Diffraction

Often when we start out as photographers we hear more experienced folks talk about the importance of not stopping a lens down too far as to avoid the effects of diffraction.

We may find this confusing because from a composition standpoint we often want to achieve deep depth-of-field, especially with landscape photography. Stopping a lens down to achieve this degree of depth-of-field seems quite logical.

This begs the question whether we really need to care about diffraction. Some photographers are almost fanatical about avoiding the effects of diffraction, while others are not very concerned about it at all as their objective is to achieve an overall look with their composition. They don’t concern themselves with pixel peeping and detailed examinations of their images.

What is diffraction?
The full explanation of diffraction can be a highly technical exercise. Basically it is an optical effect which limits the total resolution of your images, regardless of how many mega-pixels your sensor may have. Diffraction happens when the light coming into your camera body begins to disperse or ‘diffract’ before it hits the sensor in your camera. This is caused by the light passing through a small hole, or aperture, in your lens. The amount of dispersion of the light landing on your camera’s sensor can cause softening of your image.

Does it affect all cameras identically?
The effects of light dispersion (diffraction) are affected by the size of the pixels on your camera’s sensor. Higher pixel density sensors will have smaller sized pixels so diffraction will be more noticeable than with a lower pixel density sensors which have larger pixels. For example, diffraction will become noticeable on a Nikon D800 at f/11 but it would not be apparent when shooting at this same aperture using a Nikon D3 or D700, even though all three cameras utilize full frame sensors.

Cameras with smaller sized sensors like Nikon 1, M4/3 and point and shoot models will suffer the effects of diffraction at lower f-stops when compared to cameras with larger sensors, assuming of course that the size of the pixels on those larger sensors is bigger than those found on the smaller sensors.

What does diffraction look like?
Rather than get into a lengthy technical discussion about diffraction I’ve taken some test images to help demonstrate the effect. These images were shot with a Nikon 1 V2 with the Nikon 1 18.5 mm f/1.8 prime lens. The camera was tripod mounted and the shutter was activated via remote.

I chose this lens for three reasons. First, it is one that is very commonly owned by many Nikon 1 users, secondly it is one of the sharpest Nikon 1 lenses, and finally it provides an equivalent filed-of-view of approximately 50 mm. This is one of the most common focal length lenses owned by photographers regardless of camera format. As is often the case with lenses, the 18.5 mm appears to be at its sharpest when shooting a couple of stops down from wide open.

The following is a series of out-of-camera jpeg images taken at f/4, f/5.6, f/8, f/11 and f/16. These are 100% crops taken from an area roughly in the centre of the frame.

Note: Click on images to enlarge.

As you look at each image pay special attention to the amount of detail and how crisp it appears to you. First let’s look at the f/4 image.

DSC_4553 f4 crop ooc

The next one was taken at f/5.6. If you examine it closely you’ll likely be hard pressed to notice any significant difference with the f/4 image.

DSC_4554 f5.6 crop ooc

Now an image at f/8. You will begin to see some softness in the image.

DSC_4555 f8 crop ooc

Now at f/11, it becomes more apparent.

DSC_4556 f11 crop ooc

And, finally at f/16 it is quite noticeable.

DSC_4557 f16 crop ooc

Is is possible to eliminate diffraction with post processing?
Let’s see what happens when using the RAW file from the f/16 image and run it through DxO’s auto lens corrections.

DSC_4557_DxO crop auto

Not much of a difference…let’s add some sharpening by taking Global to 1.20 and Detail to 70, then add some micro contrast to +10 to see what happens.

DSC_4557_DxO-1 dxo corrections

Let’s remind ourselves what the image looked like when shot at f/5.6 as an out-of-camera jpeg.

DSC_4554 f5.6 crop ooc

And finally, how would the image look when shot at f/5.6 with the same DxO adjustments made to it that we did with the f/16 file.

DSC_4554_DxO-1 f 5.6 corrections crop

As we look through these images in succession it becomes pretty clear that we can try to hide the effects of diffraction in post but we can’t eliminate them.

Should you care about the effects of diffraction?
Depending on subject matter and image use you may, or may not, find the level of image softness that is created by diffraction objectionable. As mentioned earlier, many photographers are quite prepared to have a bit of softness in their images as long as they can achieve the overall composition they desire. As we did in some of the previous images, they often apply some sharpness to their images to try to minimize the effects of diffraction.

Practical shooting considerations
It is critical that you understand your camera body and lenses in order to get the best results from them and to minimize the effects of diffraction.

It may be useful for you to do your own hands-on testing with your camera and lenses. Set up your gear on a tripod, lock the mirror up if you are shooting with a DSLR, and use a shutter release or shutter delay. Then, shoot a series of identical images with each of your lenses beginning with each of them fully open (i.e. the lowest f-stop number) then at each standard aperture setting as your lens may allow (e.g. f/1.4, f/2.0, f/2.8, f/4, f/5.6, f/8, etc.). By doing this you will learn where each of your lenses is at its sharpest, and also where diffraction begins to set in and how bad it gets.

This will lead you to some practical understanding of your gear. For example, when I shoot landscape images with my Nikon 1 gear I typically shoot at f/5.6 as I find this gives me the best combination of image sharpness and depth-of-field. For other types of subjects I occasionally shoot as high as f/8 but almost never go beyond that point.

As with most things photographic diffraction is one of those trade-offs that can happen. Being aware of its effects allows us to make an informed decision in terms of our choice of camera settings and balancing those with our image objectives.

My intent is to keep this photography blog advertising free. If you enjoyed this article and/or my website and would like to make a modest $10 donation through PayPal to support my work it would be most appreciated. You can use the Donate button below. Larger donations can be made to tom@tomstirr.com through PayPal.

Article and all images Copyright Thomas Stirr. All rights reserved. No use, duplication of any kind, or adaptation is allowed without written consent.

4 thoughts on “Diffraction”

  1. Hi Tom,

    as always a great written article, but as an physicist in the making I miss some points.

    Diffraction is a phenomenon which causes a limited angular resolution – in the result you need a bigger lens to get the same resolution for more distant objects. Well I dont want to get to deep here, but for diffraction “distance” is at least as important as the open lens diameter – a point you completely miss here. With distance and resolution of course focal length is also relevant – and also since the f-stop only gives a ratio of opening diameter and focal length, the f-stop itself is really not giving any information for diffraction.

    I think this can be best observed in macro photography – never seen any loss in sharpness even with the smallest possible aperture. In comparison I once made a few tests with my wall calendar (distance about 4-5 meters and was shocked how bad the sharpness got even at around f11 (I think I used the 10-30 or 30-110 on my V3).

    Your demonstration makes diffraction look like a small factor, but scientifically you demonstrate nothing and the results are way better then what I tested once. But well, actually I also never made a true measurement or any calculations I could present…^^ For this I would suggest to make pictures and compare them by opening diameter (focal length divided by f-number) and distance of your object. Here you get a real idea of diffraction and could also try the extreme; small aperture with high distance. Since this combination is very uncommon in photography you may be surprised how bad your pictures actually will get.

    1. Hi Max,

      Thanks for your comment and adding some additional information to the topic. I actually thought that the image at f/16 looked quite bad :-)…I guess we all have different viewpoints. I certainly agree that details at a distance will certainly suffer even more. Some of this can be due to the softness of a zoom lens at maximum extension as well. I once mistakenly took a distant shot of a hawk in flight at f/16 with my D800 and you’re right…it looked absolutely horrid.

      Tom

  2. Such an excellent article with images demonstrating diffraction!

    Thank you for setting out the parameters of your test. I like how you progressed through different f-stops and let us note what was (or wasn’t different) between the images. I really appreciate your going the extra distance in showing what happens when one tries to compensate for diffraction in the f16 image.

    I never did realize that at f16 the diffraction effects could overwhelm the DOF, compared to a much lower aperture.

    I shoot w/ a Nikon APS sensor, so I now need to test my camera/lenses with a different target subject, rather than a straight-on one. I’ll be shooting a scene with more DOF than the bark you shot. And within my DOF range, I’ll be looking at my DOF versus diffraction. Now that you’ve shown me the way, I now know what to look for.

    Again, Thank You.

    Wei Chong

Leave a Reply

Your email address will not be published. Required fields are marked *