Diffraction limit and focal length

[devoh]

Hey Gang,

The August 2014 issue of Outdoor Photographer makes an interesting claim on page 104.. They are talking about diffraction and say this..

"Ansel Adams and Group f/64 took their name from that tiny aperture , but remember that they were using large-format cameras, where a 120mm lens was wide-angled. The physical size of the aperture (which determines the occurrence of diffraction) is larger for a 120mm lens stopped down to f/64 than a 24mm lens stopped down to f/22"

This is the first time I have heard the claim that focal length has a part in the equation. I can believe that a large-format camera would be able to be stopped down more than a DX or FX camera, before diffraction is an issue.. but not for the reason they state. What are your thoughts?

Regards
-Devoh

[ajohnw]

Hey Gang,

The August 2014 issue of Outdoor Photographer makes an interesting claim on page 104.. They are talking about diffraction and say this..

"Ansel Adams and Group f/64 took their name from that tiny aperture , but remember that they were using large-format cameras, where a 120mm lens was wide-angled. The physical size of the aperture (which determines the occurrence of diffraction) is larger for a 120mm lens stopped down to f/64 than a 24mm lens stopped down to f/22"

This is the first time I have heard the claim that focal length has a part in the equation. I can believe that a large-format camera would be able to be stopped down more than a DX or FX camera, before diffraction is an issue.. but not for the reason they state. What are your thoughts?

Regards
-Devoh

Sounds like an attempt to redefine physics - after a fashion. The F ratio on it's own determines the defraction spot size. The focal length determines the image scale. The size of the sensor determines just how many diffraction spots can be fitted in.

At F64 the spot size is about 0.09mm so a lot more can fit into a 10x8 plate camera than in 35mm format size. The standard lens for a 10x8 camera is around 300mm, 35mm around 55mm. If exactly related both would have the same angle of view.

5x4 which is what I believe they mostly used has a standard lens focal length of about 160mm

Actually I suspect f64 would waste the resolution available on the film that was available in their day.


John
-

[ajohnw]

As an add on there may be another reason why they might use F64. A different method of determining depth of field. Say they were using as stated a 120mm lens. At F64 that would resolve 120/64=1.9mm in a recognisable fashion right up to the camera - where ever it was focused. Curious thing is that if the camera is focused on infinity it will try to resolve 1.9mm even at that distance but can't due to diffractions effects but the far horizon will be as sharp as it can be for that setting.

In practice when this method is used with focal lengths like this the immediate foreground is unlikely to be visible to the camera so the resolution is know to be better than 1.9mm and reaches it's best value at the point the lens is actually focused on and then drops of at the same rate as it improved past that point.

John
-

[DanK]

What I know about this does not go much beyond what is in the relevant tutorial on this site, but based on that, the quotation is incorrect. I suggest you read the tutorial. It includes this:

Technical Note: Independence of Focal Length
Since the physical size of an aperture is larger for telephoto lenses (f/4 has a 50 mm diameter at 200 mm, but only a 25 mm diameter at 100 mm), why doesn't the airy disk become smaller? This is because longer focal lengths also cause light to travel further before hitting the camera sensor -- thus increasing the distance over which the airy disk can continue to diverge. The competing effects of larger aperture and longer focal length therefore cancel, leaving only the f-number as being important (which describes focal length relative to aperture size).

However, the large film size in those cameras would matter, as it would affect the amount of magnification needed for a given size of print.

[Glenn NK]

http://en.wikipedia.org/wiki/F-number

Maybe I don't understand the question but, the key statement is:

In optics, the f-number (sometimes called focal ratio, f-ratio, f-stop, or relative aperture)of an optical system is the ratio of the len's focal length to the diameter of the entrance pupil.
It is a dimensionless number that is a quantitative measure of the len's speed, and an important concept in photography.

f/stop = focal length divided by diameter of entrance pupil.

The focal length appears to be quite important.

Glenn

[devoh]

What I know about this does not go much beyond what is in the relevant tutorial on this site, but based on that, the quotation is incorrect. I suggest you read the tutorial. It includes this:



However, the large film size in those cameras would matter, as it would affect the amount of magnification needed for a given size of print.

Right, I have read and use the tutorial on this site. My surprise is this magazine (which should know what it is talking about) disagrees with the tutorial. So which is right? I suspect this site and tutorial is correct, as I have never heard the claim made in that magazine before. So I wanted to ask others if they have ever heard this "new" claim.

[george013]

Right, I have read and use the tutorial on this site. My surprise is this magazine (which should know what it is talking about) disagrees with the tutorial. So which is right? I suspect this site and tutorial is correct, as I have never heard the claim made in that magazine before. So I wanted to ask others if they have ever heard this "new" claim.

Diffraction limitation depends on 2 values: first the size of the airy disk created by the beam and second the sensorpixel size in digital photography. If the airy disk is larger or much larger than the sensorpixelsize, than the sharpness wil disappear.

Quote from above:
"Since the physical size of an aperture is larger for telephoto lenses (f/4 has a 50 mm diameter at 200 mm, but only a 25 mm diameter at 100 mm), why doesn't the airy disk become smaller? This is because longer focal lengths also cause light to travel further before hitting the camera sensor -- thus increasing the distance over which the airy disk can continue to diverge. The competing effects of larger aperture and longer focal length therefore cancel, leaving only the f-number as being important (which describes focal length relative to aperture size). "
This explanation doesn't make sense to me.

What I've discovered is that not the f-number is used, but the gonimetrical relation F and the distance to the film/sensor. It/s an expression of the topangel of that ray comming through the aperture. Just for simplicity the f-number is used.

https://en.wikipedia.org/wiki/Airy_disc
http://en.wikipedia.org/wiki/Diffraction-limited_system

George

[dje]

Dave I agree with what the others have said and believe that the CiC tutorial is correct. The maths are explained in the first link that George quotes above. It shows that the size of the Airy disk is dependant on wavelength, distance between aperture and sensor (d) and inversely on aperture size (a). The distance between aperture and sensor is equal to the focal length of the lens (f) (except for very close macro type focus). When this is put into the equation, the size of the Airy disk becomes proportional to wavelength and f stop (f/a) only.

So how did Ansel and co get away with f/64 ? i think it is because they were using much larger format cameras and hence the reduction factor of the image was much less. Ie the lens didn't need to be as sharp as the detail didn't have to be resolved as finely. I'd be interested to hear your view.

Dave

[ajohnw]

What I put in my first post is EXACTLY correct. Perhaps another way of putting it.

Say there are 2 lenses both F4 that have the same angle of view and one is for a larger format than the other so it's focal length is also longer.. Say that the larger format is twice as big so the focal length will also be twice as big. If for instance F4 means that resolution limits the shorter lens to some figure the same figure will apply to the larger format so it will out resolve the smaller format by a factor of 2. The limit of resolution on the sensor is related to the size of the diffraction spot which in turn is set by the F ratio and nothing else.

It's a bit like comparing crop and full frame sensors. If the same lens is used on both they resolve the same and the image is exactly the same size. If the full frame camera is fitted with a longer lens so that the image is bigger so the resolution will also be higher given qualities are identical.

So F64 is less of a problem on a large plate camera than it would be on any dslr because the diffraction spot size is a relatively small proportion of the size of the sensor or film in this case.

John
-

[TonyW]

What John says is quite correct. The diameter of the Airy disc does not depend on the focal length for a given f number. At f/22 the diameter if the Airy disc is 0.03mm compared with 0.09mm for f/64, i.e. the diameter is proportional to the f number, irrespective of focal length. However, with a large format camera they are presumably not enlarging the picture as much as with a 35mm camera and hence a larger f number is possible without serious effects.

The original quote "The physical size of the aperture (which determines the occurrence of diffraction) is larger for a 120mm lens stopped down to f/64 than a 24mm lens stopped down to f/22" is actually correct too, but does not explain the situation.

[dje]

What I put in my first post is EXACTLY correct. Perhaps another way of putting it.

Say there are 2 lenses both F4 that have the same angle of view and one is for a larger format than the other so it's focal length is also longer.. Say that the larger format is twice as big so the focal length will also be twice as big. If for instance F4 means that resolution limits the shorter lens to some figure the same figure will apply to the larger format so it will out resolve the smaller format by a factor of 2. The limit of resolution on the sensor is related to the size of the diffraction spot which in turn is set by the F ratio and nothing else.

It's a bit like comparing crop and full frame sensors. If the same lens is used on both they resolve the same and the image is exactly the same size. If the full frame camera is fitted with a longer lens so that the image is bigger so the resolution will also be higher given qualities are identical.

So F64 is less of a problem on a large plate camera than it would be on any dslr because the diffraction spot size is a relatively small proportion of the size of the sensor or film in this case.

John
-

John as far as I can see, we are saying the same thing - just in a different way.

And as you say, the same issue comes up when comparing FF and crop cameras. The same issue also comes up when people question why small sensor cameras usually have f/8 as their smallest f stop setting - because the resolving power of the lens is more critical with the smaller image sensor, not because the smaller physical apertures create more diffraction.

Dave

[ajohnw]

True Dave

The other point of course is depth of field effects as the format size is changed. As the format size increases the dof for the same aperture decreases when the angle of view remains the same. On 5x4 plate cameras f64 will be needed to get a decent depth of field. 35mm full frame has a crop factor of say 4 compared with that.

That odd dof calc I mentioned in the quick post after the 1st one does seem to work. I tried it recently and it resolved what I expected close to the camera. Only problem was that I knew I would be cropping it but had no idea where. It turned out that the background was so sharp I applied a weak Orton effect to it to tone it down. I'd guess that method of handling dof might have been common knowledge in some quarters in their time.

John
-

[george013]

What John says is quite correct. The diameter of the Airy disc does not depend on the focal length for a given f number. At f/22 the diameter if the Airy disc is 0.03mm compared with 0.09mm for f/64, i.e. the diameter is proportional to the f number, irrespective of focal length. However, with a large format camera they are presumably not enlarging the picture as much as with a 35mm camera and hence a larger f number is possible without serious effects.

The original quote "The physical size of the aperture (which determines the occurrence of diffraction) is larger for a 120mm lens stopped down to f/64 than a 24mm lens stopped down to f/22" is actually correct too, but does not explain the situation.

See it some other way and subtitute the F-number by the topangel of the lightcone. When I computed it good, it's 0.895 degree. It's the only constant. The statement that the physicall size of the aperture determines the occurence of diffraction is wrong.

And if the unsharpness due to diffraction is visible, that could also depend on the printsize etc.

George

[dje]

The other point of course is depth of field effects as the format size is changed. As the format size increases the dof for the same aperture decreases when the angle of view remains the same. On 5x4 plate cameras f64 will be needed to get a decent depth of field. 35mm full frame has a crop factor of say 4 compared with that.

Yes indeed.