Pages: [1] 2 3 4   Go Down

Author Topic: Understanding Lens Diffraction  (Read 8861 times)

HSakols

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1239
    • Hugh Sakols Photography
Understanding Lens Diffraction
« on: June 05, 2019, 11:30:26 am »

This article was written a number of years ago but I still think about it.  Yes, there is nothing like finding that sweet spot on your lens, but in the world of landscape photography, there are tradeoffs.  Of the successful landscape photographers I know they still stop down anywhere from f 16 to even f32 in some cases. I surprised I'm not seeing more landscape photographs that are focused stacked? 
Logged

Dave (Isle of Skye)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
  • I've even written a book about it
    • SkyePhotoGuide.com
Re: Understanding Lens Diffraction
« Reply #1 on: June 05, 2019, 04:44:25 pm »

I don't worry about lens diffraction at all these days, because I have found with modern high capacity sensors and a reasonably good sharpening technique, as far as I can see, it is no longer an issue IMHO - unless you are a canvas sniffer that is  :) A jocular term I once heard someone using to refer to other landscape photographers (that would be you and me!), that he would often see inspecting his work in his gallery, with their noses about an inch away from the pictures.

In fact I would even go as far as to say, that with modern sensors (of the Sony persuasion at least) the uncompressed Raw straight out of the camera and without any in-camera sharpening, can still look a little too sharp to me on occasions, so sometimes I find myself even adding a little bit of negative clarity ;)

Great article by MR though and enjoyable to see it and read it once again, but my advice, if you have a modern high capacity sensor, good quality glass, a sturdy tripod and you use a cable release, then use what ever Dof you think you need to get the shot, and if it is a bit soft around the edges, then you can probably fix it quite easily in post and no one will ever notice it but you and perhaps the occasional passing canvas sniffer.  ;)

Dave
« Last Edit: June 05, 2019, 04:48:50 pm by Dave (Isle of Skye) »
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: Understanding Lens Diffraction
« Reply #2 on: June 05, 2019, 07:36:07 pm »

I don't worry about lens diffraction at all these days, because I have found with modern high capacity sensors and a reasonably good sharpening technique, as far as I can see, it is no longer an issue IMHO - unless you are a canvas sniffer that is  :) A jocular term I once heard someone using to refer to other landscape photographers (that would be you and me!), that he would often see inspecting his work in his gallery, with their noses about an inch away from the pictures.

In fact I would even go as far as to say, that with modern sensors (of the Sony persuasion at least) the uncompressed Raw straight out of the camera and without any in-camera sharpening, can still look a little too sharp to me on occasions, so sometimes I find myself even adding a little bit of negative clarity ;)


I think you have it backwards. To take advantage of modern high resolution sensors one should keep the Airy disc diameter no greater than twice the pixel pitch. George Douvos has published an excellent article on this subject. Stopping down to f/16 is a good way to convert a 46 MP Nikon D850 to the equivalent of a 7 MP camera such as the Nikon D70. See figure 3 of this post.

Cheers,

Bill
Logged

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: Understanding Lens Diffraction
« Reply #3 on: June 05, 2019, 07:42:59 pm »

... Stopping down to f/16 is a good way to convert a 46 MP Nikon D850 to the equivalent of a 7 MP camera such as the Nikon D70...

Apart from that seemingly scholastic argument, has anyone actually done any comparison in prints of various sizes to prove the above? At which print size people wouldn’t be able to distinguish 7 Mp from 46 Mp?

LesPalenik

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5339
    • advantica blog
Re: Understanding Lens Diffraction
« Reply #4 on: June 06, 2019, 12:13:12 am »

I wonder how representative are the Rodenstock lens examples relative to other lenses. So while the article is interesting and insightful, a Canon 50mm F1.4 lens, used by many  more shooters, might have very different characteristics.

Maybe Lula could open up a new technical section - Lens Performance Examples, and various members could contribute similar images based on their experiments for Nikon, Canon, Sony, Sigma, FF, APS-C and M43 lenses - 20mm, 35mm, 50mm, 85mm, etc.

Crowd-sourcing project, though it would require a part-time administrator, but the result could be a very comprehensive and useful lens reference library. Available Only On LuLa.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Understanding Lens Diffraction
« Reply #5 on: June 06, 2019, 04:12:05 am »

Hi,

Diffraction is only affected by aperture, or I would think the diameter of the inlet pupil. That is the reason astronomers buy telescope by the diameter. The brightness of a star is dependent on diameter^4.

So, if you have a perfect lens, it is sharpest at maximum apertures. But, few lenses are perfect. The lens is limited by aberrations (uncorrected optical errors), most aberrations reduce with stopping down. So stopping down improves aberrations but causes diffraction. Some of the best lenses get diffraction affected around f/4, or even larger.

So stopping down causes loss of sharpness, due to diffraction, the better lens you have the more you have to loose.

But, diffraction is pretty benign to sharpening. So much of the sharpness lost to diffraction can be restored by advanced sharpening.

I would like to elaborate, but cannot find the time right now...

Best regards
Erik


I wonder how representative are the Rodenstock lens examples relative to other lenses. So while the article is interesting and insightful, a Canon 50mm F1.4 lens, used by many  more shooters, might have very different characteristics.

Maybe Lula could open up a new technical section - Lens Performance Examples, and various members could contribute similar images based on their experiments for Nikon, Canon, Sony, Sigma, FF, APS-C and M43 lenses - 20mm, 35mm, 50mm, 85mm, etc.

Crowd-sourcing project, though it would require a part-time administrator, but the result could be a very comprehensive and useful lens reference library. Available Only On LuLa.
Logged
Erik Kaffehr
 

HSakols

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1239
    • Hugh Sakols Photography
Re: Understanding Lens Diffraction
« Reply #6 on: June 06, 2019, 10:14:20 am »

Quote
Apart from that seemingly scholastic argument, has anyone actually done any comparison in prints of various sizes to prove the above? At which print size people wouldn’t be able to distinguish 7 Mp from 46 Mp?

This would be an interesting comparison and would make for an interesting article. 

Quote
So much of the sharpness lost to diffraction can be restored by advanced sharpening.

I would like to elaborate, but cannot find the time right now...

Both William Neill and Charlie Cramer don't seem too concerned with diffraction by stopping down. 

Logged

Dave (Isle of Skye)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
  • I've even written a book about it
    • SkyePhotoGuide.com
Re: Understanding Lens Diffraction
« Reply #7 on: June 06, 2019, 10:45:27 am »

I think you have it backwards. To take advantage of modern high resolution sensors one should keep the Airy disc diameter no greater than twice the pixel pitch. George Douvos has published an excellent article on this subject. Stopping down to f/16 is a good way to convert a 46 MP Nikon D850 to the equivalent of a 7 MP camera such as the Nikon D70. See figure 3 of this post.

Cheers,

Bill

Well I am sorry to disagree with you Bill, but I think you are entirely wrong in this case if all you are suggesting we do is base our view of photography on such meaningless technical issues, which may indeed be absolutely correct in a science over aesthetics sort of way. Because I would argue long and hard, that photography is no longer about attaining eye watering sharpness anymore, or the technical performance of exotic overpriced lenses and camera bodies, as that ship has sailed and now even the most basic system has evolved way beyond the abilities of your average Joe photographer these days. So it is now once again (and as it should be IMHO), more about the picture and the art of photography than it is about the kit and so anyone should be able do it reasonably well with a keen eye and a reasonably priced system, that compared to the equipment of old, has gone well beyond what most of us should ever need. I mean if Ansel was alive today, I don't think you or I or anyone else for that matter, would be telling him that his pictures aren't very good, because they're not very sharp or a high enough pixel density?

Dave
« Last Edit: June 06, 2019, 10:50:34 am by Dave (Isle of Skye) »
Logged

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: Understanding Lens Diffraction
« Reply #8 on: June 06, 2019, 11:20:43 am »

Has anyone else noticed that depth of field appears unchanged in all the examples in the article?

Alan Klein

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15850
    • Flicker photos
Re: Understanding Lens Diffraction
« Reply #9 on: June 06, 2019, 11:42:45 am »

When I shoot landscapes with 120 medium format film using by Mamiya RB67, I calculate the DOF I need then stop down one more stop. 

josh.reichmann

  • Administrator
  • Sr. Member
  • *****
  • Offline Offline
  • Posts: 441
Re: Understanding Lens Diffraction
« Reply #10 on: June 06, 2019, 12:53:59 pm »

Has anyone else noticed that depth of field appears unchanged in all the examples in the article?

Well, what is the general correlation between stopping down and depth of field with a constant focal point? Depth of field should be increased no? I think there is too much shadow or dark undifferentiated negative space here to tell. (?)
Logged
Compassion and wisdom are inextricably linked.

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1852
    • Frank Disilvestro
Re: Understanding Lens Diffraction
« Reply #11 on: June 06, 2019, 02:05:22 pm »

Aperture is you variable AA filter. The more you stop down, the stronger AA.

Stopping down to f/16 is a good way to convert a 46 MP Nikon D850 to the equivalent of a 7 MP camera such as the Nikon D70.

Except that in the case of the D850, Moire will be much less of an issue

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1852
    • Frank Disilvestro
Re: Understanding Lens Diffraction
« Reply #12 on: June 06, 2019, 02:20:20 pm »

Has anyone else noticed that depth of field appears unchanged in all the examples in the article?

The article says those are 100% crops. Most likely is that the entire subject was inside the DOF at f/5.6, so stopping down did not have a visual effect in this regards. Instead, stopping down further, only degrades the sharpness due to the increased diffraction.

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: Understanding Lens Diffraction
« Reply #13 on: June 06, 2019, 03:04:08 pm »

Well I am sorry to disagree with you Bill, but I think you are entirely wrong in this case if all you are suggesting we do is base our view of photography on such meaningless technical issues, which may indeed be absolutely correct in a science over aesthetics sort of way. Because I would argue long and hard, that photography is no longer about attaining eye watering sharpness anymore, or the technical performance of exotic overpriced lenses and camera bodies, as that ship has sailed and now even the most basic system has evolved way beyond the abilities of your average Joe photographer these days. So it is now once again (and as it should be IMHO), more about the picture and the art of photography than it is about the kit and so anyone should be able do it reasonably well with a keen eye and a reasonably priced system, that compared to the equipment of old, has gone well beyond what most of us should ever need. I mean if Ansel was alive today, I don't think you or I or anyone else for that matter, would be telling him that his pictures aren't very good, because they're not very sharp or a high enough pixel density?

Dave

Dave,

When I made the post, I already knew that you would disagree. An interesting article on megapixels was published in the New York Times in 2007 (I think that is before they began posting fake news :) ). He did a blinded study and concluded that 6-8 megapixels was sufficient. If this were true, I don't know why anyone would pay US $50,000 for a Phase One IQ4 that is advertised on the opening page of this forum. In any case, LuLa is happy to accept their advertising fee.

Lloyd Chambers (Diglloyd) has written extensively on the effects of diffraction on his pay web site. In a study with the Nikon D7100 (a 24 MP crop frame whose pixel density would result in 56 MP in a full frame camera) with the Zeiss 135 mm f/2 plan apo (perhaps the sharpest available lens at that time) he noted dulling of the image at f/8, even with aggressive deconvolution sharpening. A dull haze was noted at f/11 and f/16 and f/22 showed severe degradation.

I have done my own tests with this lens with my Nikon D850 using Bart Vanderwolf's sinusoidal Siemen's star. Wilth this target, resolution at a MTF of about 10% is proportional to the diameter of the extinction radius. These images are considerably cropped from the full frame to show the area of interest.  At f/4 the system resolves to Nyquist, 102 lp/mm with good contrast. Deconvolution sharpening with FocusMagic improves contrast. At f/22 resolution drops to 72 lp/mm with low contrast. FocusMagic can improve contrast, but resolution remains at 72 lp/mm.

Image at f/4:


Image at f/4 with FocusMagic:


Image at f/22:


Image at f/22 with FocusMagic:

Proper deconvolution sharpening can partially reverse diffraction, but not completely as shown by my test and by Diglloyd's work.

Of course, you will say that proper deconvolution was not applied. What algorithm and PSF do you use?

Cheers,

Bill
Logged

Dave (Isle of Skye)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2514
  • I've even written a book about it
    • SkyePhotoGuide.com
Re: Understanding Lens Diffraction
« Reply #14 on: June 06, 2019, 04:16:01 pm »

Dave,

When I made the post, I already knew that you would disagree. An interesting article on megapixels was published in the New York Times in 2007 (I think that is before they began posting fake news :) ). He did a blinded study and concluded that 6-8 megapixels was sufficient. If this were true, I don't know why anyone would pay US $50,000 for a Phase One IQ4 that is advertised on the opening page of this forum. In any case, LuLa is happy to accept their advertising fee.

Lloyd Chambers (Diglloyd) has written extensively on the effects of diffraction on his pay web site. In a study with the Nikon D7100 (a 24 MP crop frame whose pixel density would result in 56 MP in a full frame camera) with the Zeiss 135 mm f/2 plan apo (perhaps the sharpest available lens at that time) he noted dulling of the image at f/8, even with aggressive deconvolution sharpening. A dull haze was noted at f/11 and f/16 and f/22 showed severe degradation.

I have done my own tests with this lens with my Nikon D850 using Bart Vanderwolf's sinusoidal Siemen's star. Wilth this target, resolution at a MTF of about 10% is proportional to the diameter of the extinction radius. These images are considerably cropped from the full frame to show the area of interest.  At f/4 the system resolves to Nyquist, 102 lp/mm with good contrast. Deconvolution sharpening with FocusMagic improves contrast. At f/22 resolution drops to 72 lp/mm with low contrast. FocusMagic can improve contrast, but resolution remains at 72 lp/mm.

[snip]

Proper deconvolution sharpening can partially reverse diffraction, but not completely as shown by my test and by Diglloyd's work.

Of course, you will say that proper deconvolution was not applied. What algorithm and PSF do you use?

Cheers,

Bill

Hi Bill, well you totally lost me there after the first sentence, as I am more than happy to steer well clear of falling down into that kind of in-depth analytical rabbit hole, which I think sort of proves the sentiments outlined of my original post, which is that I think going into this type of analytical depth, is completely at odds with the requirements for your average Joe photographer these days and is to all intents and purposes, totally unnecessary for improving the aesthetic quality of our images, and which I also think can actually become a severe hindrance to it if you let things like this clutter your mind.

Here is an image a took a couple of nights ago, I was sat in the house watching some rubbish on TV and thought no, I am going out with my camera instead, so grabbed the camera and one wide angle zoom lens. I then went down to my local patch and without a thought in my head (perhaps the tweeting of birds or something as equally mindless), where I just setup in a spot I hadn't shot before (that was my only compositional choice BTW) and grabbed a few frames, before going back to the car for a cuppa with a couple of white chocolate chunky biscuits - which will be the death of me because I just can't stop eating them). The lens was already set on f/16, the camera on fully manual with daylight white balance and single shot mode. I stuck on my 10 stop ND, focussed on the rock near to the vertical rocky outcrop about 2/5ths up the frame on the left. I then wound up the exposure time to 30 secs and the ISO to give me a half stop underexposed, I then pressed the remote switch and continued listening to the bird song in my head and all-in-all, the shot didn't turn out too bad. Is it sharp enough? Who knows, it certainly looks OK to me and would still look OK printed out a 48 inch or even more. Is there a focal point or a central point of interest? Not really, but again I am OK with that, because when I look at it now, I realise that it represents my mood more than it does anything else and if I judge the image in this way, then yes it is a perfectly good image.

So how did I sharpen it? Well I haven't, although I did add some clarity by running it through ACR for a second time. I then duplicated the layer and added some despeckling to the upper layer and then created a sharpness mask for that layer, so that only the areas without detail (sea and sky etc) got the despeckling. Took me about half an hour to finish it and that included dust spotting.

I suppose what I am saying Bill, is that yes you are correct in everything you are saying, but for me it isn't my thing and I have been doing this long enough to realise that I really don't need it - you could say I have managed to find my comfy place and where I no longer have to worry or even think about such things as deconvolution sharpening, Airy disc diameters, or diffraction, because my reasonably priced equipment is good enough, that I can empty my head and let my emotions do the work.

All the best Bill and thank you for allowing me to realise by describing it to you here, how I now work and how happy I have become with it  :)

Dave
« Last Edit: June 06, 2019, 04:22:26 pm by Dave (Isle of Skye) »
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Understanding Lens Diffraction
« Reply #15 on: June 06, 2019, 04:25:35 pm »

Hi Slobodan,

I would say that 24MP was pretty for A2 but I have some pretty decent A2 size prints 12MP. Lets say we need 12MP for good A2, than we would need 6MP for A3.

Another way to see it would be to say that we need 180 PPI for a great print. If we have 6MP, that would yield about 2000x3000 pixels and that would mean an 11”x16.5” print.  But those pixels need to be sharp.

So, I would say that 6-7 MP would be good enough for 11x16.5, but may need an unhealthy amount of sharpening.

Best regards
Erik

Apart from that seemingly scholastic argument, has anyone actually done any comparison in prints of various sizes to prove the above? At which print size people wouldn’t be able to distinguish 7 Mp from 46 Mp?
Logged
Erik Kaffehr
 

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: Understanding Lens Diffraction
« Reply #16 on: June 06, 2019, 04:48:51 pm »

Thanks, Erik, but that is not what I am asking. That’s just another scholastic calculation. As a practicing photographer, who did 24” x 36” from an 8 Mp camera, I got tired of theoretical calculations which would say such a print would look like sh*t. It doesn’t.

What am am suggesting is a visual comparison, a blind test, involving a group of people, photographers and non-photographers, trying to differentiate at which size they would start seeing a difference in prints between:*

- a 7 Mp camera shot at a difraction-optimal aperture and

- a 46 Mp camera shot at f/16

Ceteris paribus, of course.

* something similar to what Michael Reichmann did with Canon G10 and a medium-format Hasselblad, years ago.

David Sutton

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1345
    • David Sutton Photography
Re: Understanding Lens Diffraction
« Reply #17 on: June 06, 2019, 05:48:58 pm »

Apart from that seemingly scholastic argument, has anyone actually done any comparison in prints of various sizes to prove the above? At which print size people wouldn’t be able to distinguish 7 Mp from 46 Mp?

Sort of. Years ago I ran a comparison of diffraction at various f stops for a camera club. Either a Canon 40D or 5D2, can't remember.
Enlarged to 100% you could clearly see the loss of sharpness at f/16 on the projected image.
However, after sharpening (maybe with smart sharpen in CS4?) I would have been quite happy with f/22.
F/32 was still complete mush.
David
« Last Edit: June 06, 2019, 05:52:02 pm by David Sutton »
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Understanding Lens Diffraction
« Reply #18 on: June 06, 2019, 06:50:56 pm »

Hi,

The question is complex.

You could do such a test, but the results may vary by subject.

I have done some experiments, like scaling down an image from 42 MP to 10.5 MP and back from 10.5 MP to 42 MP and printed in a large size, like A2, the image that was downsampled and upsampled looked better than the original, but was sort of brittle.

The way things are, I would think that we can throw away a lot of the information. And get very good prints. But, sharpening is involved. At some level, detail will be lost. But detail matters much less to image quality than most would believe.

Just to say, the 180 PPI for excellent prints is based on a lot of experimental work. But it assumes pretty sharp pixels.

But, I would say that the difference between excellent and good lenses starts to go away around say f/11.

So, it may make less sense to buy a very sharp lens when it will be used mostly stopped down.

But, there are other factors than sharpness, like flare and ghosting. These are not very much effected by stopping down.

Best regards
Erik

Thanks, Erik, but that is not what I am asking. That’s just another scholastic calculation. As a practicing photographer, who did 24” x 36” from an 8 Mp camera, I got tired of theoretical calculations which would say such a print would look like sh*t. It doesn’t.

What am am suggesting is a visual comparison, a blind test, involving a group of people, photographers and non-photographers, trying to differentiate at which size they would start seeing a difference in prints between:*

- a 7 Mp camera shot at a difraction-optimal aperture and

- a 46 Mp camera shot at f/16

Ceteris paribus, of course.

* something similar to what Michael Reichmann did with Canon G10 and a medium-format Hasselblad, years ago.
Logged
Erik Kaffehr
 

nirpat89

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 661
    • Photography by Niranjan Patel
Re: Understanding Lens Diffraction
« Reply #19 on: June 06, 2019, 07:45:40 pm »

"Sharpness is a bourgeois concept."

“He had his little Leica,” [fashion photographer Helmut] Newton remembers, “and he simply would point and shoot.” Since Cartier-Bresson’s hand isn’t as steady as it used to be, some of the pictures were a bit fuzzy. “Sharpness,” he told Newton, “is a bourgeois concept.” Newton sits back and laughs: “I thought that was just divine.”
– Dana Thomas, Newsweek, 6/1/03

Probably tongue-in-cheek.  But shaky hands didn't stop HCB from taking a picture.  Let's not not take a picture because the scene does not fit into the f/8 sweet spot.
Logged
Pages: [1] 2 3 4   Go Up