DPR Forum

Welcome to the Friendly Aisles!
DPRF is a photography forum with people from all over the world freely sharing their knowledge and love of photography. Everybody is welcome, from beginners to the experienced professional. Whether it is Medium Format, fullframe, APS-C, MFT or smaller formats. Digital or film. DPRF is a forum for everybody and for every format.
Enjoy this modern, easy to use software. Look also at our Reviews & Gallery!

Does 16 crop factor on digital actually change focal length of lens

The answer to Mr Marc Williams's question is yes. Indeed, the distortion or 20 mm "pull" will disappear straight away.
You can even crop it to equal a 600mm or 1200mm tele effect, provided you do enough cropping and the small remain of the photo have enough resolution for you to see the ''tele'' effect.
 
But surely, if it is the edge which is cropped,as Marc has suggested, the distortion in the original picture from the extreme wideangle effect, will remain.
 
The wide angle effect will be lost with cropping. We have gone though about this question many times last few days and I am not explaining again. If you really wish to understand this you might review our previous postings. It is easiest to understand this if you understand the angle of view of an image.
 
I have followed the previous posts Chi.
What is now suggested is that the distortion will remain since it is the distorted edge of the picture which would be emphasised and even magnified by the crop. This is not the same as cropping an area which has no distortion and I don't think that it has already been covered. Apologies if I have missed it.
 
Strictly,I wish to see how Marc made his crop,like in what shape,in what way. I will answer your question without asking him to rephrase.
I hope you understand the term angle of view of an image or photo. The AOV of the diagonal of the photo is the diagonal AOV of the lens which produces the photo,so are the vertical AOV and horizontal AOV.
There is also an AOV between any two points in the photo. The AOV from midpoint to the tip of one corner is half of the diagonal AOV.
In a photo produced by a wide angled lens, the diagonal AOV is greater than the normal angle of the eye (about 42 degrees).
Imagine you have a pyramid with the photo as its base and all 4 lateral sides are equal. The two diagonal AOVs,two horizontal and two vertical AOVs form the apex of the pyramid.
The diagonal AOV of the photo made by Marc is about 92 degrees which is a lot bigger than the normal angle of view. No wonder why the most pronounced distortion is seen along the diagonals.
Any two points in the photo have an AOV,which goes from one point to the apex of the pyramid and back to the other point. You can predict whether distortion or wide angled effect is going to happen by looking at the AOV of two points. If AOV is much bigger than the normal angle,distortion and wide angle effect will happened between the points.
Distortion will not happen between two points if the AOV of these points does not exceed the normal angle. Distortion does not occur at a single point,at least one line is involved.
If you wish to know if cropping affects the distortion. Draw a line along the distortion.If cropping intersect with the line,distortion will be affected.
Cropping in general reduces the DAOV,VAOV HAOV and distortion.
 
Gosh Folks,

We no sooner get one issue more or less settled when someone comes along and introduces a new variable. Well it's worth discussing compression and distortion.

First of all, compression and distortion are not the same thing. Every lens compresses the image information to get from a large object FOV down to the few mm. dimensions of the film. (The exception being a macro lens like the 60mm S Planar when focused to produce an image at 1:1). Assuming the same vantage point, the greater the AOV of the lens, the more compression it must produce. The ideal lens produces this compression perfectly uniformly from the center of the image to the periphery and in the process causes no distortion. Unfortunately such ideal lenses do not exist. Barrel and pincushion distortion exist in every lens and are usually more prominent in wide angle lenses; the effect is most pronounced at the edges of the image. (The distortion produced by fish-eye lenses is an altogether separate subject.)

To the extent that such distortion is present at the edges of a wide angle image, it will certainly be reproduced when one crops and enlarges as suggested by Marc. In doing so there is no way to duplicate the image produced by a tele. Marc and John are quite right, but note they have changed the ground rules. Earlier we were discussion only enlarging and cropping the very center of a WA image, where distortion is minimal or nil.

Chi, as usual will be a little slow to get on board, but I believe he will. In his last post he speaks of the "normal" AOV, by which I assume he means the AOV of the human eye. He seems to suggest that distortion can only occur when this normal angle is exceeded, and cannot occur when the AOV is less. Wrong on both counts. Just because a WA image gives more information than can be taken in by the human eye does not mean that the image is necessarily distorted. Distortion is produced by the lens; good ones produce less and poor ones produce more at any angle. Cropping cannot reduce distortion unless, of course, you simply crop out the distorted portion of the image.

And to all, a good night.
Mike.
 
> You added some clarity, Bluemsan!

Here's again as simple as I can write it a summation of where we are:

1. Image taken at same position with wide angle versus telephoto lens =20=

of subjects at some mid distance, say 10-50 ft.

a) Angle =F8 degrees subtended at the lens front element = central axis =20 is identical for all focal lengths as it only depends on the distance =20=

to the subject and the diameter of the detail being observed.

b) The perspective on the film plane is defined by angle = =F8 and is not =20 changed by cropping.

c) Only apparent field of view changes when the image = circle is not =20 fully used.

c) So, only apparent field of view and apparent angle of = view are =20 altered by cropping. What would be included in a human's field of view =20=

is the "normal" field of view for a lens, and that depends on the =20 proportion of the wider field of view of a wide angle lens image circle =20=

which is used by the film of CMOS/CCD sensor in the image circle of the =20=

lens.

d) One can take the center of the wide angle image and = enlarge it to =20 overly the image taken with a telephoto lens and they will be =20 geographically close to identical. However, they will differ in =20 contrast, depth of focus and 3D effect depending on lens quality and =20 aperture and ratio of diameter of object and distance to the object. =20 Thus difference in this quality will be more apparent at wider =20 apertures closer to the camera but vanished at infinity (eg if taken of =20=

the moon at f11).

2. Image taken at distances such the subject fills the frame with =20 either a wide angle lens or telephoto lens

a) The image taken with the wide angle lens will be very = close to =20 match the field of view captured on the sensor with the telephoto lens.

b) The wide angle lens image will cover the same amount = of the =20 subject but will have an angle =DF which is much larger than our = previous =20 angle =F8 where BOTH images were taken from the SAME distance from the =20=

subject.

c) The larger angle =DF means that the perspective of = the image taken =20 with the wide angle lens, is this time distorted compared to the =20 previous image pair, where the camera to subject distance was =20 identical.

d) This distortion caused by the new and larger angle =DF = cannot be =20 altered by cropping the image.

At close and medium distances, the image with the wide angle lens will =20=

always have less compression of sequential subjects at progressive =20 distance from the camera than images taken with the telephoto lens, =20 which will progressively compress more distant subjects.

At infinity, the compression effects will tend to equalize and pictures =20=

of the moon will have closely similar compression effects with all our =20=

lenses.

Wide angle lenses tend to have greater depth of field and less =20 compression and therefore look more 3D. Telephoto lenses have a =20 narrower plane of focus, include less information about the =20 environment, have less adjacent reference point for depth signals to =20 the brain, and so a perceived as flatter.

In all these conditions, cropping does not alter distortions, depth of =20=

field, bokeh, contrast or perspective, just "apparent" field of view. =20=

If one crops a wide angle image to the real field of view of a human, =20=

then that portion of the image plane may masquerade as the normal field =20=

of view, however, the perspective will always be dependent on the angle =20=

=F8 that any subject makes with the camera based on merely the subjects =20=

diameter and distance form the lens.

Asher
 
(reposted for clarity without formatting symbols, LOL!)

Distinguishing apparent FOV from optical persective is needed to understand what is going on. Also what signals to the brain 3D info (beside color, mist and blur).

Here's again as simple as I can write it a summation of where we are:

1. Image taken at same position with wide angle versus telephoto lens of subjects at some mid distance, say 10-50 ft.

a) Angle ø degrees subtended at the lens front element at the central axis is identical for all focal lengths as it only depends on the ratio of the distance to the subject and the diameter of the detail being observed.

b) The perspective on the film plane is defined by angle ø and is never changed by cropping.

c) Only the apparent field of view changes when the image circle is not fully used.

c) So, only the apparent field of view and apparent angle of view are altered by cropping. What would be included in a human's field of view is the "normal" field of view for a lens. Cropping an image mimicks this, but doesn't change the perspective of how 3D images are rendered on the flat image plane, since the capture angle ø is unchanged!

d) One can take the center of the wide angle image and enlarge it to overlay the image taken with a telephoto lens and they will be geographically close to identical. However, they will differ in contrast, depth of focus and 3D effect depending on lens quality and aperture and ratio of diameter of object and distance to the
object.

Thus difference in this quality will be more apparent at
wider apertures closer to the camera but will vanish at infinity (eg images taken of the moon at f11).

2. Image taken at distances such the subject fills the frame with
either a wide angle lens or telephoto lens

a) The image taken with the wide angle lens will be very close to
match the field of view captured on the sensor with the telephoto lens.

b) The wide angle lens image will cover the same amount of the subject but will have an angle ø which is much larger than our previous angle ø where BOTH images were taken from the SAME
distance from the subject.

c) The larger angle ø means that the perspective of the image
taken with the wide angle lens, is now distorted compared to
the previous image pair, where the camera to subject distances were identical.

d) This distortion caused by the new and larger angle ø and cannot be altered by cropping the image.

At close and medium distances, the image with the wide angle lens will always have more separation closer to the camera and less compression of sequential subjects at progressive distance from the camera than images taken with the telephoto lens.
The latter will progressively compress more distant subjects.

At infinity, the compression effects will tend to equalize and
pictures of the moon will have closely similar compression effects with all our lenses.

Wide angle lenses tend to have greater depth of field and less compression and therefore look more 3D.

Telephoto lenses have narrower plane of focus, include less information about the environment, have less adjacent reference point for depth signals to the brain, and so a perceived as flatter.

In all these conditions, cropping does not alter distortions, depth of field, bokeh, contrast or perspective, just "apparent" field of view.

If one crops a wide angle image to the real field of view of a human, then that portion of the image plane may masquerade as the normal field of view, however, the perspective will always be dependent on the angle ø that any subject makes with the camera based on merely the subjects diameter and distance form the lens.

Asher
 
Asher:

I am happy that you believe I have added some clarity. I still have some small reservations with your most recent post:

Your definition of the angle Ø is very different from AOV.

<<Angle ø degrees subtended at the lens front element at the central axis is identical for all focal lengths as it only depends on the ratio of the distance to the subject and the diameter of the detail being observed.>>

That's fine with me. I assume that "diameter of the detail" has the same meaning as what I call Field of View. However at (2b) you go on to say:

<<The wide angle lens image will cover the same amount of the subject but will have an angle ø which is much larger than our previous angle ø>>

That's true because you maintain the same diameter of detail (to use your term)or FOV (to use my term) but you shorten the distance to the object. That's simple geometry.

But you go on to say: the "perspective is distorted". I agree the perspective is different, but that is not the same as distorted. This difference is caused by the difference in AOV of the two lenses, as I have shown above.

<<Wide angle lenses tend to have greater depth of field and less compression and therefore look more 3D.>>

Greater dof and more 3D; yes indeed because of the additional information. But they need to compress more not less in order to get that extra information on the sensor. Of course that depends on how you define compression. I use the generally accepted definition: the reciprocal of the lens magnification factor at a specified object distance. To give you an ex&le, a 35mm lens at 10 meters has a compression factor of 285:1. At the same distance with a 135mm lens it is 73:1.

I hope this adds to the clarification.

Regards,
Mike..
 
A well known cinema technique (used to great psychological effect by such directorial masters as Alfred Hitchcock), has the 35mm motion camera dolly back while the lens zooms from the slight W/A or normal end of the zoom to the telephoto end, in effect keeping the foreground subject exactly the same size. The visual effect dramatically moves the background forward (compression) behind the subject, who apparently remains "normal" looking. The Hitchcock movie "Rebecca" was one of the first widely distributed movies to utilize this technique to great effect.
 
Back
Top