Small apertures and diffraction?

[donaldamacmillan]

<p>Ladies and gentlemen<br>

I'd be very grateful for some advice and even greater understanding. I have been an enthusiastic and serious photographer for a couple of years, using film and digital SLR, and my biggest passion is the landscape photograph. I'm based in the UK and much of my background reading, magazine publications mostly, often advises 'me' to stop my lens right down so as to ensure maximum depth of field, especially if i'm taking my photo in bright sunshine. That makes sense to me, i think. Before i go much further i should say that i am aware of the hyperfocal method and i do as much as i can try to manually focus one third in to my scene. <br>

Recently i have been reading a landscape special publication and i have been finding out about how using very small apertures to maximize depth of field actually degrades the quality of the image by increasing the amount of diffraction ... How worried should i be about diffraction? I have Nikon lenses (film photography) and Canon lenses (DSLR photography), maybe it's my inexperience but i don't think i've noticed much diffraction in my images? I do resort to stopping right down frequently, should i avoid doing this? Are small apertures bad technique? Or is this kinda defect very lens specific; are some lenses fine while others are badly affected? <br>

Regards from Scotland<br>

Donald</p>

 

[JDMvW]

<p>If you look at a pixel-peeper level at lens tests where this sort of thing is quantified (as in many cases at Photozone.de), you will see that there is a fall-off in image quality, even as there is otherwise an increase in the depth of field. What you are probbly seeing is that increase in depth of field, which gives the appearance of sharpness, even as real sharpness of edges, etc. is lost.</p>

<p>This is a phenomenon created by the very nature of light, so is the case so long as the absolute diameter of the opening is below a certain size.</p>

[jeg]

<p>I think that smaller than f/8 or so diffraction slowly becomes an issue on APS-C cameras.</p>

[Alan Marcus]

<p >When light rays are restricted by a small aperture they no longer strictly obey the laws of geometrical optics. Light rays that closely encounter the edges of the aperture restriction scatter and become misdirected. This scatter light is said to be diffracted. Diffraction occurs at all aperture settings however when tiny apertures are employed a higher percentage of the rays are forced to skim by the edges of the restriction. Thus if the aperture is made too small, the image becomes less sharp.</p>

<p > </p>

<p >As a rule of thumb, the sharpest setting is about 2 f/stops down from maximum. Since the camera lens works by bending light rays inward, rays that hit at the periphery of the lens must be bent the most. Thus the curve (figure) of the lens is greatest at the edges. Rays that transverse the edges are thus more likely to be misdirected whereas central rays are least likely to be misdirected. </p>

<p > </p>

<p >Lens makers are attuned to all this and likely the lens won’t allow the user to go into the truly tiny apertures (f/32 – f/45 – f/64). All lenses are not created equal and lenses of the highest quality are expensive. Likely your lens will do a good job at all provided apertures. The true test occurs when you attempt to make big over the mantel prints. </p>

[lex_jenkins]

<p>Even with the APS format (or DX, with Nikons) sensor, lens diffraction is usually the limitation before the sensor size becomes an issue. With my DX sensor dSLR I can see differences in diffraction between my various lenses. For example, with a 55mm macro lens f/32 is noticeably softer. But f/32 on my 300mm telephoto is still usable. (Same subjects, same distances or compositions.) Another 300mm telephoto I tried in a local shop was significantly softer at f/22 than my 300mm. So you'll just need to test your lenses for yourself and decide whether the diffraction is significant enough to matter.</p>

<p>Another issue is viewer perception. For example, in a portrait as long as the eyes are in focus or, at least, the leading eye is in focus, the average viewer will perceive the photo as "sharp", even with very shallow DOF. In other cases the viewer may perceive a photo at f/8 as "less sharp" than at the minimum aperture, because less of the subject is in focus. They may not even notice softness due to diffraction, as long as the DOF covers the entire subject range. Photographers tend to see things differently than ordinary viewers. We look for faults that others don't.</p>

[paul_de_ley]

<p>Not sure if this matters since you mention landscapes as your favorite subject, but there is an interesting reversal of the principle of stopping down for greater sharpness when you're dealing with real macro- and microphotography lenses (magnifying more than actual size, as opposed to closeup lenses that don't surpass 1:1). At high magnifications you actually obtain the greatest sharpness with diaphragm wide open, closing down even a single stop results in diffraction effects that noticeably deteriorate sharpness as a tradeoff against increased DOF.</p>

[garydem]

<p>when you stop down excessively, you get more dof; BUT it will be with the price of more distorion. this comes from the diffraction effect of the small hole that the light is passing through at very small apertures. only you can decide if the price is worth the distorsion.<br>

generally at any aperture beyond f11.0 in a c senpor dslr or beyind f16.0 for film or FF dihgital yoiu will be gin to see diffraction distorsion. personally, with my c sensor dslr, i do not go beyond f11.0 unless i absolutely have to. also if you compare dof charts at the same fstop with the same lens you are getting the same dof with a c sensor dslr as you would with a film or FF dslr but at fstop more. roughly the c sensor dslr is giuving you the same dof at f11.0 at the FF/film dslr is at f16.0. so you are getting more dof than you think you have. further the max sharpness of a lens certainly is not its max closed down point. it is going to be in the range from f8-11. maybe edging towards f5.6 for some lenses. so even if diffraction did not exist, when you close the lens down you get more dof but you are also going away from the fstop of max sharpness. remember the old saying "f8.0 and be there". f8 was not chosen in that saying for no reason.<br>

for almost all my shots, unless i want less dof for some reason, i try to keep the fstop between f8.0 and f11.0.<br>

note the following- with a 1.5 c sensor dslr and a 35mm lens at f11.0 focused at 18feet, everything from 8.97feet to infinity will be in focus. so why would you want more dof, and a higher fstop, for landscapes for? the info was from this website-http://www.dofmaster.com/dofjs.html</p>

 

[steven_f1]

<blockquote>

<p>For example, with a 55mm macro lens f/32 is noticeably softer. But f/32 on my 300mm telephoto is still usable. (Same subjects, same distances or compositions.)</p>

</blockquote>

<p>At F32 the 55mm lens will have an aperture opening of 1.7mm. The 300 at F32 will have aperture opening of 9.4mm. This means that diffraction is more significant on the 55mm than it is on the 300mm. </p>

 

<blockquote>

<p>Another 300mm telephoto I tried in a local shop was significantly softer at f/22 than my 300mm.</p>

</blockquote>

<p>On that lens at F22 the aperture is 13.6mm. In a camera store you typically have less light and typically I find that I cannot get shutter speeds high enough to avoid camera shake. On a APS-C camera you need a minimum shutter speed of about 1/450 of a second to avoid camera shake. Or you are seeing quality differences in the lenses. For the 300mm lens to have the same diffraction as the 55mm at F32 you need to have the 300mm lens set to an aperture of F176. Diffraction is typically not an issue with telephoto lenses. Subject motion, focus accuracy, and camera shake are serious issues with telephoto lenses but are not so bad with wide angle lenses.</p>

 

<blockquote>

<p>maybe it's my inexperience but i don't think i've noticed much diffraction in my images?</p>

</blockquote>

<p>In my experience with landscape photography diffraction has not had a noticeable effect on my photos. You are not alone in not noticing the effect of diffraction. In my experience focusing errors and camera shake have a far greater effect. I have compared two photos (one on a tripod and one without) with a wide angle lens at greater than 1/100 of a second shutter speed. The picture with the camera on a tripod was significantly sharper than the one that was hand held. I had a decade of experience in photography before I learned about diffraction. However now that I know about it I don't use narrow apertures if I don't need the depth of field. </p>

<p>The effect of diffraction is typically small and not noticeable in most images. However if you are taking a close up of a subject with a lot of fine texture you are more likely to see it. Somewhere on the web I saw a series of close up pictures of fabric. At narrow apertures you couldn't clearly make out the individual fibers. At wider apertures you could see all the fibers. If the photo has details close to the size of diffraction you are more likely to see an effect. </p>

 

<blockquote>

<p>note the following- with a 1.5 c sensor dslr and a 35mm lens at f11.0 focused at 18feet, everything from 8.97feet to infinity will be in focus. so why would you want more dof, and a higher fstop, for landscapes for?</p>

</blockquote>

<p>Frequently in landscape photography you get more dramatic photos when you have something close to the lens and maintain objects at infinity in focus. For me that typically means something within 2 feet of the lens. Frequently F11 does not give me enough depth of field</p>

<p>My advise is keep the following in mind:</p>

<ul>

<li>Diffraction is typically worse with wide angle lenses set to narrow apertures.</li>

<li>Diffraction is mostly noticeable in photos of subjects with fine texture.</li>

<li>Focus accuracy, camera shake, and subject motion typically have a much greater effect than diffraction.</li>

<li>Diffraction will be more noticeable in large prints than it will be with small prints.</li>

</ul>

<p>If you keep this in mind don't worry about using narrow apertures unless you are taking a photo where diffraction does matter or where you don't need the maximum depth of field. Since you have a DSLR I encourage you to do some test photography to determine how much of an effect it has. Use a wide angle lens with a high shutter speed and take a number of photos at several aperture settings. Make sure focus is correct and subject motion are not present when you do so. That way you will be able to see what effect it has and will be able to judge if it is something to worry about. </p>

[lex_jenkins]

<p>FWIW, I always test lenses from a tripod in camera stores. Camera shake wasn't an issue. But your points are very good. The main difference between the two 300mm lenses was optical quality (different Nikkor designs).</p>

[JDMvW]

<p>Stephen has got it, I think. This is not the least of the reasons the old f/64 school <em>could</em> stop down so far without damaging IQ. They were working with large lenses and f/64 was still a reasonably large hole in absolute terms (so the diffraction from the edge was a low percentage of the image).</p>

[dave_s]

<p>Sorry to rain on the parade, but the comparison of the 50mm lens and the 300mm lens draws an incorrect conclusion. The absolute (millimeter) size of the two apertures at f/32 is different, but the rear pupil of the 300mm is six times farther away. This leads to the diffraction effect being magnified correspondingly at the film plane. Macro workers know this well, because their effective aperture diminishes when they hang the lens out on a bellows, without even touching the aperture ring.</p>

<p>The reason Edward Weston could use f/64 on his view camera was that he was exposing a big film sheet, and contact printing. (This is the same reason he could work with a $5 second-hand Rapid Rectilinear lens.) At f/64, diffraction limits resolution to about 25 lines per mm on film, which is plenty for contact prints or modest enlargements from an 8 x 10 negative. </p>

<p>To get back to Donald's OP, the advice to maintain a minimum aperture size is right on. Your lenses probably work best between f/4 and f/8, but you won't notice much degradation at f/11, or f/16 on 35mm. Eschew 22. </p>

[JDMvW]

<p>But Dave, regardless of distance, isn't it the <strong><em>absolute</em> </strong> diameter of the hole in relation to the <strong><em>wavelength</em> </strong> of the light that makes the diffraction a relatively large part of the image? That is, it is an <em>interference</em> effect?</p>

<p>Yes, the effective aperture of a lens in terms of exposure is "smaller" when out on a bellows, but the aperture of f/16 marked on a 300mm lens lets through the same amount of light as f/16 on a 50mm lens. Otherwise, we wouldn't be able to meter exposure without knowing the focal length of a lens.</p>

<p>I'm not at all sure, I confess, which of us is "whistling Dixie", but I don't think that the distance of the aperture from the film plane really has anything to do with diffraction, as such.</p>

<p>I'm not arguing, and I really want to know what's right here. I've worked on this problem for a long time; no, actually, I've <em>worried</em> about this question for a long time, but unfortunately that's not the same thing (Charlie Brown's little sister).</p>

[james_ashby]

<p>The distance of the aperture from the film plane doesn't affect the diffraction but it does increase the distance from normal that the light travels, as the light 'bends' round the edge of the blades it travels in a straight line at the same angle, the further from the film plane the blade is the further from normal the light travels, trigonometery etc etc<br>

Thats a bit physics'y for a photo phorum though.<br>

f/11 is usually pretty good for landscape stuff, just don't expect your feet and the horizon to be in focus (unless you have a tilt/shift device)</p>