Diffraction at small apertures, a limiting factor in sharpness?

heartyfisher

1) I would have thought that the amount of diffraction for a given aperture is different between lenses of different focal lengths. but it sounds like that is in contention.
Paperman : Definite no as Ade has explained. All that matters is f-stop

This asserts that the diffraction at say F16 is the same for a 800mm lens and a 16 mm lense .. This cant be correct.. as far as i know the smaller the sise of the physical hole the more diffraction there is so the diameter of the aperture at F16 is at maximum 1mm for the 16mm lense and 50mm for the 800mm lense so the resulting diffraction from the 16mm must be much more than for the 800mm lense

TaoTeJared

1) I would have thought that the amount of diffraction for a given aperture is different between lenses of different focal lengths. but it sounds like that is in contention.
Paperman : Definite no as Ade has explained. All that matters is f-stop

This asserts that the diffraction at say F16 is the same for a 800mm lens and a 16 mm lense .. This cant be correct.. as far as i know the smaller the sise of the physical hole the more diffraction there is so the diameter of the aperture at F16 is at maximum 1mm for the 16mm lense and 50mm for the 800mm lense so the resulting diffraction from the 16mm must be much more than for the 800mm lense

You are correct @heartyfisher - where diffraction becomes an issue does change by lens. It can even change for two different lenses of the same focal length - with different optical designs. Some are trying to ignore what the discussion is around, "becoming an issue" as it is a practical application not some sort of clinical mental exercise.
Calculations utilize the "perfect lens" and do not account for the ability of changing directions of light through advanced lens/optical design.

This is like watching a 12yr old argue that a bowling ball and a feather both fall at the exact same speed because gravity is constant. Other variables always exist.

SquamishPhoto

Optical theory says that a perfect lens will be perfect wide open, and that diffraction will start to take its toll as the lens is stopped down. Some of the very best 35mm lenses show this. I'm thinking of the Canon 300mm f/2.8L IS, which measures as well wide open as stopped down one or two stops. There may be others. But most lenses need to be stopped down to improve some of their other optical characteristics, and so like most things in life there are compromises to be made.

-- http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml

Seems to suggest that diffraction may set in at different apertures depending upon the nature of the lens design, but that the perfectness of lens is only relevant to whether you'd need to stop it down in order to see it perform at its best. A perfect lens is at its best wide open and then diffraction sets in slowly from there as you stop down, whereas most lenses need to be stopped down a little to get to their optimal performance and then only after that point does diffraction begin to set in. The Sigma 35mm f1.4A, and essentially every lens from Nikon's telephoto lens lineup come to mind.

Ade

@SquamishPhoto

Not really. See my graph on previous page on why some lenses need to be stopped down to reach maximum sharpness. Basically, these lenses can't reach the diffraction limit at wide open apertures, due to other aberrations of the lens.

But the diffraction limit itself remains constant for all lenses.

Ade

@heartyfisher

For photographers what we care about is to know when two objects being photographed are so close together that their airy disks overlap.

Since the two objects will be very close to each other, and the angular separation between them will be very small, Wikipedia has a formula to approximate this distance on the sensor:



where x is the closest distance before the airy disks overlap, f is the focal length, λ is the wavelength, and d is the diameter of the entrance pupil of the lens.

Solving for x, we get:



Notice the f/d term above is simply the aperture (f-number) of the lens!

So yes the diffraction limit is the same for an 800mm lens or a 16mm lens.

Paperman

@TaoTeJared
"This is like watching a 12yr old argue that a bowling ball and a feather both fall at the exact same speed because gravity is constant. "

Only that the the the argument/diffraction theory is by scientists/optical physicists who have spent their lives on subjects like this and not by a 12 year old. If you think long enough, you will be able to see who is being the 12 year old here.

One also wonders how something you have so easily figured out - all this possible diffraction related improvement in lens design - has gone amiss by the same scientists who have been involved with the subject maybe for decades. Why and how on earth don't they know anything about this ?? Maybe they are just too lazy to rewrite those papers/articles or modify the theory hoping no one notices.

I remind you that you have NOT come up with a SINGLE scientific article/paper/link backing your opinion or that diffraction shiftable ZEISS ( finding it more suitable to add a quote by some famous person ) where as I and a few people have been bombarding you ( can provide tens more ) with links on diffraction theory.

TaoTeJared

If you actually read every reputable paper instead of half-assed blogs and pretending you did you will ALWAYS see the disclaimer that it uses the "perfect" lens scenario because every lens is different. Copying and pasting wikipedia doesn't prove any point either. In fact it proves how incorrect you are. Lens design is done with only a couple of variables? I don't think so. Even a 12 year old understands the world doesn't work in a vacuum.

Different designs, different materials, even the f-stop is actually different due to accepted tolerances. Ever see the transmission measurements? Part of the loss of transmission is the measured size of the aperture. It is not unusual to see cine specific models of lenses to show a different f-stop as the accepted tolerance for a given aperture is less in filming as the effects are greater on the effected medium. I have seen 2.8 lenses when measured are actually f/3.3. That not light transmission, that is the physical measurements.

Since you are too lazy, this is what 15 second search pulls.
cambridgeincolour
Furthermore, this limit is only a best-case scenario when using an otherwise perfect lens; real-world results may vary.

Diffraction_Limited_Photography_Pixel_Size_Aperture_and_Airy_Disks
(assuming an otherwise perfect lens, when viewed at 100% on-screen)

Nikon's published The Diffraction Barrier in Optical Microscopy
The resolution limits imposed by the physical laws that govern optical microscopy can be exceeded, however, by taking advantage of "loopholes" in the law that underscore the fact that the limitations are true only under certain assumptions. There exist three particularly important assumptions that are brought into play during assessment of resolution criteria, including the conventional geometry in which light is gathered by the objective, the uniformity of excitation light throughout the specimen, and the linear characteristics (absorption and emission) of fluorescence involving a single photon. In brief, resolution can be enhanced by gathering light over a larger set of angles around the specimen or using excitation light that varies with position. The diffraction barrier can also be broken using fluorescence processes that involve two or more photons in a non-linear manner.


I didn't keep the e-mail nor the link from Zeiss as a computer upgrade (or two) and I sold my 50mm 1.5 lens so I didn't try to find it again. I don't keep stuff just in case some asshat from a forum demands it. If you email them with a serial number they will give you the information. I would email them for you but solving your ignorance because you are lazy is not worth my time.


A free falling bowling ball = (F = m•a) (bet everyone remembers that one)

A free falling feather (which is not simplistic) *Drag equation


No one learned that in your basic science class.

Ade

@TaoTeJared

Those technical links completely contradict what you are saying. That you don't even realize this... well what can I say.

E.g., the Nikon article you linked to says so in plain English:

"Even in cases where an optical microscope is equipped with the highest available quality of lens elements, is perfectly aligned, and has the highest numerical aperture, the resolution remains limited to approximately half the wavelength of light in the best case scenario."

Yup, diffraction is still there. Even the best optical system in the world cannot exceed the Abbe limit. Zeiss certainly can't sell a diffraction-buster lens improvement for $100.

All the "loopholes" they talk about to "get around" diffraction limit involve non-optical methods, like shooting individual photons at biological cells. E.g., the section you bolded involves fluorescence microscopy, not reflected-light photography.

Paperman

Well, a 15 second search pulled what WE have been saying in every diffraction discussion - nothing to back your views there... Did also mention before that the only way to beat diffraction could come from sensor design and not lens design ( just a wild guess - with angled pixels maybe but then the angles would have to be " changable " to accommodate both large and small apertures ).

SquamishPhoto

@TaoTeJared

Those technical links completely contradict what you are saying. That you don't even realize this... well what can I say.

E.g., the Nikon article you linked to says so in plain English:

"Even in cases where an optical microscope is equipped with the highest available quality of lens elements, is perfectly aligned, and has the highest numerical aperture, the resolution remains limited to approximately half the wavelength of light in the best case scenario."

Yup, diffraction is still there. Even the best optical system in the world cannot exceed the Abbe limit. Zeiss certainly can't sell a diffraction-buster lens improvement for $100.

All the "loopholes" they talk about to "get around" diffraction limit involve non-optical methods, like shooting individual photons at biological cells. E.g., the section you bolded involves fluorescence microscopy, not reflected-light photography.

This pretty much says it all, doesn't it. Im eager to see him give a rational and reasoned response to this.

Im not holding my breath.

heartyfisher

@heartyfisher

For photographers what we care about is to know when two objects being photographed are so close together that their airy disks overlap.

Since the two objects will be very close to each other, and the angular separation between them will be very small, Wikipedia has a formula to approximate this distance on the sensor:



where x is the closest distance before the airy disks overlap, f is the focal length, λ is the wavelength, and d is the diameter of the entrance pupil of the lens.

Solving for x, we get:



Notice the f/d term above is simply the aperture (f-number) of the lens!

So yes the diffraction limit is the same for an 800mm lens or a 16mm lens.

??? the diameter d of a 16mm lense at F16 is 1mm and the diameter d of a 800mm lense at F16 is 50mm of course they are different .. no?

Paperman

f/d = focal length / diameter = 800mm/50mm = 16mm/1mm ....

TaoTeJared

The limit where diffraction becomes a issue to sharpness is different on every lens.

Nothing you have showed addresses the unique design characteristics of any lens other than the "perfect lens."

Here is an idea - go out and shoot photos and find out. I have had over 50 lenses and test them all to see where that limit is. They have all spanned from f/5.6-f16 where diffraction has become an issue.

We don't live in a test lab vacuum - this is the real world.

Ade

For everyone else:

Refer to my chart from the previous page for a refresher on the behavior of "real world" lenses (blue line); and a followup post further down with reference to ZEISS with data from the "real world" Planar 1.4/85 ZA.

That's all from me on this thread, thanks.

heartyfisher

f/d = focal length / diameter = 800mm/50mm = 16mm/1mm ....

That makes sense but doesnt make sense.. :-) will think on it .. I went to look at the original article on wikipedia and I think my issue is with the assumptions the formula is based on. will post when its clearer in my head..