• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Lens corrections equivalent to audio corrections when possible

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
For most of my life listening to music and playing around with cameras and lenses have been my principle hobbies.
This means I have inevitable supplementary hobbies of HiFi and Photography.

Recent posts about equalisation and preference in audio remind me of the similar equivalents I have been looking into in optics for years.

My particular interest has been lenses, and I have studied them for over 50 years.
Whilst sharpness, field flatness, chromatic aberration, vignetting and distortion are easily measured their importance is application dependant.
Distortion is very important in architectural photography, and shortcomings obvious on subjects with straight lines but is non-obvious on most subjects and perspective distortion an inevitable result of wide angle shots so not correctable as such anyway but distortion, in general, is easily corrected in digital photography.
Flatness of field is important for copying but probably never noticed photographing 3D subjects, even with fast lenses, since most of the field won't be sharp anyway.

There are a couple of, IMO, crucial aspects of lens performance that I have never seen a convincing way of measuring.
One is known as boke, the rendition of the out-of-focus (oof) part of the field. With fast lenses this is very important to the look of the picture since almost the whole field will be oof. The Canon Lens Work books state that having both sagittal and radial MTF graphs running close together is an indication of good boke, but no technical explanation is given and almost zoom lenses achieve this though several of their excellent tele lenses do.

The other lens feature that is super critical IME is flare sensitivity, both to in frame and out of frame light sources. I have never seen a measurement parameter for this yet IME it is the shortcoming most likely to completely ruin a photograph. :(

When the four-thirds standard was agreed for digital photography since it was a new, post digital standard, digital correction of lens aberrations, mainly distortion, was incorporated. This meant lenses could be optimised much smaller and lighter than they could otherwise have been.
Legacy lenses could not do this since they would be fitted to cameras with film or no standard for digital correction.

My guess is now that almost nobody uses film and all camera makers have correction files for their lenses to supply to Adobe etc. maybe all lenses are more compromised in some areas to make the more extreme ones feasible.

AFAIK distortion, vignetting and chromatic aberrations are correctable to a certain degree.

Flare certainly isn't, nor is flatness of field. Poor boke probably isn't either, two of these are crucial to photography.

So it seems to me that optics which are so often discussed simply in terms of sharpness, when this is only one of many important lens characteristics, are much less "correctable" than audio, which is much simpler to measure everything audible.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,053
I think flare is mostly a quality of coatings issue.
 
OP
Frank Dernie

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
I think flare is mostly a quality of coatings issue.
Coatings can improve contrast and reduce flare, but it is inherent in the lens design and caused by reflections from element surfaces, so is usually worse with designs which use a lot of elements, either zooms or multiple elements to avoid using very expensive optical glass.
High contrast lenses, ie ones with few elements usually, but not always, have less flare.
I have seen calculated predictions of the likelihood of flare, and contrast measurement gives a clue, but not ever seen anybody measuring it.
I am familiar with the Zeiss publications.
I was an avid enthusiast of their clever designs in the 60s, like the hologon, and the reason they stuck with the with the Sonnar design for so long.
I have sold my collection now, but used to own at least one of all the Zeiss designs except the Mirotars, and pretty well every Leitz lens too.
I have books going back decades.
It always surprises me when people write about the importance of round aperture blades for boke when it is wide open that boke matters most and all "apertures" are round then!

The dichotomy for me is lenses have a lot of important characteristics, but people seem to be satisfied by "sharp" whereas audio is much simpler but we get flowery language and complex discussions :)
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
Coatings can improve contrast and reduce flare, but it is inherent in the lens design and caused by reflections from element surfaces, so is usually worse with designs which use a lot of elements, either zooms or multiple elements to avoid using very expensive optical glass.
High contrast lenses, ie ones with few elements usually, but not always, have less flare.
And then there were those early Canon 24-105 f/4 IS L samples where they'd forgotten to paint a screw black.
 

sq225917

Major Contributor
Joined
Jun 23, 2019
Messages
1,360
Likes
1,610
I used to work for a computer graphics hardware and software manufacturer who developed a raytracing chip, we sold rack mount graphics servers for photoreal visualisation for cad cam, manufacturing, architecture and cgi.

It was mostly staffed by Cambridge uni rainbow labs grads, very smart guys. One of the tricks we could do was simulate the visual effects of various lenses by modelling the physical elements and characteristics in maths and then providing a software interface to the various settings on the actual lenses. All very high end at the time late 90s. And all very accurate, good enough to replace real models and studio photography in most cases.

As a party trick the same box could be used to model the acoustics of cad modelled environments to a very high level. Sound and light is closer than we think. Its all just math...
 

NTK

Major Contributor
Forum Donor
Joined
Aug 11, 2019
Messages
2,656
Likes
5,819
Location
US East
This post is a bit off-topic, but nonetheless optics related :D

I used to work in the semiconductor manufacturing equipment industry, and have a fascination in the optical technologies of semiconductor lithography. Zeiss is the undisputed top dog in this field, and along with its partner AMSL ASML, the only game in town for EUV. I am definitely not an opticist, and know next to nothing about optics and the underlying theories.

The technology is super fascinating. Here is a link to an article written 15 years ago during the early stage of EUV litho development. There is no known material with adequate transparency to EUV to make refractive lenses. (Even air is too absorptive for EUV and the EUV litho process has to be performed inside vacuum chambers.) All the optics are reflective, i.e. using shaped mirrors. Quote from the article https://optics.org/article/22383 :
Requirements for the optical quality of the mirror surfaces are also staggering. For a mirror 100 mm in diameter, the acceptable root-mean-square surface roughness is only 0.2 nm. In addition, the thickness variations of the individual layers of coating must be controlled to within 0.1% - a mere 7 pm.
'pm' is picometer — a billionth of a millimeter. Imagine the advancements made in the past 15 years since the article was written. This technology is absolutely mind-boggling.

This is a 2009 paper from Zeiss on using "ion beam figuring" to manufacture their optics. It is machining the optics surfaces by knocking away molecular layers at a time using an ion beam. Wonder if these technologies will ever trickle down to the manufacturing of photography lenses. :)
https://www.researchgate.net/publication/229121942_Ion_beam_figuring_for_lithography_optics
 
Last edited:

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,437
Likes
4,686
This is a 2009 paper from Zeiss on using "ion beam figuring" to manufacture their optics. It is machining the optics surfaces by knocking away molecular layers at a time using an ion beam. Wonder if these technologies will ever trickle down to the manufacturing of photography lenses. :)
https://www.researchgate.net/publication/229121942_Ion_beam_figuring_for_lithography_optics

RCOS has been offering ion milled telescopes for a very long while. They used to sell to amateur astronomers, not sure they do now, and they have been acquired by another company, also apparently sleeping.

https://www.rcopticalsystems.com/telescopes/ion_milling.html

They had a very good mechanical and optical reputation... and prices to match.
In the amateur astronomy market, they were overkill I think. Under normal seeing conditions, the eventual difference between 1/10 lambda to 1/100 lambda is completely lost, physical aperture (unobstructed diameter) trumps everything else if there is no major optical issue and thermal stability is a bitch.

I wonder what would be the benefit of near-perfect surfaces is devices such as zoom lenses where there are twenty of them which are expected to operate anywhere between -25C and 60C. (the argument for canon white pro lenses were thermal btw, not sure if it was marketing/differentiation or real added value though).

OTOH, I think ion beam milling is used in the finishing of some microlenses (the small lenses that typically go above the CMOS wells to refocus the light that would otherwise be lost to the electronic, non sensing part of the cells). No immediate reference/link here.
 

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,437
Likes
4,686
So it seems to me that optics which are so often discussed simply in terms of sharpness, when this is only one of many important lens characteristics, are much less "correctable" than audio, which is much simpler to measure everything audible.

I guess it's a bit like music the scene/subject, the composition, the lighting, the creativity of the photographer is what matters.
Not unlike music, where the piece being played, the artists playing it and the recording/mastering engineers are the main factors.

In photography, we also have "pixel peepers" who will shoot their awful dogs with a wide-angle lens and argue endlessly about top right corner sharpness or CA.

I guess a lot of us here could be called "dB peepers"
 

MDAguy

Senior Member
Forum Donor
Joined
Nov 26, 2020
Messages
404
Likes
405
Location
San Francisco Bay Area
Perfect timing for this topic Frank... I just got ahold of a Leica Noctilux-M 50mm f/0.95 ASPH, a lens not know to be very sharp, and yet creates the most amazing photos (when you're in focus!!) and been learning about how extreme optics behave ... going to follow this thread.
 
Last edited:

sq225917

Major Contributor
Joined
Jun 23, 2019
Messages
1,360
Likes
1,610
Silicon chip fab tech blows my mind, the scales are barely imaginable.

Of course they pale into insignicance with the accuracy of a hole in one at golf....
 

Berwhale

Major Contributor
Forum Donor
Joined
Aug 29, 2019
Messages
3,933
Likes
4,922
Location
UK
Silicon chip fab tech blows my mind, the scales are barely imaginable.

In the early 90's, I applied for and was offered a programming role at company called Edwards High Vacuum. At the time, Edwards supplied semicon fab vacuum pumps to IBM, amongst others. The thing that always made me chuckle was that Edwards also designed vacuum pumps for the meat processing industry that were used to suck the gibblets out of chickens...

Anyway, I ended up taking another role in the finance sector and that ended my chance of a direct association with the semicon/chicken gibblet industry :)
 
Last edited:

NTK

Major Contributor
Forum Donor
Joined
Aug 11, 2019
Messages
2,656
Likes
5,819
Location
US East
Silicon chip fab tech blows my mind, the scales are barely imaginable.

Of course they pale into insignicance with the accuracy of a hole in one at golf....
How about sub nano-radians accuracy? Almost hole in one to the moon :D
(From: http://euvlsymposium.lbl.gov/pdf/20...EUV Lithography Extendibility/S9.1_Kaiser.pdf)

euv1.JPG
 

JeffS7444

Major Contributor
Forum Donor
Joined
Jul 21, 2019
Messages
2,347
Likes
3,508
I won't claim to understand exactly how they achieve it, but modern prime lenses such as Sony G Master and Olympus Zuiko Pro deliver some of the smoothest out of focus results that I've ever seen, with only modest distortion, chromatic aberration, vignetting and corner softness. Not copies of classic Zeiss Distagon, Sonnar or Planar designs, some are mind-bogglingly complex:

https://www.sony.com/electronics/camera-lenses/sel100f28gm#product_details_default
https://www.getolympus.com/us/en/lenses/m-zuiko-ed-25mm-f1-2-pro.html (19 elements in 14 groups!)

No doubt advances in optical coatings, optical materials, production techniques, and a lot of cheap computer power have really allowed lens designers to achieve a level of correction which would've been impossible in the past without compromising contrast / flare resistance. But having said that, I still have a soft spot for my prewar 6 x 4.5 Zeiss Ikonta folding camera with it's uncoated Zeiss Jena Tessar lens too!
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,348
Likes
3,462
Location
San Diego
I think the biggest similarity between lenses and audio is the "magical thinking" that goes on in the high end of both hobbies. The "Gestalt" of an unubtainium Leica lens and the unequaled musicality of an original Western electric 300b tube for example.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
I won't claim to understand exactly how they achieve it, but modern prime lenses such as Sony G Master and Olympus Zuiko Pro deliver some of the smoothest out of focus results that I've ever seen, with only modest distortion, chromatic aberration, vignetting and corner softness. Not copies of classic Zeiss Distagon, Sonnar or Planar designs, some are mind-bogglingly complex:

https://www.sony.com/electronics/camera-lenses/sel100f28gm#product_details_default
https://www.getolympus.com/us/en/lenses/m-zuiko-ed-25mm-f1-2-pro.html (19 elements in 14 groups!)

No doubt advances in optical coatings, optical materials, production techniques, and a lot of cheap computer power have really allowed lens designers to achieve a level of correction which would've been impossible in the past without compromising contrast / flare resistance. But having said that, I still have a soft spot for my prewar 6 x 4.5 Zeiss Ikonta folding camera with it's uncoated Zeiss Jena Tessar lens too!

Computer aided design and new coating technology have been a big part of this, as you suggest. Another factor is the ability to design and manufacture complex aspheric elements. Some of it is also design goals. Outside of the cinema industry, photographers didn't used to talk much about how out-of-focus areas looked, and lens designers didn't pay attention to it. So when a lens had nice blur, it was usually a happy accident, typically due to undercorrected spherical aberration. Nowadays, photographers care about out-of-focus performance and are willing to pay for it. This has shifted priorities in the industry.

Another shift is the expectation of high performance over a full range of apertures and magnifications. It used to be more common for a lens to have a fairly narrow sweet spot that you'd learn to work with. One of my favorite lenses is old-school like this. It's a Schneider 28mm shift lens for Nikon. You have to limit yourself to a couple of apertures, and you can only shift it about 6mm (out of a possible 12). If you agree to these constraints, you get state-of-the-art results. If you don't, it looks like a bad lens. I can work with this, but contemporary standards demand much more flexibility.
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
I think the biggest similarity between lenses and audio is the "magical thinking" that goes on in the high end of both hobbies. The "Gestalt" of an unubtainium Leica lens and the unequaled musicality of an original Western electric 300b tube for example.

This is true, but I think that photography is a bit less susceptible to bias, at least in the digital age when you can prepare raw files identically and look at them side-by-side on the same screen. You don't have the complexity of blind audio testing, and you don't have to remember what something sounds like, and you don't have the mental stress of trying to hear infinitessimal details in a medium that's flying by at the speed of sound.

With two images next to each other it's pretty clear if there are differences that matter or that exist at all.

But not everyone will do this. Magical thinkers gonna magically think.
 
OP
Frank Dernie

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
I won't claim to understand exactly how they achieve it, but modern prime lenses such as Sony G Master and Olympus Zuiko Pro deliver some of the smoothest out of focus results that I've ever seen, with only modest distortion, chromatic aberration, vignetting and corner softness. Not copies of classic Zeiss Distagon, Sonnar or Planar designs, some are mind-bogglingly complex:

https://www.sony.com/electronics/camera-lenses/sel100f28gm#product_details_default
https://www.getolympus.com/us/en/lenses/m-zuiko-ed-25mm-f1-2-pro.html (19 elements in 14 groups!)

No doubt advances in optical coatings, optical materials, production techniques, and a lot of cheap computer power have really allowed lens designers to achieve a level of correction which would've been impossible in the past without compromising contrast / flare resistance. But having said that, I still have a soft spot for my prewar 6 x 4.5 Zeiss Ikonta folding camera with it's uncoated Zeiss Jena Tessar lens too!
It is the computer programmes.
The "standard" designs which had been shown to have known strengths and weaknesses were then calculated for focal length using log tables.
Once computer ray tracing was fast enough any element layout could be investigated without it being almost a life's work so now we do indeed have some fantastic lenses which are complex.
It has always been the case that prime lenses have much better boke than zooms as well as having less distortion, flare and a flatter field. Zooms are getting better but I have a big psychological problem using a zoom for anything other than snapshots of grandchildren wizzing about!
 

paulraphael

Active Member
Joined
Dec 18, 2020
Messages
262
Likes
367
Location
Brooklyn, NY
It has always been the case that prime lenses have much better boke than zooms as well as having less distortion, flare and a flatter field. Zooms are getting better but I have a big psychological problem using a zoom for anything other than snapshots of grandchildren wizzing about!

I did too, until a few years ago when I needed a couple of very wide lenses for a project. I consulted with some architectural photographers and dove into reviews and measurements, assuming I'd end up renting Zeiss primes. But everything pointed me to the Nikon 14-24. I rented it for a day and was so impressed that I sold my most expensive lens to buy a copy of my own. It's one of the best lenses I've used of any type for any format. And it's not even state-of-the-art anymore ... I believe it was designed in 2007. There are now better zooms.

I have no idea about its bokeh, because I've never used it for that (and it doesn't come up much with ultra-wides). It's not world class at the long end of its range, and it's indeed susceptible to flare. And it weighs a ton, and has a very vulnerable bug-eye front element, so not perfect, and not right for everyone. But for my purposes I don't think a better lens was available.
 
Top Bottom