Lens diffraction - truth or urban legend?

[Sonnar Brian]

I would suggest that to spend the sums that some here spend on their cameras and lenses (presumably, to achieve higher image quality), and then to pretend that diffraction does not matter, and to use those cameras and lenses under conditions that reduce their performance to that of a Canon Rebel with the kit zoom, might be pathological.

This statement threw me. On first reading, it seems like "Why stop down an expensive lens and get the performance of a kit zoom". By F8, most lenses are quite good and give reasonable performance. I'm hard pressed to tell the difference between a $1000 50mm lens and a $100 50mm lens at F8. But I'll stop both down to F8 if I need the DOF.

 

[semilog]

...I'll stop both down to F8 if I need the DOF.

Of course!

 

[Sonnar Brian]

In the 70s, I worked with a Physicist that specialized in X-Ray Crystallography that referred to "Conservation of Inconvenience". Seems appropriate here, DOF gain vs Diffraction Loss.

 

[ampguy]

Hi Brian

Hi Brian

good summary. The only other real world issues I have concerned myself with relating to this is I never know what images I may eventually want ultimate resolution, and I have been in situations where I needed to use the aperture to stop down, specifically for some long exposures with water where I didn't need more DOF, and didn't have slower film/sensor or ND filters on-hand, but needed the slow shutter speed for the effect I wanted.

If you browse say the u4/3 and smaller sensor forums and galleries, you will see diffraction limited images (especially reds), no pixel peeping needed.

I could put in a few formulas and numbers regarding diffraction and resolution from Neblette and Hecht, but it would not be of much value to the discussion. Personally, I almost never stop down below F5.6 and almost always shoot wide-open and at F4. I really want to put the new 12.5mm F1.3 Cosmicar on the EP2 because the aperture closes down all the way, and then hit it with a LASER. Remarkable Restraint.

But- if you need depth of field in an image, the way to get it is to stop down. The benefit gained by having a larger depth of field for the overall image will outweigh the resolution lost due to diffraction. If you focus at 10ft, and stop down to F11, an object at 5ft will be better resolved even with the increased effects of diffraction compared with opening the lens up to F4. The object at 5ft will not be as resolved at F4 as it would be at F11. That is the call that the photographer must make for the sake of overall quality of the image. If you are shooting test targets at a fixed distance, then DOF does not matter. But in the real world, it is DOF vs diffraction.

 

[semilog]

If you browse say the u4/3 and smaller sensor forums and galleries, you will see diffraction limited images (especially reds), no pixel peeping needed.

Once more, with spirit:

If you can see it in micro 4/3, you can see it in APS-C. As I demonstrated above, diffraction can make a 200% or even 300% difference in linear resolution in APS-C, while the difference between 4/3 (or micro 4/3) and APS-C is 17 to 20%.

 

[semilog]

I have no idea what you know about sensors.

A major chunk of my professional work since 1998 has depended on ultra-high resolution (position detection at or below 1 nm accuracy, using light with wavelengths of 530 nm [green] to 1064 nm [IR]) and/or ultra-high sensitivity (single molecule) microscopy. Sensor efficiency and diffraction (and, less germane to the present discussion, the chemical and physical properties of fluorescent probes) are the limitations in these kinds of work. So yeah, I've spent a lot of time thinking about this stuff.

 

[Doug]

...If you browse say the u4/3 and smaller sensor forums and galleries, you will see diffraction limited images (especially reds), no pixel peeping needed.

Ok, I'm curious... Ted, how do you detect diffraction limited images? What characteristics do you look for to see the effects of diffraction? Truly, I'd like to know what to look for...

 

[semilog]

Ok, I'm curious... Ted, how do you detect diffraction limited images? What characteristics do you look for to see the effects of diffraction? Truly, I'd like to know what to look for...

Apropos of this discussion, this video (about perhaps the most beautiful experiment in physics, which concerns diffraction) is just wonderful.

http://www.youtube.com/watch?v=DfPeprQ7oGc

 

[Doug]

Apropos of this discussion, this video (about perhaps the most beautiful experiment in physics, which concerns diffraction) is just wonderful.

Yes, but... 🙂 wouldn't a closer examination of the results of the single slit be more relevant here? The amount/nature of the spreading & diffusion as the waves/particles are affected by the edges of the slit, with varying slit widths?

 

[semilog]

Yes, but... 🙂 wouldn't a closer examination of the results of the single slit be more relevant here? The amount/nature of the spreading & diffusion as the waves/particles are affected by the edges of the slit, with varying slit widths?

The double-slit experiment is highly relevant for photography. Any analysis of an optical system using waves instead of linear ray-tracing (that is, a simplification where photons are treated as ballistic particles instead of waves) must deal with diffraction and interference. Interference explains why a diffraction-limited optical system will always draw a point source of light (a distant star, for example) not as a point, or as a gaussian, but as a point spread function with a visible Airy disk.

 

[ampguy]

nope

nope

4/3 diagonal is 21.6mm, Nikon/Pentax/Sony APS is 28.4mm, 35mm FF is 43.3mm.

4/3 will in theory, and in practice become diffraction limited before a larger sized sensor of the same technology and relative pixel pitch size and spacing.

Once more, with spirit:

If you can see it in micro 4/3, you can see it in APS-C. As I demonstrated above, diffraction can make a 200% or even 300% difference in linear resolution in APS-C, while the difference between 4/3 (or micro 4/3) and APS-C is 17 to 20%.

 

[ampguy]

summary please?

summary please?

Would you mind giving a couple sentence summary of the youtube content?

The last 2 youtube links you posted were about the video quality of a hacked Panasonic GH1, which is interesting for video folks, but was unrelated to the thread topic.

Apropos of this discussion, this video (about perhaps the most beautiful experiment in physics, which concerns diffraction) is just wonderful.

http://www.youtube.com/watch?v=DfPeprQ7oGc

 

[ampguy]

Hi Doug

Hi Doug

It looks like a smearing, but also has a characteristic of reduced resolution and contrast, but different than high iso noise characteristics.

Take a look at this demo that shows diffraction limitation from f8 to f22 or so:

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

scroll down to "what it looks like"

Ok, I'm curious... Ted, how do you detect diffraction limited images? What characteristics do you look for to see the effects of diffraction? Truly, I'd like to know what to look for...

 

[semilog]

4/3 diagonal is 21.6mm, Nikon/Pentax/Sony APS is 28.4mm, 35mm FF is 43.3mm.

4/3 will in theory, and in practice become diffraction limited before a larger sized sensor of the same technology and relative pixel pitch size and spacing.

We've now established that (1) you can do arithmetic; and (2) you get the same answers as me and everyone else who can also do arithmetic. Congratulations.

Here is the question that you are avoiding: how do you get from there to being "able to see diffraction" in micro 4/3 pictures at web resolution (I'll be generous: 800 pixels wide), without (you claim) pixel peeping?

(Sneak preview: you don't.)

Here's the problem that you face: At f/22, with a 10 mpix 4/3 sensor, the Olympus Zuiko 50 mm macro delivers a measured resolution (MTF-50) corresponding to >740 line pairs/ph along the picture's short axis (and hence 1000 lp/ph along the long axis).

That's at f/22, as far as the lens stops down.

Here are the measurements.


In other words, no, you can't see diffraction in images from a 4/3 sensor at web resoution - unless by "web resolution" you mean images larger than 1500 x 2000 pixels, and images taken at f/22 or smaller.

At a more reasonable aperture, f/8, the same lens is Nyquist-limited (right out to the corners!) for the same 3648 x 2736 sensor.

So you can "see" the diffraction limitation of images (and separate this limitation from other sources of noise and contrast degradation) taken on 4/3 cameras on web-based images, exactly how?

I'm deeply skeptical.

 

[semilog]

One last point: I encourage anyone who still thinks that there's a huge difference between the resolution of 4/3 and APS-C systems under diffraction-limited conditions to check out this comparison: a tale of two macros. Set both lenses to f/22 and see what you get.

Seriously. Check it out.

By the way, these lenses are not cherry-picked. You'll get pretty much the same result with any two similar (or even dissimilar) lenses of decent quality.

 

[semilog]

Would you mind giving a couple sentence summary of the youtube content?

It's a nicely-illustrated short lecture on the classical experiments that demonstrated the particle-wave duality for electrons (and, in similar experiments, for photons and other elementary particles), with no math.

For a longer, but still non-mathematical treatment of this and related quantum weirdness, I can't suggest Richard Feynman's brilliant, thin book QED: the strange theory of light and matter too strongly. It's really wonderful.

 

[Sparrow]

Ok, I'm curious... Ted, how do you detect diffraction limited images? What characteristics do you look for to see the effects of diffraction? Truly, I'd like to know what to look for...

Yep me too, an example would be good, I set up the worst combination I have, my 12mm and my daughters GF1, the camera was on auto and while I can see there are differences between f5.6 and f22 I'm unsure what I'm actually looking at

this was the set-up




these are the too shots one wide open at f5.6 and the other at f22




[URL="http://www.flickr.com/photos/19217760@N05/4940859321/sizes/l/in/photostream/"][/URL]

 

[Sonnar Brian]

The above image is like finding the "missing objects between these two drawings".

The focus scale on the J-8 is missing! The white lettering has disappeared when compared with the second image. The grain on the table is missing.

Was the lighting held constant?

A guy that worked in my building during the 80s won the Nobel prize for work in diffraction and X-Ray Crystallography. He recently retired.

 

[surfer dude]

Well, I can never understand the hate that non-users of 4/3 cameras have for the format. If I listened to you ampguy I'd just trash the 1000s of photos I took with my E3 and E1 because obviously they're rubbish (technically at least). And to say that you can TELL the inferiority of the format just by looking at a web post just indicates the inane sort of garbage that gets spoken.

 

[Sparrow]

The above image is like finding the "missing objects between these two drawings".

The focus scale on the J-8 is missing! The white lettering has disappeared when compared with the second image. The grain on the table is missing.

Was the lighting held constant?

A guy that worked in my building during the 80s won the Nobel prize for work in diffraction and X-Ray Crystallography. He recently retired.

Well you know my skill with digi-cams, it was on auto it did one at f22 1/30 and 400 asa, the other f5.6 1/40 at 100 asa, for whatever reason

The lighting was the same on both