General Lens Distortion

:D
 
Last edited:
Its not really the lens focal length but the working distance, at least I think. So the 35mm has a FOV of a 50mm lens - so you would be further away (maybe) to get the same framing as a full frame with a 35mm lens. So, the distortion may not be there because you aren't as close to your subject. I may be wrong.....about all of it.


This quote is from a post about the perfect portrait lens and how a book showed a picture taken with a 50mm lens and there was distortion compare to a picture take. With a longer lens (around 100mm).

Dwig said:
You must keep in mind that it is not the lens focal length that causes this. Its the working distance. In order to get similar framing, he shot the example pictures at different distances and its the differing distances that produces the differing perspective.

As others have pointed out, you must always take the format of the camera into consideration. Any mention of focal length without information about format is meaningless. If you are reading a book discussing 35mm film usage and you are shooting a different format, such as that of a "crop sensor" DSLR, you need to "translate" the focal lengths to equivalent focal lengths that yield the same field of view on your format that the specified focal lengths did on the author's.

When it comes to perspective, whether discussing landscape, architectural, or portrait photography, you must keep in mind that its working distances that affect the perspective and not the focal lengths themselves. For portraits, you generally get the best perspective when shooting between 6-10ft (2-3m). The proper focal length is the one that gives you the framing you desire (face only, head & shoulders, 3/4's length, or full length) at some distance in that range. The looser the desired framing the shorter the focal length you want to use. The tighter, the longer.

Also, in the comments of that link - someone said to take a picture with 3 different focal lengths (like you did) at the same distance with each one, 5-6 ft away, then crop the image so you have the same framing of the face and the images would be identical. Basically it's to prove that it's the distance not the focal length - I don't know if it's true but.....
 
OOoooh Ok. So maybe not with 3 lenses but one of my zooms, might actually get nearly identical images that way :D

Distance makes sense. Because the curvature of the lens glass right?

Ever read too much about something and then it all jumbles?
 
Stop thinking about this as distortion and lens distortion. This is about lens/subject distance. Yes, the focal length is a player because without the wide angle view you couldn't get that close and frame a head and shoulders and without the narrower field of view of the long lens you couldn't back up enough and keep a head and shoulders portrait. But the variation in appearance you're experiencing here is a function of variation in relative significant distances.

The term wide angle distortion is commonly used but it can cause a lot of confusion. Lenses have barrel distortion not wide angle distortion. Extreme focal lengths allow us to position the camera at distances that don't work for the fixed lenses in our heads. Put you eye eight inches from your daughter's nose and she'll look a lot like the shot from the wider lens. The fact that our camera's can view the world in ways that we can't isn't a distortion.

There's another factor involved in this. The reproduction or scale of the viewed image and the distance from the viewed image. This will also alter your perception of the degree to which the subject's 3D proportions are different than your common experience. What people frequently call lens distortion is often due to scaling and print/image viewing distance. Take a photo of your daughter and reproduce it to size X1 and then view the image from the same distance as the camera lens when the photo was taken. As you begin to alter this ratio, eg. move closer to image, reproduce image smaller, you see the proportions of your daughter's face change. If you re-size the image and move closer and keep the ratio constant, the preceived proprtions will stay the same.

Joe
 
Its not really the lens focal length but the working distance, at least I think. So the 35mm has a FOV of a 50mm lens - so you would be further away (maybe) to get the same framing as a full frame with a 35mm lens. So, the distortion may not be there because you aren't as close to your subject. I may be wrong.....about all of it.


This quote is from a post about the perfect portrait lens and how a book showed a picture taken with a 50mm lens and there was distortion compare to a picture take. With a longer lens (around 100mm).

Dwig said:
You must keep in mind that it is not the lens focal length that causes this. Its the working distance. In order to get similar framing, he shot the example pictures at different distances and its the differing distances that produces the differing perspective.

As others have pointed out, you must always take the format of the camera into consideration. Any mention of focal length without information about format is meaningless. If you are reading a book discussing 35mm film usage and you are shooting a different format, such as that of a "crop sensor" DSLR, you need to "translate" the focal lengths to equivalent focal lengths that yield the same field of view on your format that the specified focal lengths did on the author's.

When it comes to perspective, whether discussing landscape, architectural, or portrait photography, you must keep in mind that its working distances that affect the perspective and not the focal lengths themselves. For portraits, you generally get the best perspective when shooting between 6-10ft (2-3m). The proper focal length is the one that gives you the framing you desire (face only, head & shoulders, 3/4's length, or full length) at some distance in that range. The looser the desired framing the shorter the focal length you want to use. The tighter, the longer.

Also, in the comments of that link - someone said to take a picture with 3 different focal lengths (like you did) at the same distance with each one, 5-6 ft away, then crop the image so you have the same framing of the face and the images would be identical. Basically it's to prove that it's the distance not the focal length - I don't know if it's true but.....

Megan -- got it right. You beat me to it while I was typing and uploading today's photos.

This is about camera/subject distance. However, extreme focal length lenses are unique critters in that our eyes aren't physically capable of seeing the world as they do.

Take this a step further. If you run a test and sample thousands of people as they look at photos and determine at what distance they seem to comfortably settle on to view the photo you'll discover that people back up about double the long side of the image. With that info in hand we can do the math to determine precisely what lens focal length will render a subject such that viewing the print/image will show the proportions of the subject as seen from the lens position. With a full frame 35mm camera that would be a 75mm lens.

Joe
 
Dwig said:
When it comes to perspective, whether discussing landscape, architectural, or portrait photography, you must keep in mind that its working distances that affect the perspective and not the focal lengths themselves. For portraits, you generally get the best perspective when shooting between 6-10ft (2-3m). The proper focal length is the one that gives you the framing you desire (face only, head & shoulders, 3/4's length, or full length) at some distance in that range. The looser the desired framing the shorter the focal length you want to use. The tighter, the longer.

There is no "best" perspective for portraits. There is only the perspective intended by the photographer. A friend of mine shoots professional fashion and he'd never get 6 -10 feet near a model. He shoots long lenses from 15 to 20 feet and even further. His open studio floor is over 50 feet long and I've seen him use it. On the other hand look at the portraits of A. Newman or A. Liebowitz. Look at Newman's Great British -- he's shooting a 90mm lens on a 4x5 right in their faces.

http://images.npg.org.uk/790_500/2/6/mw08426.jpg

Joe
 
Ysarex said:
Megan -- got it right. You beat me to it while I was typing and uploading today's photos.

This is about camera/subject distance. However, extreme focal length lenses are unique critters in that our eyes aren't physically capable of seeing the world as they do.

Take this a step further. If you run a test and sample thousands of people as they look at photos and determine at what distance they seem to comfortably settle on to view the photo you'll discover that people back up about double the long side of the image. With that info in hand we can do the math to determine precisely what lens focal length will render a subject such that viewing the print/image will show the proportions of the subject as seen from the lens position. With a full frame 35mm camera that would be a 75mm lens.

Joe

Glad I got something right!! ;)
 
wow.. two pages already... you guys have been busy! :)
 

Most reactions

New Topics

Back
Top