How Lens Focal Lengths Will Affect Background Compression

Published May 17, 2016


When I decided to write about the “perfect” portrait lens, I really thought I had a grasp on what I was going to talk about. As a Canon shooter, I imagined that I would spend all the time talking about the sharpness of the Canon 135mm f/2.0, or the amazing bokeh of the Canon 85mm f/1.2L II.

But only a few days into working on this project did I realize how much we talk about models of lenses, but omit an understanding of lens optics themselves. Like many photographers, I have lost track of the time I have spent reading, and watching lens reviews. Then saving up money to buy a certain lens, but not asking the most important question, will this focal length help me in my photography.

And I am not alone in this problem. At Shooting the West, a photo workshop and symposium in Nevada, I spoke about and demonstrated different focal lengths for a room full of photographers. In this lecture, I found that many photographers, regardless of experience, really didn’t understand why they are reaching into their camera bag for one focal length over another. Selecting a lens just more than getting closer or further away from your subject, or how soft the bokeh of a lens can be.

So I am not going to waste your time with pixel peeping, or bokeh comparisons. I am going to introduce the concepts of lens focal length and how it relates to compression, distortion and why this should be the first thing you think about when reaching into the camera bag, or purchasing a lens.

The first thing I think about when selecting a lens is compression, or how much of the background I am going to see behind my subject. To test lens compression, in the case of portrait photography, I asked my good friend, and go to model, Travis Stewart to stand in the same place as I took several pictures of him. For each shot, I would change the focal length while keeping him composed in the same space of the frame.

16mm-Distortion-Test-Small 24mm-Distortion-Test-Small 35mm-Distortion-Test-Small

I started with wider-angle lenses first and then moved up to more telephoto. While the 16mm, 24mm, and 35mm focal lengths are not thought of as portrait lenses, look at how much of the background is displayed at these ranges. Travis is still present in the frame, but if the portrait was also about the area or place the subject lived or worked in, these focal lengths really show off the location and the subject at the same time. Of course, there is distortion at these focal lengths, which we will get to later, but even at 24mm, the distortion is not so radical that you couldn’t live with it, or just make a better composition, say full length.

50mm-Lens-Distortion-Test-Small 70mm-Lens-Distortion-Test-Small

At 50mm and 70mm there is a very nice balance of showing off a large part of the landscape, but making Travis really stand out in the frame. Still good focal lengths to use when you need to showcase that background.

100mm-Lens-Distortion-Test-Small 135mm-Lens-Distortion-Test-Small 200mm-Lens-Distortion-Test-Small

After 100mm two things start to happen, first, the compression of the background is so great, that we have lost the mountain range, and now we are just seeing the hills directly behind Travis. And at 135mm and 200mm not only do only see the hills, but also those hills have moved right behind Travis. As my focal length become more telephoto, it looks like those hills are just over Travis’ shoulder, when in fact they were at least a half a mile away, if not more.

Another thing to look at in these shots is image distortion. With the 16 and 24mm, we see serious distortion in Travis’ face. I took another set of images for more of a headshot composition, which really shows off this distortion. It is interesting that Travis’ nose does get larger because of the distortion while his body and face are narrower. As we get to 50 and 70mm that distortion starts to go away, and the nose is back to a normal size, and all those hours in the gym are showing off.

Most portrait photographers avoid wide-angle lenses when doing portraits, because of this very distortion, but as a fun test, I had Travis hold out his water bottle to show how distortion could aid you with certain subjects. At 16mm, the water bottle is huge, and is the subject of the photo, but when I took the exact same picture at 200mm, the water bottle is now much smaller and lacks punch. What if you are hired by a sports team or athletic apparel company, and they want a portrait of a pitcher or quarterback holding a ball out towards camera? Shooting the wrong focal length could put emphasis on the wrong subject. A wide, or at least wider-angle lens would make that ball larger and make it stand out from the subject. A telephoto lens would make the image more about the athlete holding the ball.


And of course, there is the relationship of the depth of field to focal lengths. Notice in all of these photos, as the focal range increases, the depth of field gets shallower. All of the shots were taken at f/10. I do want to point out how close Travis was to the camera, which has a significant impact on the depth of field as well. Even at 16mm, the background is out of focus because of the distance of subject to the camera.

While much of this information was not news to me, it was interesting to put a subject in one location and see how the focal lengths not only impacted distortion, and depth of field, but also the environment I am working in. Right now my go-to lenses for portrait work are 35mm and 100mm primes lenses for this very reason.

It is these concepts you need to think about when purchasing and using lenses. If you are working in a small studio, or your location is not great, then a telephoto lens will not only reduce distortion, and give you a shallow depth of field, but it will also compress that background, so less of it is present. But if your location is on the assembly line of a factory and your subject is the CEO, you probably want to find a focal range, like 35-70mm, that will both showcase your subject and their location.

Lens data charts are great, and comparing the bokeh of one lens to another makes for great information, but when it comes to your work, make sure you are telling your story in the right voice.


M.d. Welch

Guest Contributor

Author: M. d. Welch

Posted in Equipment
  • obican

    Statement is still false. Changing focal length does not affect your position, you have to manually walk back and forth, you may still choose not to.

    Whole arsenal of focal lengths is called a zoom lens.

  • Stretch too much? You mean wide angle distortion? Surely that’s just an issue with the lens and can get some lenses with less distortion? Or do you mean perspective? Plenty of arch photographers get by with ultra wide or tilt shift wides, better still tilt shift and pan stiching.
    Besides that still doesn’t change the fact that in ordinary shooting focal length affects the photogs position therefore perspective and dof. If you are lucky enough to have a whole arsenal of lenses then great but most people don’t. Therefore they choose a lens for the shot they want and stand in a positon they want, not choose a standing position and then the lens.
    By that that logic 99% of the time lens focal length affects perspective and dof because it affects your position relative to background and subject. Therefore the statement is pretty much true. People like to jump up and down at perceived errors.

  • obican

    It may be easier to explain at first but it leads to more complicated situations. I have a client who demands I shoot everything with a narrower lens because my wide angle lens makes the interior stretch too much. Can’t really do it because I’m backed off as much as I can and simply chose the narrowest lens which gets the whole interior in. Many people think if you simply change the focal length, everything will compress or decompress.

    Photographer should choose where to stand and choose the focal length after that.

  • Although true,in practice focal length does affect compression because choice of lens affects where the photographer will stand, thus it indirectly changes compression/perspective. Its just easier to explain it as focal length changes compression /perspective no? For me its similar to sensor size and dof debate, what you shoot with determines your distance to subject which affects dof

  • Scott

    Right – as I said in my comment to my comment, if you want to fill the frame with the same subject – you need to step back and use a longer focal length. Given the same subject size in the frame, which needs a longer focal length, you will get a compressed background.

    I noted in my original comment that the author stepped back.

  • obican

    No, it will be compressed because you are stepping back.

    If there is a McDonalds way behind you, would you say getting close to McDonalds does compress the background?

  • Scott

    Correct, but if you want to fill a frame with a subject at a given focal length- the perspective will be different because you ARE changing perspective, and the background will be compressed at longer focal lengths.

  • obican

    Yeah but that’s another story 😉

  • S.Yu

    The main issue right now is that you can’t shoot at 16mm and get a 200mm crop and get anything more than a thumbnail.

  • In the wide angle shots not only is the camera closer to the subject, it is positioned lower as well. There is visibility under the chin and the height relation between eyes and ears show the perspective as well as the person’s gaze is looking at a higher level above the camera although hes is looking straight forward horizontally.
    As the lens changes and the camera moved further back it also moved higher, about the level of the eyes. The subject now looks straight at the camera, although he has not changed angle. Again the level of ears and eyes show the perspective as well as the background hills becoming higher, whereas in the wide angle shots the sky was there.

    There should be a third set of pictures from the side, depicting the relationship and distance between subject and camera position and height.

  • obican

    You are wrong, the effect is there only because the photographer stepped back.

    Had he his 16mm lens on, where he was standing for the 200mm shot, the shot would exactly the same when cropped. Assuming the 16mm is not a fisheye of course.

  • Scott

    Focal length absolutely affects background compression. Perspective does not change, but compression comes with longer focal lengths. Yes, the author obviously stepped back as focal length got longer, if he hadn’t, at 200mm you would have only seen his nose in the first set photos. But you will not, and cannot get background compression with a 16mm.

  • l_d_allan

    Interesting and helpful. Thanks. I don’t recall seeing humans plus background treated this way. I included an article link to a DPR question about portraits of a group of 20.

    Also interesting … what happens if the human is at or close to the side of the image .. perhaps with the arms spread out so the fingers one one hand are at the edge, and the fingers of the other are at the center of the sensor frame?

    The article title mentions “background compression” but the article also dealt with “foreground apparent distortion”, which seems a somewhat different issue.

    I’d not pay much attention to the purist with their “panties in a bunch” who fuss about positioning of the camera, subject, and background. They are technically correct but downplay cropping. I think the point of the article was real-world visualization of what happens. And yes, many modern cameras have plenty of mpx for lots of cropping.

    It might help to have small “stick figures” showing where these elements were, to appease the purists.

    Yes, I liked this less technical article, and would enjoy more. And I’d also enjoy more of RC’s geeky articles. More LR articles … lots more !!!!

  • Greg Dunn

    Exactly – this is something which was illustrated extremely well in a 1970s issue of Popular Photography. It’s the relative position of the camera, the subject, and the background. The lens (as long as it’s rectilinear) has NO effect.

    Incidentally, using “Effect” as a verb in the blog title means “to cause” or “to bring about” whereas “Affect” means to change or alter. I suggest that “Affect” is the correct word for this usage.

  • SpecialMan

    Awesome. More like this please.

  • BernhardAS

    Thank for your post. I always enjoy watching this kind of examples carefully.
    Let me add one thought: going through the pictures the 50mm is exactly the point where motive and background are for me best in balance.

  • Kristian Wannebo

    Just in case you haven’t already planned something like this, I would like to offer the following suggestions.
    And I am certain you would explain much more understandably!

    I think many of your readers would appreciate a follow-up with similar examples of aperture choice.

    Perhaps you have time to reshoot the portrait series with a (somewhat) distant background with some structure and some highlights making noticeable bokehs:

    1) Keeping the bokehs of the background the same size in relation to the image of the person – and so also to the final photo if you frame the person similarly in all cases.
    This could be summed up (to a first approximation not considering differences in the way lenses render) as keeping

    (aperture diameter)/(size of focused subject) = constant

    over different choices of sensor size, focal length and photographing distance – regardless of how much space you frame around the main subject.

    2) Keeping the blur of the background a constant proportion of the size of it’s details. (E.g. a forest with the foliage a green haze and the stems similarly recognizable in all photos. Bokehs would then of course increase proportionately when zooming in.)
    This could be summed up (to a first approximation not considering differences in the way lenses render) as keeping

    (aperture diameter)/(distance lens – focused subject) = constant

    over different choices of sensor size, focal length and photographing distance – regardless of viewing angle, so even if you crop.

    3) I would also suggest a reminder of the more well known approximate rule that

    keeping the aperture diameter constant
    gives the same DOF

    over different camera choices for the same motif in the same viewing angle, even if you then need to crop to achieve that viewing angle.

    And when zooming from a fixed standpoint

    (aperture diameter)*(focal length)/(sensor size) = constant

    gives the same DOF (defined as a given maximum size of the blur circles in a final photo of a given size).
    For close ups, of course, use distance sensor-lens instead of focal length.

    Photographers, naturally, seldom have reason to think about this, but some of these rules of thumb can aid a budding photographer in choosing sensor size and speed of lenses to suit the kind of photos he wants to shoot.

    The well known math. behind:
    The lens formula ( 1/d + 1/D = 1/f ) gives

    p = ( a*d/(2s) ) * (B-F)/(B*F)
    p ~ ( a*f/(2s) ) * (B-F)/(B*F) for D>>f

    p = ( a*d/(sD) ) * (B-D)/B
    p ~ ( a*f/(sD) ) for B>>D>>f

    a = aperture diameter
    s = sensor size (linear)
    ( or, more correctly, size of the used part of the sensor in case of cropping in post)
    p = (diameter of blur circle)/(sensor size)
    f = focal length
    d = distance sensor – lens
    ~ f (except for close ups)
    D = distance lens – focused subject
    F = distance lens – front end of DOF
    B = distance lens – back end of DOF

  • DrJon

    I’d say firstly that it’s the FoV losing the background not compression and second that perspective is the key thing to consider, as how the person will look depends on how far away you stand, so then you need a lens to get the framing you want when standing there.

  • Omesh Singh

    There is a difference between perspective, projection, keystone and distortion.

  • almeich

    Yes, you are right and this cannot be repeated often enough. Perspective does not change with lens focal length, only the field of view does. To change perspective you need to move. As the photographer did in this case.

  • obican

    It is not the focal length that affects the background compression, it’s where the photographer stands in relation to the subject and background. You could’ve taken the 200mm lens out in the last shot and put the 16mm and the image would be exactly the same after the crop.

  • asad137

    These photos illustrate the concept really well.

    However, I would be careful to distinguish the distortion you’re referring to in these photos, perspective distortion, from the “distortion” you often see talked about in lens reviews, geometric distortion, aka the barrel or pincushion distortion of straight lines into curved lines. This geometric distortion comes from lens’ optical design.

    Perspective distortion, which is what makes portraits taken with wide lenses look so funny and what this post is talking about, comes simply from the fact that things that are closer to us appear larger. In photographs, it arises *only* from camera-to-subject distance and really has nothing to do directly with focal length. If you had a sensor and lens (and universe) with infinite resolution and you took the shot of your model from the same distance as you took your 200mm shot, if you cropped in to to the same framing as the 200mm shot it would have exactly the same perspective distortion.

    The reason that people think wide angle lenses cause distortion in portraits is that in order to fill the frame with your subject with a wide angle lens, you need to shoot from very close — and so the nose, which is significantly closer to the camera, looks larger than the ears, which are further away.

    Or, another way to put it is that the distance from the nose to the ears is a relatively much larger fraction of the camera-to-subject distance when filling the frame and using a wide angle lens than when using a telephoto lens.

Follow on Feedly