Depth of Field Revealed
I ran a contest this week, with the prize being a Think Tank Photo Urban Disguise 40 camera/computer bag. I initiated the contest based on a question about depth of field. It all started earlier in the week when I told you the story about my efforts trying to explain to a mentor-client how focal length has very little if any effect upon the depth of field. To help him understand this concept better I laid out three pencils on the floor, then lay down and shot two images. I shot each image at different focal lengths leaving the f/stop constant. When I reviewed my results I was surprised to see that the depth of field seemed to change noticeably. This perplexed me. I was sure that focal length did not affect depth of field, yet it sure seemed to here. I went home and did some research. I discovered that I had made one slight error: I did not keep the subject at the same magnification in each shot, i.e.– I did not physically move closer to the subject to make it the same size in the viewfinder.
To get my readers involved I decided to make a contest out of this whole mystery. On Monday I asked you this seemingly simple question: “You’re shooting at 18mm at f/4, then you decide to shoot the same subject with the same magnification (meaning the subject fills the frame in the same way) at 200mm at f/4. What happens to your depth of field?” Then the winner would be drawn from amongst the right answers.
So what is the right answer? Just what I suspected–Nothing— the depth of field is not affected by focal length by any real significance. There are people who are a lot smarter than I am who have put the math into this and have come up with the same conclusion. When all I did was to take my camera out and do a field test.
Here’s how I did my field test:
I took my Canon 5D MKIII and my 70-200 mm f/2.8 lens set the aperture to f/4 and focused on a street sign. I chose a sign that had several other signs behind it as a way show off the depth of field. After taking a photo at 200 mm I then switched lenses to my 17-40mm f/4 lens. Keeping the aperture at f/4 I tried to “zoom” the 17-40mm to 18mm–I ended up shooting at 19mm. And, wait for it–here is the important part, I physically moved the camera to the point where at 19mm focal length the sign was at the same magnification. In other words, I had the sign fill the same exact proportions of the viewfinder at 19 mm as it did at 200 mm.
The focus was on the signal caution sign. Here is the 200 mm image at full frame scaled down to 940 px.
The focus was still on the signal caution sign. Here is the 19mm image at full frame scaled down to 940 px.
After my field experiment, I opened up both images in Photoshop. Here’s where it gets complicated, at first I thought all I needed to do was open the two images in Photoshop and put them both in the same file on two different layers, then match the size of the two signs I had focused on. But, when I did this there still seemed to be a difference in the depth of field, or rather, the focus of the objects in the background in the two images. Then I remembered something Gavin had written in his comments earlier this week. He mentioned that depth of field is also contingent upon the size of the image you are viewing. Of course, this makes complete sense, especially when you think of an example of a billboard or hoarding. Billboards are printed at an insanely low resolution– sometimes as low as 25 dots per square inch, yet when viewed from a distance they look completely normal. Given this fact, I felt the best thing to do would be to take an object from the background in both images (I choose the intersection or road crossing sign) and try to match the magnifications in the file. But when I tried to enlarge the 19 mm image so that the background object matched the same object in the 200 mm image, the 19 mm image quality fell apart and rendered it basically unusable. I thought, “why not just do it the other way around?” Shrink the 200 mm image to match the 19 mm image. This gives a more accurate idea.
Clearly, the intersection sign in the two images match at the same magnification. No visible difference in the depth of field.
You can see by the image above that the intersection signs are almost identical. There is no real difference in focus, thus matching similarly the depth of field. I hope this helps clear up the confusion that you get more depth of field with a telephoto than you do with a wide angle. The truth is you dont. But you do get a perceived depth of field due to the image magnificaiton.
So who is the winner of the Urban Disguise 40? Congratulations to Jacob James for having the right answer and being fortunate enough to have his name drawn out of my Airport Express box. If you’d like to read more about the relationship between focal length and depth of field I’ve drawn together a few links for you below:
Luminous Landscape: Do Wide Angle Lenses Really Have Greater Depth of Field Than Telephotos?
Cambridge in Colour: TUTORIALS: DEPTH OF FIELD
Norman Koren: Understanding image sharpness part 6: Depth of field and diffraction
Stack Exchange- Photo: How does changing focal length affect depth of field?
Subscribe to my newsletter and be among the first to hear about new workshops.
Good job, guys. As soon as the first person posted this answer, I knew he was right. :). Great challenge!
Thanks for commenting Robert.
Great article Matt. Photo one and two are a great lesson on compression. Shows what happens to background when changing lenses.
Compression is a wonderful thing 😉 Thanks for reading.
Learnt something new today. Thanks.
Thanks for dropping by! Hope you will keep checking back!
Excellent article – thanks Matt – I have shared it on my FB page and also the FB page of the Photographic Society of New Zealand.
Shona, Thank you so much for sharing this post. You might find some other posts worth sharing as well. New Zealand is on my ‘bucket list’ of places to go shoot someday.
Nice article, but I think the conclusions are incorrect.
DOF is defined in terms of the final image viewed and the distance of that image from the human eye, so when you zoom in (either by cropping and magnifying later or using the digital zoom on-camera) you are in fact changing the picture even if the actual pixels haven’t changed. What once looked sharp now looks blurred, so the depth of field in the viewed image has in fact changed.
If, however, you are only interested in the DOF that affects the pixels – irrespective of what you do with the image afterward – this is quite another question (pixel peeping). This isn’t DOF!
Perhaps it should be, or perhaps we need another term, but you really need to look at the bigger picture 🙂