Luminous Landscape Forum
The Art of Photography => Discussing Photographic Styles => Topic started by: stamper on February 06, 2013, 07:56:53 am
-
I think I have a good understanding of diffraction and it's drawbacks. As a rule of thumb an aperture smaller than f/11will mean softness in an image. Yesterday I was looking at a book by David du Chemin and he has a couple of images that look sharp - at least to me - that he took at f/22. The reason he used f/22 was to give himself a shutter speed of 15 seconds in a long exposure. I understand what he was doing but he obviously ignored the "problems" of diffraction. Brian Peterson in his book about Exposure doesn't think diffraction is a problem. So how did David get his image to look sharp? I can't obviously post a copy.
-
Because there is nothing to compare it to. If he took another shot at, say, f/11, and you were able to see it at 100% on your screen, you would see it is sharper. Besides, we are talking about a book reproduction (offset?) at a drastically reduced size (as compared to pixel peeping). Under those conditions, almost everything looks good enough. Also, defraction effects can be sharpened away with a proper technique (deconvolution).
-
Not all scenes contain the kind of detail that is needed to spot diffraction degradation. Or if they do, our minds may be willing to forget about it if the image/scene still feels satisfying?
The distinction between true detail and apparent detail is blurry (pun intended). If we perceive the scene to be "sharp", does not mean that a really large amount of information about the true scene was preserved.
-h
-
Slobodan I understand about deconvolution and I assumed that was possible. A lot of photographers make a big issue about not going smaller than f/11. What you state makes sense. :)
-
Hi,
I agree with Slobodan.
Diffraction starts to have a negative effect already at f/8, the better the lens the earlier it is seen. But it is a gradual degradation of image contrast. Diffraction is bening to sharpening.
Check left column on the link below, it shows effect of diffraction:
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=1
The link below shows the effect of sharpening:
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=2
Best regards
Erik
Slobodan I understand about deconvolution and I assumed that was possible. A lot of photographers make a big issue about not going smaller than f/11. What you state makes sense. :)
-
In a real world, ignore the theory. Use the diaphragm stop that will give you the depth of field that you require, and that's all you need worry about.
Folks have to realise that they are dealing with a photograph and not a laboratory test. Unless they actually do see their photography as a sequence of laboratory tests, in which case, good luck to them, and have fun.
Rob C
-
If you look at William Neils work you will see that he commonly uses f16 or f22. I asked him about why he uses such small f stop and said that it is easier to "repair" resolution with sharpening software. He uses Noise Ware.
-
I think I have a good understanding of diffraction and it's drawbacks. As a rule of thumb an aperture smaller than f/11will mean softness in an image. Yesterday I was looking at a book by David du Chemin and he has a couple of images that look sharp - at least to me - that he took at f/22. The reason he used f/22 was to give himself a shutter speed of 15 seconds in a long exposure. I understand what he was doing but he obviously ignored the "problems" of diffraction. Brian Peterson in his book about Exposure doesn't think diffraction is a problem. So how did David get his image to look sharp? I can't obviously post a copy.
Although diffraction has the same effect on the image plane with any format size, its effects on the final print are greater with small sensor formats, since they have to be "enlarged" more to produce the final print. Afterall, users of 8 x 10 inch film used to expose at f/45. What sensor is Mr du Chemin using?
Bill
-
Folks have to realise that they are dealing with a photograph and not a laboratory test.
Which reminds me of what I see on Nikonians.org's D800 forum. Evidently the majority of D800 owners bought the camera to test rather than to make photographs. The forum is a laugh a minute.
-
Just in case we forgot Michael' article on the subject (from the film era):
Stop It — Down! (http://www.luminous-landscape.com/essays/stop-d.shtml)
And its conclusion:
Is this shot sharp enough for almost any purpose? Yes!
Would it have been even sharper if I'd shot at f8 or f/11. Yes, probably.
Does it matter? No, not in the slightest.
Stop worrying! Take Pictures!
Enough said.
-
There used to be a group of photographers who always (?) used f:64... Very sharp photographs. They were mostly using 8x10, by the way.
There is no rule that diffraction starts at f:11 or f:8 or whatever, as it depends on the sensor/film size and the wavelength we are most interested in (green usually). There is always diffraction, but it does not bother us until the sensor is sharper than the diffraction spoiled image. With D800 the limit is around f:8, with best APS-C cameras at f:5.6, with tiny sensors in some P&S and video cameras even below f:4. Actually some small sensor cameras can not resolve as good as the sensor pixel count promises, as the low speed lenses have too much diffraction even at full open.
Like already said, diffraction has always been with us and it does not bother us unless we know better = there is something you can compare it to. We have not been able to see much of it before digital photography and today's sensors, which give us the ability to pixel peep at huge enlargement size and make fast comparison series at different apertures without cost. Nobody bothered to shoot Technical Pan-X to make tests, and if somebody did, there was no Internet to spread the bad news...
Here is the tutorial again, look at the table near the end...
http://www.luminous-landscape.com/tutorials/resolution.shtml
-
Hi,
Somebody actually did...
http://www.photodo.com/topic_138.html
In real life, sharpening also matters a lot. Diffraction is benign to sharpening and sharpness is far from always the most important factor in imaging.
Best regards
Erik
Nobody bothered to shoot Technical Pan-X to make tests, and if somebody did, there was no Internet to spread the bad news...
Here is the tutorial again, look at the table near the end...
http://www.luminous-landscape.com/tutorials/resolution.shtml
-
Hi,
Somebody actually did...
http://www.photodo.com/topic_138.html
No surprise really, but now EVERYBODY is doing them...
-
Hi,
What is interesting with the quoted article is that the authors demonstrates the effects by microscope photographs on film. (135, 120 and 4x5" film)
Best regards
Erik
No surprise really, but now EVERYBODY is doing them...
-
In a real world, ignore the theory. Use the diaphragm stop that will give you the depth of field that you require, and that's all you need worry about.
Folks have to realise that they are dealing with a photograph and not a laboratory test. Unless they actually do see their photography as a sequence of laboratory tests, in which case, good luck to them, and have fun.
Rob C
Absolutely! Once they have mastered the basics of the technical side, a photographers work is going to be judged by the subject, the composition, the light, the mood - anything except the difference in diffraction effects between f8 and f22. Technical competence is important, but technical excellence is secondary to the other qualities of a photograph.
Jim
-
Although diffraction has the same effect on the image plane with any format size, its effects on the final print are greater with small sensor formats, since they have to be "enlarged" more to produce the final print. Afterall, users of 8 x 10 inch film used to expose at f/45. What sensor is Mr du Chemin using?
Bill
In the photograph he used a Canon 5D mark II. I think he now has jumped ship to Nikon. For anyone who hasn't read his books he is - imo - a very fine photographer and worth a read. I often do long exposure work with ND filters which mean I use f/16 a lot to get the shutter speeds I want. Reluctant to got to f/22 because of possible softness but not now!
-
I often do long exposure work with ND filters which mean I use f/16 a lot to get the shutter speeds I want. Reluctant to got to f/22 because of possible softness but not now!
If you have sufficient ND filtering available, why do you have to use f/16?
-h
-
I like to control my initial shutter speed to 1/125 second in aperture mode which I know will give me about 20 seconds in manual mode. This tends to give you a small F stop. In sunset shots for clouds and water It suits what I am shooting using a 10 stop ND filter. With the 10 stop you have to focus and expose in aperture mode - screw the filter on - and switch to manual so setting exposure means you are limited. If I were to use a larger F stop then the shutter speed would be shorter but if I use a 6 stop or 3 stop or both combined then I am more flexible about setting the F stop because exposing and focusing can be achieved with the filters on the lens.
-
Just in case we forgot Michael' article on the subject (from the film era):
Stop It — Down! (http://www.luminous-landscape.com/essays/stop-d.shtml)
And its conclusion:
"Is this shot sharp enough for almost any purpose? Yes!
Would it have been even sharper if I'd shot at f8 or f/11. Yes, probably.
Does it matter? No, not in the slightest.
Stop worrying! Take Pictures!
Enough said."
Michael makes a point, but according to the excellent tutorial (http://www.luminous-landscape.com/tutorials/resolution.shtml) by Osuan and Garcia if Michael stopped down his Phase One IQ180 he would turn his expensive 80 MP camera into an 8 megapixel device with hardly more resolution than an iPhone.
BTW, an iPhone can take nice pictures.
Regards,
Bill
-
Hi,
I think it is OK to stop down when needed, but I really think that stopping down beyond f/16 causes considerable loss of sharpness, that can be compensated with sharpening to e certain extent.
Best regards
Erik
Michael makes a point, but according to the excellent tutorial (http://www.luminous-landscape.com/tutorials/resolution.shtml) by Osuan and Garcia if Michael stopped down his Phase One IQ180 he would turn his expensive 80 MP camera into an 8 megapixel device with hardly more resolution than an iPhone.
BTW, an iPhone can take nice pictures.
Regards,
Bill
-
The first time I noticed diffraction in real life was when doing some macro photography, where stopping down is necessary for DOF. Lo and behold, pictures actually started getting softer after f:11 (shooting with 16 MPix FF body) and f:22 was already quite bad and f:32 even more so. These were not for exhibition prints and for a half page reproduction even f:22 was usable, but the effect of diffraction was crystal clear.
-
Hi,
This is a good example of loss of image detail:
(http://echophoto.dnsalias.net/ekr/images/DMBFigures/F56_vs_F16_cropped.jpg)
The line pattern is not visible on the f/16 image on the left but clearly resolved in the f/5.6 image. Both images deconvolution sharpened in LR.
Full article: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/68-effects-of-diffraction
Best regards
Erik
The first time I noticed diffraction in real life was when doing some macro photography, where stopping down is necessary for DOF. Lo and behold, pictures actually started getting softer after f:11 (shooting with 16 MPix FF body) and f:22 was already quite bad and f:32 even more so. These were not for exhibition prints and for a half page reproduction even f:22 was usable, but the effect of diffraction was crystal clear.
-
In a real world, ignore the theory. Use the diaphragm stop that will give you the depth of field that you require, and that's all you need worry about.
Hi Rob,
The problem with such broad sweeping statements is, that they are worthless advice in specific circumstances that many professionals and amateurs alike encounter on a daily basis. Granted, some of those are not producing 'artistic images' where blur and lack of resolution can be coined to be a feature rather than a flaw.
Folks have to realise that they are dealing with a photograph and not a laboratory test. Unless they actually do see their photography as a sequence of laboratory tests, in which case, good luck to them, and have fun.
That is also not helpful for those who want to make a more deliberate choice of their camera settings, understanding (all) the trade-offs involved. Mastering one's tools is also part of becoming a better photographer, and it's more fun than being mastered by one's tools. Do you shoot with your camera in Program mode, or do you deliberately choose a shutter speed or aperture priority for a given situation?
There apparently is still a lot of confusion about diffraction, which BTW is the topic of this thread, so if a few laboratory tests can help to improve one's understanding, I don't see the downside of that ...
Cheers,
Bart
-
Now that we have exposure bracketing (with in-camera hdr synthesis), flash bracketing, WB bracketing and what not, when are we going to have ergonomic focus bracketing?
"I want reasonable sharpness from 2m to 6.2m. I define sharp to mean a PSF not exceeding effective 3 pixels. Use the focus motor to take however many images needed, and let me know when you are done".
Or:
"I am touching the areas on my liveview touchscreen that matters to me (in terms of sharpness) in this image. Now make me an image that is sharp in those spots".
Some people would object to such automation, claiming that it is not "real" photography, and "where is the art in it when the camera does it all for you". Hopefully, Samsung does not listen to those, and choose to make cameras that help people make better images instead.
-h
-
Wasn't Contax N (or whatever the designation was) providing focus brackating already years ago? Also, some Canon models used to have a mode where you select two focusing points and the camera calculates both the needed f/stop and sets the necessary distance to keep both sharp.