STILL LIFE, and show me how one lens exhibits more "pop" than another. Both with the exact same subject, lighting, focal length. Then we can dissect the proverbial "pop", if there really is any with any lens.

I watched the Cooke vs Leica cinema vid, and I still can't see it. When the speaker says "see how more rounded the subject looks....see how the wall in the back is flatter...blah" What? Where? Am I the only one that wants to say The King Has no Clothes!?
Click to expand...

I too would like to see more examples. My mind is mostly made up at this point, and this isn't any kind of mysterious phenomenon, but a simple property of lenses. I'd love to have my mind changed.

I couldn't see anything in the Cooke vs Leica video that would make me want to spend money to buy the more expensive lens. I watched the video on a connection that is not very fast, so I'm sure compression removed some of the distinctions. This kind of hairsplitting just bothers me sometimes. I see the differences, but the preference of one over the other is often faddish.
 
NickTrop said:
No. It's not. It's a standard unit of measure that describes the light-gathering ability of a lens for subjects focused at infinity, though the "infinity" part of the equation is largely ignored today. It's no more a "marketing tool" than shutter speed. It was defined in 1867 by Sutton and Dawson. It is calculated thusly:

N= F/D

Where F = focal length, and D is the diameter of the aperture at its widest.

I bet it's rounded because if this division results in an irrational number, the size of lens barrels would become quite unwieldy or the print so small so as to be illegible. By the way, your eye has an f-stop range of 2.1 to 8.3.
Click to expand...

So you think a f1.45 lens wouldn't be sold as an f1.4 lens? You know that a 50mm lens doesn't have a 50mm focal length as well. Likewise, until modern electronic shutters, 1/500s wasn't actually that fast.

There's a reason that T stops are used in cinematography. It's because they're an actual measurement of the light transmissibility of the lens, which historically has been incredibly important for a medium which consists of spliced segments of film cut from many scenes. They can't rely on a manufacturer's statement that the lens is f1.4 to mean anything.

Fortunately, for still photography, with the latitude available in film and digital sensors, a few percent fudge factor on the largest aperture doesn't matter.
 
What's to stop Nikon from producing a lens with an f stop of 1.2-1.3 or so, and saying it's a 1.4 so the T stop is 1.4? Without technically taking apart a lens who is to know. I sincerely doubt that Nikon's lens is f1.4 with a T stop of 1.4...

Joe
 
You must log in or register to reply here.
Back
Top Bottom