Have optic design changed in the digital age?

[Oren Grad]

I can say with authority that it’s hogwash, having designed optics since the start of the digital era. Nothing against you, of course.. it’s an internet myth or misconstruing of a minor technical detail. As a linear system, the lens has no idea what the imaging media is, so the light strikes both imaging media the same. I speculated on a recent podcast interview that the myth may have sprung from the description of pixels as quantum wells, which makes one think of a water well with some “depth” that needs to be “filled”. That is, however, a failure of analogy. The focal plane array is flat with no structure and the light-sensitive silicon detector starts at the surface.

If nothing else, the degraded corner and edge performance of short-focal-length M-mount rangefinder lenses used on mirrorless digital cameras suggests a need to adapt optical designs to the attributes of the sensor cover glass and microlenses.

 

[Theo Sulphate]

Trendland, I mean no disrespect at all but let me be blunt: I seriously can't understand what the hell you're writing 90% of the time. For the sake of effective communication could you please convert comments in your native language into google translate and paste the English version in until you've gotten better with grammar? I would love to respond to your comments but I can't tell what you're trying to say. You may want to do the same converting back to your native language because you're missing the full meaning of what you read...

Yes. Trendland, please seriously consider what has been suggested. Many people here truly can't understand what you're trying to say and often you misinterpret what others write.

This is not disrespect, but a constructive suggestion to help both you and Photrio.

I will be the first to offer that if I'm misunderstood here, I will happily translate it to Hungarian for them.

 

[trendland]

It's a fool's errand to try and figure out what he is talking about. Just skip over his posts. They are gibberish.

Yeah faberryman you got it - applause!

with regards

PS : Faberryman don't like me? Perhaps he is unable to fill his opinion with arguments?
Because I have the impression- folks he often have conflicting viewpoints?
That is total OK - from my point! But I can count the post 30 from faberryman here with lack
of topics to talk about?......hmmm!

PPS : It is glibberish yeah - but it is also a question from lack of language skills from you
Faberryman - isn't it?

I for my part have no idea what a " fool errant " could mean! But I have not to translate it because I can compensate that missing grammar! And be sure I unterstand the content you
wrote 100%!

PPS : Missing the sentence :" Set him on ignore " - hmm? Perhaps it will come next?

 

[trendland]

Yes. Trendland, please seriously consider what has been suggested. Many people here truly can't understand what you're trying to say and often you misinterpret what others write.

This is not disrespect, but a constructive suggestion to help both you and Photrio.

I will be the first to offer that if I'm misunderstood here, I will happily translate it to Hungarian for them.

Guess you have the job....!

with regards

People - you Americans should train your skills in foreign languages - that can help!

 

[faberryman]

If nothing else, the degraded corner and edge performance of short-focal-length M-mount rangefinder lenses used on mirrorless digital cameras suggests a need to adapt optical designs to the attributes of the sensor cover glass and microlenses.

I get the impression that the only problem is with certain wide angle Leica M-mount lenses. I have not heard of any other manufacturer's wide angle lenses being a problem. If you want to use Leica wide angle M-mount lenses, you best bet is to use a Leica M-mount digital camera.

 

[Theo Sulphate]

...
People - you Americans should train your skills in foreign languages - that can help!

Not everyone here is American, nor a native English speaker.

 

[trendland]

Or, like me, just put him on IGNORE and not waste time trying to understand the gibberish.

Welcome to the club!
silveror0 I am waiting for your response!

BTW : Have optical design changed in the digital age?

Guess there is no idea from you right? Sit down please 5-....

 

[trendland]

Not everyone here is American, nor a native English speaker.

Theo I know - you are from Hungaria!

To explain : the first 1/3 of post#15 isn't understandable! That would be the same if it would be
in hungarian native speach! It is because the content!

with regards

PS : It is full ironical 100% with some science fiction physics! Just a joke!

 

[Oren Grad]

I get the impression that the only problem is with certain wide angle Leica M-mount lenses. I have not heard of any other manufacturer's wide angle lenses being a problem.

Same issue with many C/V M-mount lenses or LTM lenses. Short-register wides in general are especially vulnerable.

If you want to use Leica wide angle M-mount lenses, you best bet is to use a Leica M-mount digital camera.

I agree with this. But I am interested in Jason's perspective on the optical design considerations that apply to this problem.

 

[faberryman]

I agree with this. But I am interested in Jason's perspective on the optical design considerations that apply to this problem.

I understand that the Leica M mount sensor has special microlenses to counteract this issue. Whether such microlenses introduce shortcomings with respect to other lenses, thus precluding general adoption in other cameras, is unknown to me. My own feeling is use your Leica M-mount lenses on Leica bodies; that is what they are designed for. If you want to use Leica M-mount lenses on other bodies, accept that it is an issue. Other camera makers shouldn't be taken to task for not providing optimum performance for a handful of lenses from other manufacturers owned by a small group of photographers on cameras they weren't designed for. For best results, use the proper tool for the job and all that.

 

[GLS]

In my opinion, you have to step up to the Zeiss Otus to beat it.

The Zeiss Milvus 85mm is already a considerably better lens, let alone the Otus.

The Sigma Art 85mm is also very good.

To address the question in the OP though: I would stick with the 85mm f1.4D. Its bokeh is outrageously good.

 

[neilt3]

So when Konica Minolta came out with D lens, just a marketing ploy, no real difference between the D and older lens? I don't shoot much color film anymore, but when I have and use D lens on Minolta 9 I don't see any difference in color. Can say the same with Sigma, the Sigma 50 1.4 (Older Version but post film) worked just fine with color film on SA9.

You wouldn't . It's got nothing to do with colour , or if the lens was designed for film or digital .
When Minolta brought out the "D" lenses they put a distance encoding chip in the lens so it could report to the camera at what distance it is focused at .
The "D" is for Distance , not digital .
You won't see a difference between "D" and non D lenses as it;s an electronic feature , not optical .
Some earlier lenses were updated to include the chip , nothing else on the lens changed .

This was brought out when they introduced ADI flash control , an update from TTL-OTF flash control .
Rather than measuring flash exposure off the film and cutting it when sufficient had bounced back , ADI works by the camera knowing what focal length lens is used ( or a zoom set at ) , knowing what aperture is set and knowing how far away the subject is , and knowing the power of the flash .
With this information the camera set the correct output of the flash and isn't fooled by such as reflective surfaces .
have a read here ; https://www.mhohner.de/sony-minolta/flashcomp_mm.php

All well and good , but doesn't function with bounce flash or when a filter is used on the lens , instead you have to revert back to TTL-OTF metering .

Non of this is any use to you using a Dynax 9 as Minolta didn't introduce ADI flash control until the Dynax 7 , so you camera can't make use of the information and just used TTL-OTF flash metering .

All this is off topic anyway as Minolta using the "D" designation has no relevance to Nikon using the "D" designation the O.P is asking about regards to if there is an optical difference between the two .

 

[film_man]

I am looking at getting either the Nikkor 85/1.4D or 85/1.4G for my Nikon F5. While I know the G is a much newer design, I am wondering if the older D, designed in the era of film would actually be a better lens for me. Here is why I ask:

Many, many years ago (early 2000's) I recall reading that long, fast lens designed for film photography where not idea of digital photography because of DOF and the media being used... What I recall was that the lens design for films took into account that the RGB layer in the film was physically at a different distance (plains) from the lens. Because of that, company, especially in the large format world, started coming out with lens for digital sensors where all three colors needed to focus at the same plain.

My questions:

1. Is this a true fact that lens designed for film don't work so well on digital and thus lens designed for digital don't work so well on film?
2. If this is the case, does anyone know if the Nikkor 85/1.4G is a digital lens design? I assume it is since it came out only a few years ago long after digital became the go to format.

To answer your questions:

1. Some old lenses do not work well on digital cameras because the angle of light is too extreme, causing severe vignetting to said digital cameras, smeared corneres or, in some extreme cases, colour shifts (see Zeiss 21/4.5 ZM on Leica M8 and M9). This is generally an issue with wide lenses, longer lenses are fine. On the other hand, modern lenses designed to fix this issue will work fine with film with no side effects. You also get the benefit of much sharper optics due to newer glass/lens design over the years.
2. The 85/1.4 G is a modern design suited for digital but will do just as well on film. You may or may not benefit from the improvements over the D model but you certainly won't be worse off. Your main concern is G lens compatibility on older cameras but your F5 will be fine.

 

[Dahod]

In keeping with the title of your thread ( Have Optic Design Changed in the Digital Age) I suspect the answer is yes. Maybe someone already brought this up and I missed it - sorry if I did. Increasingly, manufacturers are turning to in-camera software for improving lens performance - you can't even test the lens without the camera hooked up. Focus-by-wire lenses are another example. Such lenses would be difficult (if not impossible) to apply satisfactorily to a film camera.

Somebody once said you can focus a pop bottle if you know the math and I fear that's the direction we're headed. I can still buy and use a 150 year old lens on my film cameras but my kids will be using my digital lenses for paperweights.

 

[film_man]

I can say with authority that it’s hogwash, having designed optics since the start of the digital era. Nothing against you, of course.. it’s an internet myth or misconstruing of a minor technical detail. As a linear system, the lens has no idea what the imaging media is, so the light strikes both imaging media the same. I speculated on a recent podcast interview that the myth may have sprung from the description of pixels as quantum wells, which makes one think of a water well with some “depth” that needs to be “filled”. That is, however, a failure of analogy. The focal plane array is flat with no structure and the light-sensitive silicon detector starts at the surface.

Please, authority? At the bare minimum there is a piece of glass in front of the sensor, you can see it and you can clean it when dirt lands on it and shows up in the images. Just take the camera, fire the shutter on bulb and look at it. That makes it 3D. It is there, you can see it yourself and touch it.

By the way, here's what Canon has to say.
https://cpn.canon-europe.com/content/education/infobank/capturing_the_image/microlenses.do

 

[Sirius Glass]

It's a fool's errand to try and figure out what he is talking about. Just skip over his posts. They are gibberish.

Copy that but did you have to defame gibberish?

 

[David Brown]

Surely optical designs have changed. It seems the new digital lenses are much bigger and have more elements. ... The new Panasonic 50mmm is also huge and heavy.

Not to go off-topic, but, I held one of these Panasonics just this morning. It is larger than the 85mm "normal" lens on my RB67! The rep "explained" that it was because it was f1.4. I pulled out my Takumar 50mm f1.4 when I got home just to assure myself I wasn't crazy ...

 

[Nodda Duma]

Please, authority? At the bare minimum there is a piece of glass in front of the sensor, you can see it and you can clean it when dirt lands on it and shows up in the images. Just take the camera, fire the shutter on bulb and look at it. That makes it 3D. It is there, you can see it yourself and touch it.

By the way, here's what Canon has to say.
https://cpn.canon-europe.com/content/education/infobank/capturing_the_image/microlenses.do

Yes, please, authority. I’ve been designing optics for quite some time and know very well what I’m talking about. Of course, we all know that expertise doesn’t count for squat on the internet and *certainly* not on photrio.

In any case, not sure what point you’re trying to make but it has been noted.

 

[Nodda Duma]

Not to go off-topic, but, I held one of these Panasonics just this morning. It is larger than the 85mm "normal" lens on my RB67! The rep "explained" that it was because it was f1.4. I pulled out my Takumar 50mm f1.4 when I got home just to assure myself I wasn't crazy ...

That’s good humor. Joking aside, 50 / 1.4’s are pretty much all the same. If the Panasonic lens is legitimately larger, my guess would be the designer tried not to use vignetting (i.e. reducing the lens element diameters) to correct aberrations. That’s a common approach for double gauss types. If that’s the case, the light falloff would be non-existent but the lens would be softer wide open.

 

[Theo Sulphate]

...
In any case, not sure what point you’re (film_man) trying to make but it has been noted.

That article seems to address a different issue entirely, viz. microlenses on the sensor, whereas the idea being debunked involved the angle of light from the lens.

 

[StepheKoontz]

As a linear system, the lens has no idea what the imaging media is, so the light strikes both imaging media the same. I speculated on a recent podcast interview that the myth may have sprung from the description of pixels as quantum wells, which makes one think of a water well with some “depth” that needs to be “filled”. That is, however, a failure of analogy. The focal plane array is flat with no structure and the light-sensitive silicon detector starts at the surface.

I know from use, on a mirrorless body like a sony A7, a lot of the older design rangefinder wide lenses smear and basically look pretty horrible due to the light ray angles. This is a pretty well known issue that these lenses that work fine with film, are almost useless on a digital body.

 

[wiltw]

I get the impression that the only problem is with certain wide angle Leica M-mount lenses. I have not heard of any other manufacturer's wide angle lenses being a problem. If you want to use Leica wide angle M-mount lenses, you best bet is to use a Leica M-mount digital camera.

In the early days of digital, with APS-C sensors, it was noted in the media that super wide angle FL had a problem with the different wavelengths of light being spread at different angles, and this cause the individual color-specific sensels to see light a bit more 'misaligned' (my term) and causing color fringing. So WA lenses for digital were designed so that the light rays struck the sensor surface a bit closer to perpendicular to the surface in an effort to reduce color finging with extremely short FL lenses.

 

[Alan Johnson]

System resolution Rs, 1/Rs ~ 1/Rlens + 1/Rfilm (or sensor) [some use R squared in this equation}
For the quality lenses considered here, unless a special high resolution film or sensor is used, it appears likely that the system resolution will not be affected much by the aerial resolution of the lens.It is not the usual practice to give aerial resolution of lenses in lppm but what little data exists suggests this may be upward of 200lppm for a good lens compared to circa 100 lppm for a film or sensor. If one of the two Nikon lenses considered has a slightly higher resolution than the other it wont usually have much effect on system resolution.

 

[film_man]

Yes, please, authority. I’ve been designing optics for quite some time and know very well what I’m talking about. Of course, we all know that expertise doesn’t count for squat on the internet and *certainly* not on photrio.

In any case, not sure what point you’re trying to make but it has been noted.

The point is that saying the sensor is flat and the angle of light does not matter is what is hogwash. The sensor may be flat but the microlenses and protective glass in front of it make it have depth so the angle matters, as demonstrated numerous times with numerous lenses on bodies like the M8 and the Sony. But anyway, you're the expert so you know all that. Bye.

 

[Peltigera]

The rough explanation I heard regarding lenses "made for digital" was that such lenses were optimized so that light would strike the sensor more straight-on rather than at an angle.

I might be missing something important, but if the light is striking the sensor/film straight-on, does that not mean that the light is unfocused? Does not being in focus necessarily mean that the light is at an angle as it strikes the sensor/film?