fredmiranda.com Login
This topic is in the 2024 archive. You can read it and like its posts here. It is read only.

Archive 2024ED elements , lens design and 3D pop

  
 
philip_pj
Offline
Details
p.2 #1ED elements , lens design and 3D pop
Start here for some excellent examples of 3D:

https://www.fredmiranda.com/forum/topic/1389884/77/

What do the longer Loxias (35-50-85) have in common?

Low element counts (9-6-7),
no aspherical surfaces,
traditional designs,
20th century heritage,
slow maximum apertures,
T* coatings,
lead.

YT title: 'Common LEAD as the 'magic' ingredient in the best lenses of Voigtlander & Zeiss'

https://www.cloudynights.com/topic/119827-characteristics-of-leaded-vs-unleaded-glasses/

Same as the better Contax-Yashica lenses, which also show the same image depth.

All wide angles are, by design, good at dimensionality - image motifs are longitudinally separated. APD glass is fine if handled with extreme care, as is done by Zeiss and Cosina.
Nov 07, 2024 at 08:49 PM
Garmadon
Offline
Details
p.2 #2ED elements , lens design and 3D pop
So the less ASPH glass -> more pop?
And how can you tell if a lens has lead in it ? does the batis line /sony zeiss line has it?

philip_pj wrote:
Start here for some excellent examples of 3D:

https://www.fredmiranda.com/forum/topic/1389884/77/

What do the longer Loxias (35-50-85) have in common?

Low element counts (9-6-7),
no aspherical surfaces,
traditional designs,
20th century heritage,
slow maximum apertures,
T* coatings,
lead.

YT title: 'Common LEAD as the 'magic' ingredient in the best lenses of Voigtlander & Zeiss'

https://www.cloudynights.com/topic/119827-characteristics-of-leaded-vs-unleaded-glasses/

Same as the better Contax-Yashica lenses, which also show the same image depth.

All wide angles are, by design, good at dimensionality - image motifs are longitudinally separated. APD glass is fine if handled with extreme care, as is done by Zeiss and Cosina.

Nov 07, 2024 at 10:09 PM
philip_pj
Offline
Details
p.2 #3ED elements , lens design and 3D pop
Maybe indirectly. Aspherical surfacing is generally seen as an unalloyed good, hence all the glowing blurbs at B&H. The asph elements absolutely anchor the designs of zooms of high quality. I've long believed makers (Leica etc.) who embraced them in virtually every high end prime and zoom make some lenses with the flat look - often seen in the blur wall behind the sharp planar subject. Images are so 'correct' many are now not using profiles.

But it remains a postulate for now, being suspicious I discuss it a bit here and there. My 100 year old design Heliar 50/3.5 produces much more depth and life in landscapes than the vaunted 35/50 APOs, as do my Contax Zeiss lenses.

There are essentially two types of what you and the video guy call 'pop' (image depth works better for me):

. a very sharp subject with high separation from the blur wall; and
. mid aperture high DOF images.

I want the second of these, the first is getting to be commonplace.
Nov 08, 2024 at 12:37 AM
philip_pj
Offline
Details
p.2 #4ED elements , lens design and 3D pop
Lead, a fascinating topic with precious little information. The EU coerced lens makers to remove lead, even though there is one exception made: optical glass. Apparently some lead isotopes are incredibly valuable for high tech glass applications (satellites etc.).

"Just for information, the use of lead (oxide) in glass for optical application is still allowed within the European Union. This exemption is mentioned in Annex III, 13a, of the European Union Directive 2011/65/EU, which regulates the lead content in various devices."

https://www.l-camera-forum.com/topic/323209-material-used-in-vintage-r-lenses/

The angry photographer (link above), Ken Wheeler I think his name is, has some controversial views on the subject and he is very knowledgeable on it. I believe it was enacted to remove it in 2007 or thereabouts. I imagine all have developed other methods to try to do what leaded glass did so well. I'd like to hear more from Ken.

"Lead oxide added to the molten glass gives lead crystal a much higher index of refraction than normal glass, and consequently much greater "sparkle" by increasing specular reflection and the range of angles of total internal reflection. Ordinary glass has a refractive index of n = 1.5; the addition of lead produces an index of refraction of up to 1.7.[1] This higher refractive index also raises the correlated dispersion, the degree to which the glass separates light into its colors, as in a prism. The increases in refractive index and dispersion significantly increase the amount of reflected light and thus the "fire" in the glass.

In cut glass, which has been hand- or machine-cut with facets, the presence of lead also makes the glass softer and easier to cut. Crystal can consist of up to 35% lead, at which point it has the most sparkle."

https://en.wikipedia.org/wiki/Lead_glass

'Glasses with lead oxide content first appeared in Mesopotamia, the birthplace of the glass industry.[4 in above link] The earliest known example is a blue glass fragment from Nippur dated to 1400 BC containing 3.66% PbO. Glass is mentioned in clay tablets from the reign of Assurbanipal (668–631 BC), and a recipe for lead glaze appears in a Babylonian tablet of 1700 BC.'

It's hard to find out anything much from the industry. As we agree, and as Tim said, many factors are involved. Many will use both modern high end and more 'vintage' lenses. We must hope to know more as time goes by.
Nov 08, 2024 at 12:37 AM
Kalainen
Offline
Details
p.2 #5ED elements , lens design and 3D pop
I have been photographing for over ten years, actively following forums, and have probably seen every possible discussion topic. I’m going to be a bit controversial here, but in my opinion, 3DPop is one of the most pointless topics in photography. It quickly leads many enthusiasts into a pseudoscientific swamp, where they might get lost for years contemplating the esoteric properties of lenses and their gear choices – all of this is off from real photography and developing as a photographer.

Over the years, while following these discussions, I have learned that where there is an active conversation about 3DPop, there is rarely any truly meaningful content regarding photography. The participants are often beginner enthusiasts (with a couple of years of photography experience and some technical knowledge under their belt) who mistakenly think they have discovered a profound and significant aspect of photography. They think they have found a secret source of knowledge that will make their photography captivating. The other half of the debaters usually consists of the gurus of the forums who maintain this belief and perhaps gain some kind of attention and influence from these discussions. I have to add that the same photographic shallowness can also be seen in image-threads where people try to find this 3DPop, you can see tree trunks, fire hydrants and other everyday items, but rarely anything photographically meaningful.

This leads to the fact that as soon as someone brings up 3DPop, I can usually immediately conclude that the speaker is often more or less a beginner in photography (affected by the forum discussions) – and the conversation will not lead to anything photographically meaningful (guy on the starting post’s video as an perfect example). People are driven forward by this subconscious idea that the right gear leads to better pictures and more refined photography. It seems to me that the fascination regarding this subject is often related to ‘a sense of sophistication’. We want to be sophisticated and discerning in our photography, and the discussion about the subtle 3DPop effect, ED elements, lead, element groups etc., gives the impression of highly polished sophistication level. It's just never shows in the pictures as people circle around endless test pictures.

Instead of looking for these exceptional lenses, it would be much more useful and constructive to focus on the things that really affect and develop photography: knowing your subject matter, being aware about the message and context you want to have, having a some sort of interesting perspective on the subject, being able to tell a story, knowing the historic photographers and their work, etc.
Someone might argue that 'technical things and *the rendering* matters', yes of course, but to what extent? When it comes to technical things, you basically need to have three things right: light, composition and moment. There is so much more to learn about the photographic message and the context. 3DPop brings nothing to this table, and nobody in the real audience is interested in esoteric 3DPop. Only hobbyists who have lost their way into this swamp are interested in it (the topic is completely comparable to the high-end audio hobby, which is of course not wrong, but nevertheless doesn’t lead one to anywhere).

I know this was a bit controversial post regarding the thread, and I am not referring to any of you, since I know some of you and your backgrounds and have all respect for you and your work, but I am describing the nature of these 3DPop discussions on a general level.

And not anyone has to agree with me, we will still be friends and I have respect for others. But for those, who happen to find themselves here and are a bit unsure if they should investigate this more in their photography, my post is intended to offer 'an exit' of this discourse. One can do so much more useful things to develop one's photography than speculate about the 3DPop.
Nov 08, 2024 at 07:51 AM
gdanmitchell
Offline
Details
p.2 #6ED elements , lens design and 3D pop
^^^
Yes, basically. :-)
Nov 08, 2024 at 08:17 AM
Garmadon
Offline
Details
p.2 #7ED elements , lens design and 3D pop
Well, ofcourse it is a rabbithole ,and ofcourse a real pro will get the job done with a simple point and shot . Yet, most of us here use expansive gear or at least gear that they put a lot of
thought into.
I dont see why comparing lenses at 500% crops or look deeply into a single bokeh ball are more vital then trying to understand how lens elements effect the final image.

Now , Im not looking for a magic lens , I already have one , the 55mm f1.8. It always delivers. Since getting it 5 years ago ,ai tested more then 20 other lenses , some were sharper,with better bokeh ,faster, but this one has this special look that I really like .
Now , I look at this video and my lens has the exact same attributes,so maybe there is somthing to learn here ,for me at least.
Maybe I should check other lenses with similar construction ?
Kalainen wrote:
I have been photographing for over ten years, actively following forums, and have probably seen every possible discussion topic. I’m going to be a bit controversial here, but in my opinion, 3DPop is one of the most pointless topics in photography. It quickly leads many enthusiasts into a pseudoscientific swamp, where they might get lost for years contemplating the esoteric properties of lenses and their gear choices – all of this is off from real photography and developing as a photographer.

Over the years, while following these discussions, I have learned that where there is an active conversation about 3DPop, there
...Show more →
Nov 08, 2024 at 08:45 AM
mojoh
Offline
Details
p.2 #8ED elements , lens design and 3D pop
Can 3D pop be created instead using software? Like some plug-in or mobile phone portrait mode?
Nov 08, 2024 at 10:06 AM
tschopp
Offline
Details
p.2 #9ED elements , lens design and 3D pop
philip_pj wrote:
There are essentially two types of what you and the video guy call 'pop' (image depth works better for me):

. a very sharp subject with high separation from the blur wall; and
. mid aperture high DOF images.

I want the second of these, the first is getting to be commonplace.


Part 2 of a discussion on pop below.

The takeaway I got from the videos about different features that generate pop.

1.High contrast, rapid bokeh falloff. My interpretation (maybe incorrect) would be high element fast modern primes.

2. A combination of vignette, distortion, and field curvature. I would place these in the low element count, low special glass group.

I also like the discussion of good pop and bad pop from the first video. With bad pop often looking like the subject was photographed in a studio with an image as a backdrop. My daughter has commented on this as some primes can produce an image where the subject looks like they were poorly photoshopped into an image. The good pop still has the separation, but it is clear they are part of the overall scene.



Nov 08, 2024 at 10:57 AM
philip_pj
Offline
Details
p.2 #10ED elements , lens design and 3D pop
Someone called this the 'green screen' effect, what your daughter saw. And much of the effects we now see are learned over time, and they are not consumer-led. Lens development is moving along quickly compared with 30 years ago.

Your #2 is what I think many cine lens makers are doing, to obtain the elusive image depth. Zeiss have been doing vignette and acceptable distortion for decades, they never wavered despite a lot of review push back. We see it now in Voigtlander reviews.

More needs to be said: even as we look for decision trees and reasons for the flat look and what exactly it is, along comes some lenses that break with what we thought. The YT author dismisses all zooms and I understand why; but the new Sigma 28-45mm, which fairly bristles with asph and APD, does a really fine job of image making at all functional apertures.

It's a 'back to the future' lens nevertheless, being 45/28 = a zoom range ratio of just 1.6 - a courageous move by Sigma. The very 3D older zooms were 35-70, so 2.0 zoom ratio and they were slow at f3.5-f4. Slow aperture lenses are just better behaved for 3D, for maybe one reason they need less exotic glass. Zooms look to have gone too far, in the commercial ones at least - Canon's 28-70/2 - nineteen elements are needed to make it work.

Then the character lenses, like Nikon's 35/1.4 (two asph in 11 elements), 40/2 (two asph, six elements) and 58/1.4 (also two asph, nine elements) and many Voigtlanders. The video maker mentions Sony's FE 55/1.8 in one video, it is also a simple lens with seven elements but with five asph surfaces.

So maybe it is easier to avoid flatness with asph elements inside a relatively simple design? The RX1 Sonnar is of this kind also, five asph surfaces in an eight element lens. No one ever claimed these two from Sony were APO, but they are popular nevertheless.

Many character lenses are anything but flat and they are very popular with a subgroup of enthusiasts. I went off exploring the asph tangent because the new asph lenses really bend light in an extreme fashion ( as they do in phones), in order to extract ever more sharpness.

But that might be a very recent development, gained from phone technology. It might be that Canon factored this in when they chose the wide mouth lens mount, as their PMO resin lenses crank the light refraction hard, out wide in the frame near the rear of the lens.
Nov 08, 2024 at 02:41 PM
gdanmitchell
Offline
Details
p.2 #11ED elements , lens design and 3D pop
tschopp wrote:
1.High contrast, rapid bokeh falloff. My interpretation (maybe incorrect) would be high element fast modern primes.


Contrast can be an element that makes images seem sharper and can be used to make certain elements stand out. The problems here are several though. First, the difference in contrast among most lenses (leaving out really poor ones of course) is not really all that great — certainly much less than the contrast control you get in your post processing apps. Second, it isn’t just a matter of increasing contrast in the image, though sometimes that can help. For the best effect you have to be selective about where in the image you do this — you may want to do it on your primary subject (if it isn’t already too contrasts) but NOT on secondary, background elements. And you also may want to do this not with the contrast control, but rather by manipulating curve.

2. A combination of vignette, distortion, and field curvature. I would place these in the low element count, low special glass group.

Vignetting can help to draw attention to what is in the center of the frame, and it is often used in this way. (Sometimes over-used, but I digress…) But it is a double-edged sword in lenses. Sometimes the amount and form of it in a particular lens are just what you want, but other times it isn’t — it can be too dark, and it can create dynamic range issues in some images, there can be color shifts on digital sensors with very low contrast subjects. When necessary, introducing vignetting in post gives you greater control and actually better IQ.

Not sure what you refer to with “distortion.” Is this image geometry distortion? Something else? Not quite seeing how that gets you the effect you want.

Similar with field curvature, which is a sort of distortion, too, in that it may make areas in the center of the frame sharp while leaving edge/corner areas soft. This is also another double-edged sword — do you always want your corners to be soft? If so, and you can find a lens that does this exactly as you want, go for it. Otherwise — again — it is easy and effective to do this in post. (Invert a radial selection and do whatever you prefer to do to reduce sharpness at the edges.)

In the end, if you are looking to produce the effects that you characterize as pop, I’m positive that the most fruitful things to investigate are not different lenses but different photographing and post-processing techniques.



Edited on Nov 10, 2024 at 08:35 AM · View previous versions
Nov 08, 2024 at 02:45 PM
tschopp
Offline
Details
p.2 #12ED elements , lens design and 3D pop
gdanmitchell wrote:
Not sure what you refer to with “distortion.” Is this image geometry distortion? Something else? Not quite seeing how that gets you the effect you want.

In the end, if you are looking to produce the effects that you characterize as pop, I’m positive that the most fruitful things to investigate are not different lenses but different photographing and post-processing techniques.



The video was referencing geometric distortion. I'm guessing barrel distortion would help with the ilusion the subject is further from the background. I guess the key to vignette, distortion and field curvature is they all change the relationship between the subject and the background.

I'm not looking for pop myself, just found the videos interesting. I would agree, as does the video, pop can be more of a distraction. I've seen some stuff about how simulated perspective control can induce pop in post.

I really don't know much about the various classic lens designs or what strengths and weaknesses they have.
Nov 08, 2024 at 03:37 PM
DWOfPaul
Offline
Details
p.2 #13ED elements , lens design and 3D pop
Something I definitely agree with is that different lenses, especially from different manufacturers, have different rendering styles. Some of us notice it more than others. Once you notice these differences, you often want to figure out what causes the difference so you can get the most out of your gear. Unfortunately, we don't seem to have many optic experts in the photography community online so people make their best guess at what might be going on and some of those guesses are much more sound than others.

Nov 08, 2024 at 03:49 PM
white
Offline
Details
p.2 #14ED elements , lens design and 3D pop
As I think I have said before somewhere, 3D pop is a slider in LR.

Too much optical naval gazing.
Nov 08, 2024 at 04:05 PM
MARKFER
Offline
Details
p.2 #15ED elements , lens design and 3D pop
Just curious, what are the lens qualities outside of focal length, that pp can't or won't in the future just be a slide away? When it gets to the point will photography be better for it?
Nov 08, 2024 at 04:13 PM
DWOfPaul
Offline
Details
p.2 #16ED elements , lens design and 3D pop
In theory, if we get back to light field cameras at some point, almost everything outside of focal length can be manipulated in software. (At a high level, light field cameras capture the rays of light as opposed to the final pixels so the depth of field information is fully preserved. You can have a perfectly in focus image from front to back or a very shallow depth of field, all applied in software and without the artifacts most DOF simulators have today)

Now if we are talking AI programs that try to enhance images and simulate things like DOF, I am sure every year they will get better. Optics also get better every year though, and my gut is AI programs will take longer than most people realize to catch up with optics, unless we get some major breakthroughs in AI.

For example, AI denoising has pretty much become the gold standard for image denoising, but we haven't seen AI image denoising get better by leaps and bounds in the last few years. Personally, I feel like I have witnessed trade offs from version to version. Some of the new AI denoising programs work better on some images and work worse on some images. Not to get too techy, but this is a base problem in a lot of AI solutions. Make a different model and it often gets better at some things and worse at others. Then, there is also the computing resource issue that has diminishing returns. Create a model that requires twice the resources, and you may only get a 10% bump in performance. Just like a $4,000 50mm prime is not 10x better than a $400 50mm prime.
Nov 08, 2024 at 05:15 PM
gdanmitchell
Offline
Details
p.2 #17ED elements , lens design and 3D pop
MARKFER wrote:
Just curious, what are the lens qualities outside of focal length, that pp can't or won't in the future just be a slide away? When it gets to the point will photography be better for it?


Arguably, focal length sort of is something that can be manipulated. We crop and that has essentially the same effect as using a longer lens. Yes, there are some potential image degradation issues, but most of the time we have high enough image resolution to crop a fair amount with excellent results.

Some kinds of filtering can’t really be replicated accurately in post. Polarizing filters, for example, work on the light as it is captured, and what they manipulate is no longer present in the image file.

We could argue that neutral density filters can’t really be replicated in post, since you can’t really increase the exposure time in post. (I can sort of imagine some future software that might try to emulate this for long exposures, but not now.)

We can’t really manipulate aperture/DOF in post either, There is some intelligent software that does simulate something like this — you’ll find it on your iPhone.

One more thing. A thought experiment…

If some particular lens is what produces this thing people want to call “3D pop,” won’t all photographs made with that less have it? And wouldn’t lenses that don’t have “it” always, in every case fail to produce photographs with so-called “3D pop?” Why or why not?

Edited on Nov 09, 2024 at 10:44 PM · View previous versions
Nov 08, 2024 at 06:35 PM
tsdevine
Offline
Details
p.2 #18ED elements , lens design and 3D pop
So here is a shot from today. Does this have any 3D pop? If it does.....is it the lens, the lighting, the leading lines, the contrast, the color, the focal length, the aperture.....some combination of the above. (That's assuming people even see 3D pop.)

https://live.staticflickr.com/65535/54128855635_9368975dc4_h.jpg

As Dan is sort of implying, if it was solely the lens, you'd see it in every shot from that lens.

Edited on Nov 11, 2024 at 06:46 AM · View previous versions
Nov 09, 2024 at 06:19 PM
philip_pj
Offline
Details
p.2 #19ED elements , lens design and 3D pop
Expert marksmen miss the occasional shot. Side-by-sides with same FL/exposure/post cancel out 'noise' (flat harsh lighting, composition differences, etc.). It has to be done this way, to be valid, too many other variables get in the way with single, stand alone images.

It's most apparent in faces which we look at and for from birth. Poor face rendering is something you see very easily, and it's a real issue in the popular very fast lenses, in particular. Alex Barrera sees it very clearly in this YouTube, start watching at 5:25 for his monitor work, through to 13:00, maybe 10-12 images.

'Canon RF 50mm 1.2 L vs Leica APO-Summicron-SL 50mm f/2'

It's literally as plain as the nose on your face, every single image. He was still in the 'more bokeh and circular bokeh equals better' mindset, which has since fallen from favor somewhat. Also, see the effect colour and contrast have on DOF, life and general impact.

The Canon 50/1.2 is an overwhelming 15 elements, the 50/2 SL has 12 elements. Really, neither are stars at image depth but you see how easy it is to observe winners and losers.
Nov 09, 2024 at 09:07 PM
gdanmitchell
Offline
Details
p.2 #20ED elements , lens design and 3D pop
philip_pj wrote:
Expert marksmen miss the occasional shot. Side-by-sides with same FL/exposure/post cancel out 'noise' (flat harsh lighting, composition differences, etc.). It has to be done this way, to be valid, too many other variables get in the way with single, stand alone images.

It's most apparent in faces which we look at and for from birth. Poor face rendering is something you see very easily, and it's a real issue in the popular very fast lenses, in particular. Alex Barrera sees it very clearly in this YouTube, start watching at 5:25 for his monitor work, through to 13:00, maybe 10-12 images.

'Canon
...Show more →

No. It is not as plain as the nose on my face. And my nose is pretty clear. ;-)

Again, if a lens is is the thing that produces this (God, I hate the term!) “3D pop,” then every photo you make with it should exhibit that characteristic, right? And no lens that lacks magic “pop” thing will ever produce a photograph that is worthy? Right?

Also, whatever the objective differences are between lenses like the two fine lenses you mention (and their performance is not identical, obviously, but sometimes a difference is just a difference and not a hierarchy, The differences will be far, far smaller than the changes we regularly introduce in post to optimize images for print or screen.

Maybe, just maybe, back when people shot transparency film and did not post-process, these small, ostensible differences between lenses were more important. Or even when shooting negative film and printing, where post-processing adjustments were not as flexible as they are today. But we no longer live in that world.

Edited on Nov 10, 2024 at 08:39 AM · View previous versions
Nov 09, 2024 at 10:48 PM
FM ForumsSony ForumJoin Upload & Sell
  
 

You are not logged inLog inRegister