Pages: 1 ... 5 6 [7] 8   Go Down

Author Topic: This puzzling business of "35mm lens equivalent"  (Read 38154 times)

AlfSollund

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Re: This puzzling business of "35mm lens equivalent"
« Reply #120 on: January 20, 2012, 03:26:32 am »

Good point! If you don't point the camera in a different direction but use a different focal length of lens instead, have you not also taken a different photo?

Why would it makes sense to compare two photos that are very different in content and composition, as a result of the focal length of lenses used being different, but not make sense to compare two photos of precisely the same subject taken from the same distance, with the shot using the longer focal length being turned at a slight angle out of necessity?

Perhaps the definition should be changed along the lines, "Perspective in photographic images is affected by both distance to the subject and angle of view." ;D

If you point the camera in the same direction and change focal lenght you take the same photo with a different crop. You might also take a different photo due to different DoF and different real-world characteristics of different lenses but thats not whats being discussed.

The difference between two photos with different focal lengths is that tele one has a subset of of the content, with exactly equal composition and perspective as the wide (DoF might differ). And nobody else than the ones ignorant of perspective would want to compare these, since the result is given as most of us know.

If you change the angle of the camera you change the distance to subjects since its a 2-D representation of 3-D. You actually tilt the angle of the 2-D plane capturing the 3-D. Try this: point the camera directly towards a building so that vertical lines of building are parallel. Then slowly tilt the camera up or downwards. What happens? The lines start to not be parallel. This is not because the building are starting to tilt when you move the camera.

So you don't need to state "by both distance to the subject and angle of view" since changing the angle means changing the distance.

I'm sorry to say that you are arguing with the last 1000 years or so of knowledge on simple geometry  ;D.
« Last Edit: January 20, 2012, 03:28:29 am by AlfSollund »
Logged
-------
- If your're not telling a story with photo you're only adding noise -
http://alfsollund.com/

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: This puzzling business of "35mm lens equivalent"
« Reply #121 on: January 20, 2012, 03:31:32 am »

"Bear in mind that by turning slightly to the left in order to take the shot of the sign, I have not changed the distance from the camera to the sign and have therefore not stepped outside of that definition that perspective is only affected by distance to subject."

But you have, Ray, you have changed everything.

The entire concept of the thing is about a single shot. The moment you introduce movement, you might as well use a ciné camera and go the whole way. Or just a paintbrush.

The matter of equivalent focal lengths isn't about moving cameras, it's about difference or sameness within a single exposure. Anything else is another matter and irrelevant to the question being posed.

But you already knew that, you advocating devil!

;-)

Rob C

« Last Edit: January 20, 2012, 04:11:13 am by Rob C »
Logged

Jim Pascoe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1131
    • http://www.jimpascoe.co.uk
Re: This puzzling business of "35mm lens equivalent"
« Reply #122 on: January 20, 2012, 03:47:19 am »

Erik,
You're a very sensible and scientific sort of guy. Can you offer any proof that the perspective of a photographic image or composition is dependent only on the distance from the lens to the subject?

So far, in this thread the only proof that has been offered is the fact that converting a wide-angle shot to a narrow angle shot of the same effective focal length, through cropping, results in identical images.

Well that doesn't seem like proof to me, the fact that two lenses of equal focal length produce identical pictures when used from the same position. That's pretty obvious, isn't it.

The problem as I see it, is the absurdity of claiming that non-identical pictures can have identical perspective, as in the two examples attached.

The wide angle shot was taken with a 24mm lens. The image of the sign was taken with a 140mm lens equivalent. Whether I took the shot of the sign with an actual 140mm lens attached to my D700, or cropped the 24mm image to simulate a 140mm lens, possibly makes no difference, except in resolution, but maybe it does. This is something I hope you can clarify.

(1) If I had used a real 140mm lens for the shot of the sign, from the same position, I would have needed to turn slightly to the left in order to capture it. Would that have changed the angle between the top of the sign and the top of the frame?

(2) In the full-frame shot it is surely evident that a fairly wide-angle lens was used. Experienced photographers would guess that the lens used was somewhere between 20mm and 28mm, based on the perspective evident in the image.

Would experienced photographers be able to guess the focal length of the lens used to capture the sign only. Does the perspective in the image of the sign only, give any clues?



Ray, by picking a 140mm lens to compare with you are making it obvious that you would have to swing the camera to get the sign in.  A better example would be to have used a 35mm lens and then the sign would appear in the top left of the image without having to move the camera.  I think you would find the perspective would then be exactly the same.  Less extreme example but one that proves the point.

Jim
« Last Edit: January 20, 2012, 03:49:20 am by Jim Pascoe »
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #123 on: January 20, 2012, 09:51:01 am »


But you have, Ray, you have changed everything.

Rob C


C,mon Rob. Let's be sensible ;D . I haven't changed everything. I haven't necessarily changed the camera body for a start, and I haven't changed the distance to an object that appears in the top right-hand corner of a 14mm frame simply by turning the camera, with a longer focal length of lens attached, slightly to the right, and slightly upwards. Nor have I changed my clothes.

I'm addressing the oft quoted maxim that focal length of lens has nothing to do with perspective, only distance from lens to subject. If I turn slightly on the same spot, in order to photograph with a telephoto lens an object which would appear at the edge in a wide-angle shot, I have not changed the distance between the camera and that object.

In fact, if I turn around 360 degrees on the same spot, I haven't changed the distance between myself and any objects surrounding me, unless I wobble a bit, of course.


Now it's clear to me, in practice, using existing of state-of-art lenses, that angle of view also has some bearing on perspective. Distance to subject is not the only factor.
Logged

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: This puzzling business of "35mm lens equivalent"
« Reply #124 on: January 20, 2012, 10:03:35 am »

... Nor have I changed my clothes....

Bingo! You finally grasped what ceteris paribus is all about! ;D

AlfSollund

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Re: This puzzling business of "35mm lens equivalent"
« Reply #125 on: January 20, 2012, 04:02:07 pm »

C,mon Rob. Let's be sensible ;D . I haven't changed everything. I haven't necessarily changed the camera body for a start, and I haven't changed the distance to an object that appears in the top right-hand corner of a 14mm frame simply by turning the camera, with a longer focal length of lens attached, slightly to the right, and slightly upwards. Nor have I changed my clothes.

I'm addressing the oft quoted maxim that focal length of lens has nothing to do with perspective, only distance from lens to subject. If I turn slightly on the same spot, in order to photograph with a telephoto lens an object which would appear at the edge in a wide-angle shot, I have not changed the distance between the camera and that object.

In fact, if I turn around 360 degrees on the same spot, I haven't changed the distance between myself and any objects surrounding me, unless I wobble a bit, of course.


Now it's clear to me, in practice, using existing of state-of-art lenses, that angle of view also has some bearing on perspective. Distance to subject is not the only factor.

Congratulations. Your the only person so far that has been able to move without moving  ;D ;D ;D. I recommend you to study the meaning of "move", "turn", "rotate" and so on, and try to do this without moving  ;D ;D ;D. The last time I looked into moving it was to change the position relative to something.

You should also contemplate what distance between camera and object means. When stating "camera" one usually means the sensor. This is not a sphere nor a 0-D point that records light from all around you so that rotating doesn't influence , but a 2-D rectangle. When you rotate or otherwise move the camera you move this rectangle relative to all objects around you. Any movement will change distance, thats what movement is.
Logged
-------
- If your're not telling a story with photo you're only adding noise -
http://alfsollund.com/

Eric Myrvaagnes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 22814
  • http://myrvaagnes.com
    • http://myrvaagnes.com
Re: This puzzling business of "35mm lens equivalent"
« Reply #126 on: January 20, 2012, 04:10:26 pm »

Now it's clear to me, in practice, using existing of state-of-art lenses, that angle of view also has some bearing on perspective. Distance to subject is not the only factor.
Aha! Now we're getting to the heart of the matter. I think some of the other factors that just might have some bearing on perspective are equally worthy of lengthy discussion in the forum. Here are a few possibilities:

1.   Now that you mentioned clothes, how do we know that changing clothes might or might not have some effect on perspective? Can you offer some proof either way?

2.   Or how about carbon fibre vs. a wooden tripod?

3.   Or whether or not the photographer in question knows enough Latin to be able to understand the meaning of ceteris paribus, all other things being equal?

4.   Or whether Michael will have to buy a new hard drive just to hold this thread? That might have some effect on his perspective on the issue.

etc.   :D

Eric

Logged
-Eric Myrvaagnes (visit my website: http://myrvaagnes.com)

feppe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2906
  • Oh this shows up in here!
    • Harri Jahkola Photography
Re: This puzzling business of "35mm lens equivalent"
« Reply #127 on: January 20, 2012, 06:23:57 pm »

It's a bit odd how on one of the most knowledgeable photography forums on the interwebs some people manage to dispute trivial linear, geometrical optic understood since 1600 AD.   :-\

It's actually worse than that: this topic is brought up every few months, and every time it's just as controversial.

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: This puzzling business of "35mm lens equivalent"
« Reply #128 on: January 20, 2012, 07:00:25 pm »

Ray,

Could you try this experiment:
1. Take a "head shot" portrait with a 35mm camera and a wide focal length, say 28mm or less, used close enough to have the face fill the frame: this should achieve the big-nose effect.
2. Repeat with the same focal length on a smaller format camera, moving back from the subject to get the same framing.
3. Repeat again with that smaller format camera but reducing the focal length so as to get the same framing with the same camera position as in 1.

Ideally, the format size difference would be extreme, like 35mm format vs a small sensor compact.

And to avoid any misinterpretation, willful or otherwise, I mean same faming of the subject in the complete, uncropped images delivered by the cameras.


My high school geometry and photographic experience both tell me that 1. and 3. (same camera-subject distance, different focal length) will be equally big-nosy, while 2. (greater camera-subject distance, equal focal length as in 1.) will be less big-nosy.
Logged

EricV

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 270
Re: This puzzling business of "35mm lens equivalent"
« Reply #129 on: January 20, 2012, 08:33:44 pm »

At the risk of adding to a thread which is already gone too far ...

Perspective as I understand it refers to the spatial relationship of objects in a scene, which of course varies with the viewpoint from which the scene is observed.  Perspective is a real physical property of the scene, and does not depend on the apparatus (human eye or camera lens) used to record the scene.  The big nose in a close-up picture taken with a wide-angle lens is an actual property of the face as viewed from that camera position.  The size of the (close) nose relative to the (more distant) eye depends on the camera position and not on the lens focal length.

I think others in this forum are using the term perspective to include what I would call lens distortion.  Pictures taken from the same location with different lenses will look somewhat different, even when cropped to the same size or angular coverage, because the lenses have different amounts of distortion.  Lens distortion changes the mapping of a scene (with its inherent perspective) to a flat sensor plane.  Pincushion or barrel distortion will make objects look different, both in size and orientation, depending on where they appear in the image.  But calling this effect a change in "perspective" caused by the lens is confusing terminology, and that is what many posters here are objecting to. 
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #130 on: January 20, 2012, 11:39:39 pm »

Ray,

Could you try this experiment:
1. Take a "head shot" portrait with a 35mm camera and a wide focal length, say 28mm or less, used close enough to have the face fill the frame: this should achieve the big-nose effect.
2. Repeat with the same focal length on a smaller format camera, moving back from the subject to get the same framing.
3. Repeat again with that smaller format camera but reducing the focal length so as to get the same framing with the same camera position as in 1.

Ideally, the format size difference would be extreme, like 35mm format vs a small sensor compact.

And to avoid any misinterpretation, willful or otherwise, I mean same faming of the subject in the complete, uncropped images delivered by the cameras.


My high school geometry and photographic experience both tell me that 1. and 3. (same camera-subject distance, different focal length) will be equally big-nosy, while 2. (greater camera-subject distance, equal focal length as in 1.) will be less big-nosy.

BJL,
You should understand better than anyone on this site that I know quite well that distance to subject affects perspective. This matter was discussed years ago. I'm certainly not disputing this fact. What gave you that idea?

What I'm disputing is that distance to subject is the only consideration as regards perspective. I think you also know, because I have the impression you have a scientific frame of mind, that one can prove that almost anything is true if one is selective in the data.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #131 on: January 21, 2012, 01:14:53 am »

You should also contemplate what distance between camera and object means. When stating "camera" one usually means the sensor. This is not a sphere nor a 0-D point that records light from all around you so that rotating doesn't influence , but a 2-D rectangle. When you rotate or otherwise move the camera you move this rectangle relative to all objects around you. Any movement will change distance, thats what movement is.

Distances are usually measured from point to point. I agree that a camera or sensor is not a theoretical point, nor am I. Nevertheless, if one wishes to be pedantic and absolutely precise about distance one needs to specify a point, or an average of numerous points.

So, for the sake of the exercise, shall we take a point in the centre of the sensor for measuring distances, or would you prefer to take an average of numerous points covering the whole sensor, representing the average distance of the sensor to any particular point in the scene?

Either way, I think you will agree that it is possible to rotate the sensor without changing the distance beteen a specified point in the far right of the scene and a specified point at the centre of the sensor.

Likewise,  if one were to use the average of say 100 points spread evenly across the whole sensor, as the measuring point, one could rotate or tilt the sensor in such a way that the distance from that average to a specified point in the right of the scene was unchanged.

In other words, in rotating the sensor to the right, the left edge of the sensor becomes closer to the specified point in the  right of the scene, and the right edge of the sensor becomes further away by the same degree. The average is: no change in distance. But we certainly get a change in perspective.

So, whilst I'd agree that there has been some miniscule alteration in distances to parts of the sensor, of the nature of 5mm or so, when I rotate the sensor to capture a part of the scene which can be captured with a wide-angle lens without rotation, the distance to the sensor as a whole has not changed.

I would think that such changes to parts of the sensor are best described as changes in angle of view, rather than changes in distance.

Below is the view out of my hotel window in Bangkok. You should be able to tell from the resolution which is the 14mm shot (cropped) and which is the 120mm shot (180mm equivalent because I used my Nikon D7000 for the telephoto shot).

Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: This puzzling business of "35mm lens equivalent"
« Reply #132 on: January 21, 2012, 03:32:24 am »

Hi,

I started to put together a small article on the issue: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/64-lnses-in-perspective

Best regards
Erik


Ray, by picking a 140mm lens to compare with you are making it obvious that you would have to swing the camera to get the sign in.  A better example would be to have used a 35mm lens and then the sign would appear in the top left of the image without having to move the camera.  I think you would find the perspective would then be exactly the same.  Less extreme example but one that proves the point.

Jim
« Last Edit: January 21, 2012, 03:58:12 am by ErikKaffehr »
Logged
Erik Kaffehr
 

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #133 on: January 21, 2012, 04:45:58 am »


Perspective as I understand it refers to the spatial relationship of objects in a scene, which of course varies with the viewpoint from which the scene is observed.  Perspective is a real physical property of the scene, and does not depend on the apparatus (human eye or camera lens) used to record the scene.  The big nose in a close-up picture taken with a wide-angle lens is an actual property of the face as viewed from that camera position.  The size of the (close) nose relative to the (more distant) eye depends on the camera position and not on the lens focal length.



I'm sorry but this just doesn't make sense to me, apart from the first sentence. I agree that perspective refers to the appearance of the spatial relationships in a scene from the point of view of the observer. However, surely it must be obvious that without an observer there is no perspective. Perspective cannot therefore be a real physical property of a scene independent of an an observer. If there's no observer, whether real person, or camera or recording device of some despription, there's no perspective.

The big nose effect in a close-up picture taken with a wide-angle lens is surely not an actual property of the face as viewed from that camera position. You could get into serious trouble telling your girlfriend what a big nose she has when you try to kiss her.

The actual size of her nose remains unchanged whatever position you're in. Perspective is an illusion. Isn't that obvious?

What the wide-angle lens does in photography is to create a sense of perspective in an image which is impossible for us to see in reality because the natural angle-of-view of our eyes is too narrow. What happens when we attempt to view a face from close-up is that we cannot simultaneously focus on the nose and the background. We have to shift our gaze from one part of the scene to another, therefore we don't get that direct comparison of the spatial relationship between different objects that are both inside and outside of the eyes' narrow FoV. Big is big only in relation to something that is perceived as being smaller.

This is I why I object to the claim that perspective has nothing to do with focal length of lens. If we try to simulate a wide-angle effect by stitching a number of shots taken with a short telephoto, similar in focal length to our eyes for example, then none of the individual shots prior to stitching will have an unusual perspective. The big nose in one image will only appear unusual in relation to the small ears in another image.

It's not until the images are stitched that the distorted perspective becomes apparent. However, after the images are stitched we have in effect a wide-angle shot. In other words, in order to get the rather amusing and absurd perspective of a very close-up portrait, one absolutely needs a wide-angle lens, or a wide-angle lens 'equivalent'.

That's why I claim that perspective is affected by both distance to subject and angle of view. I haven't yet figured out a way to get that distorted perspective effect from close-up without using a wide-angle lens, whether actual or effective through stitching.
Logged

AlfSollund

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Re: This puzzling business of "35mm lens equivalent"
« Reply #134 on: January 21, 2012, 05:26:22 am »

Distances are usually measured from point to point. I agree that a camera or sensor is not a theoretical point, nor am I. Nevertheless, if one wishes to be pedantic and absolutely precise about distance one needs to specify a point, or an average of numerous points.

So, for the sake of the exercise, shall we take a point in the centre of the sensor for measuring distances, or would you prefer to take an average of numerous points covering the whole sensor, representing the average distance of the sensor to any particular point in the scene?

Either way, I think you will agree that it is possible to rotate the sensor without changing the distance beteen a specified point in the far right of the scene and a specified point at the centre of the sensor.

Likewise,  if one were to use the average of say 100 points spread evenly across the whole sensor, as the measuring point, one could rotate or tilt the sensor in such a way that the distance from that average to a specified point in the right of the scene was unchanged.

In other words, in rotating the sensor to the right, the left edge of the sensor becomes closer to the specified point in the  right of the scene, and the right edge of the sensor becomes further away by the same degree. The average is: no change in distance. But we certainly get a change in perspective.

So, whilst I'd agree that there has been some miniscule alteration in distances to parts of the sensor, of the nature of 5mm or so, when I rotate the sensor to capture a part of the scene which can be captured with a wide-angle lens without rotation, the distance to the sensor as a whole has not changed.

I would think that such changes to parts of the sensor are best described as changes in angle of view, rather than changes in distance.

Below is the view out of my hotel window in Bangkok. You should be able to tell from the resolution which is the 14mm shot (cropped) and which is the 120mm shot (180mm equivalent because I used my Nikon D7000 for the telephoto shot).



You are incorrect. To consider the photo you need to take into account all points that make up the photo. If you change one point its a different photo.  The photo of an object (except an ideal  point with no size) will have points representing the real objects on the 2-D sensor. When you move that sensor (any kind of movement) you change the distance of these points on the sensor relative to their points at the object. Its possible to rotate the sensor along an axis so that a point or even a line will not change distance to a given line  or point at the object. But all other points distances will change, so you will change the distances and hence perspective for the photo by any kind of movement.

The average of the distances is of no importance to this.

And with this Im do not want to give any more education lessons of subjects that should be obvious at ground school levels, and that the Greeks mastered more than 2000 years ago (and probably other cultures before that).

As to "do there need to be an observer". Thats more philosophical question. I would say that if you set a camera to take a photo by timer the photo will be in the camera from he exposure. Others will claim that there is a potential for many photos, and that this potential is realized by an observer. But for this discussion the normal geometry is "good" enough, and we do not need to consider quantum physics.
Logged
-------
- If your're not telling a story with photo you're only adding noise -
http://alfsollund.com/

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: This puzzling business of "35mm lens equivalent"
« Reply #135 on: January 21, 2012, 07:23:20 am »

Ray, you seem to have missed, will fully or otherwise, the point of my 1. vs 3. comparison, which is about getting equal perspective from possibly very different focal lengths when used from at the same position relative to the subject. That is the issue arising from the misunderstanding of the original post, and to the myth that using a smaller format and proportionately shorter focal lengths will produce undesirable results like making portraits "big-nosy" even when using a lens that gives equal angular FOV to get the same framing.

But my scientific observations also include:

- once a person has posted more than a couple of times repeatedly defending a position against numerous posts full of contrary arguments, the participants' opinions almost never change in subsequent posts. Or in other words, once the arguments have been made enough that an open-minded person has the information needed to reach a reasonable decision, it is quixotic to push on in pursuit of unanimous agreement.

- threads this long are past their useful lifespan, unless they contain lots of nice photos.

I have no nice, relevant photos to post, and so I am signing of now.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #136 on: January 21, 2012, 08:21:47 am »

You are incorrect. To consider the photo you need to take into account all points that make up the photo. If you change one point its a different photo. 

I might well be incorrect. I've never heard of anyone who wasn't incorrect on at least a few points. But I fail to see how I am incorrect on the basis that I haven't taken into account all the points that make up the photo.

My argument so far has consistently been that the reason the wider angle shot has a different perspective, and from really close-up an obviously distorted perspective, is  because it is a different photo from another taken from the same shooting position and same angle, using a longer lens.

The wide angle shot has many more points and many more clues about spatial relationships than the narrow angle shot. One cannot get the 'big nose effect' with a long lens shot taken from the same close distance as a wide-angle shot.

Obviously one does get essentially the same picture of the nose, barring lens imperfections, but no clues as to the perspective of the nose in relation to other points which simply don't exist in the narrow angle shot.

If you arrange for those points (and clues to perspective) to exist, by stitching a number of shots, you are then very slightly changing the position of the sensor with each shot, which, according to you, represents a change in distance, however small that change may be.

Nevertheless, I'm not one for nitpicking. I accept that within reason it is possible to emulate a wide-angle lens with a narrow lens through stitching. And I accept also that much of the differences in my comparison image above, if not all, are due to lens imperfections.

My main thesis here is that the wide-angle shot can provide a different perspective because there are more objects or points in the scene for the eye and brain to construct a sense of spatial relationships. Cropping the wide-angle shot makes it a narrower angle shot.

Quote
And with this Im do not want to give any more education lessons of
subjects that should be obvious at ground school levels, and that the Greeks
mastered more than 2000 years ago (and probably other cultures before
that).

The ancient Greeks! They weren't all that good at the scientific method, were they!  Didn't they believe that we see, in part, because our eyes shine a light on the object we are looking at.

Didn't they tend to believe that the sun, and the entire observed universe, revolved around the earth?

Even Aristotle, one of the greatest of Greek thinkers, was confused about the number of teeth in a female human. He thought males had a greater number of teeth. Maybe his wife was abnormal, teeth-wise, or maybe he'd never bothered checking.

I can't help feeling compassion for all those countless individuals who, for over a thousand years, accepted everything that Aristotle wrote as being gospel, when we now know that most of what he wrote is sheer bunkum.

I'm reminded to some extent of the posters in this thread who blindly accept that focal length has nothing to do with perspective because certain renouned experts have said so.  ;D

Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #137 on: January 21, 2012, 10:58:12 am »

Ray, you seem to have missed, will fully or otherwise, the point of my 1. vs 3. comparison, which is about getting equal perspective from possibly very different focal lengths when used from at the same position relative to the subject. That is the issue arising from the misunderstanding of the original post, and to the myth that using a smaller format and proportionately shorter focal lengths will produce undesirable results like making portraits "big-nosy" even when using a lens that gives equal angular FOV to get the same framing.


Nope. Didn't miss your point at all BJL. We discussed this about 10 years ago. Distance to subject affects perspective, and a difference in focal length is required to maintain equal FoV when sensor size is different.

The issue is, if a different focal length is not provided and the FoV of the shot is therefore different, is perspective entirely the same? Can one sensibly talk about perspective being equal in shots with a different FoV?

Quote
- once a person has posted more than a couple of times repeatedly defending a position against numerous posts full of contrary arguments, the participants' opinions almost never change in subsequent posts. Or in other words, once the arguments have been made enough that an open-minded person has the information needed to reach a reasonable decision, it is quixotic to push on in pursuit of unanimous agreement.

You are probably right. Maybe I'm just an incurable optimist. If one re-phrases the argument in numerous ways, perhaps the penny will eventually drop. On the other hand, perhaps it's me who is wrong. Perhaps someone will rephrase the argument in such a way that it becomes apparent to me that I'm wrong.


Quote
- threads this long are past their useful lifespan, unless they contain lots of nice photos.

I have no nice, relevant photos to post, and so I am signing off now.

Nor have I. Nice photos are not my speciality, but I post them anyway for the sake of their meaning rather than their niceness.  ;D
Logged

AlfSollund

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Re: This puzzling business of "35mm lens equivalent"
« Reply #138 on: January 21, 2012, 11:00:08 am »


The wide angle shot has many more points and many more clues about spatial relationships than the narrow angle shot. One cannot get the 'big nose effect' with a long lens shot taken from the same close distance as a wide-angle shot.
You get exactly the same big nose effect. Position the nose inside the tele, then change to wide without moving sensor and the nose will appear identical. What I assume (but might be wrong) you are taking about is that the nose will appear different in the outer end of an wide. Well, thats simply the effect of trying to capture 3-D in 2-D. Its not possible to get a 2-D representation of an 3-D object without loosing something. So perfect round objects will appear oval in the outer end of photos by wides. This is called perspective. This does not mean that wides have different perspectives, just that they are wide.


Obviously one does get essentially the same picture of the nose, barring lens imperfections, but no clues as to the perspective of the nose in relation to other points which simply don't exist in the narrow angle shot.

If you arrange for those points (and clues to perspective) to exist, by stitching a number of shots, you are then very slightly changing the position of the sensor with each shot, which, according to you, represents a change in distance, however small that change may be.

Nevertheless, I'm not one for nitpicking. I accept that within reason it is possible to emulate a wide-angle lens with a narrow lens through stitching. And I accept also that much of the differences in my comparison image above, if not all, are due to lens imperfections.

My main thesis here is that the wide-angle shot can provide a different perspective because there are more objects or points in the scene for the eye and brain to construct a sense of spatial relationships. Cropping the wide-angle shot makes it a narrower angle shot.

If you are stitching several tele photos you to correspond to one wide photo each of these stitched photos will have a different content and perspective since you have moved the sensor, and the final stitched picture will be different and have a different perspective than the wide picture. You can also avoid "the nose effect" seen at the outer end of the wide photo since you are making a different photo with a different perspective by making a "tele nose photo" (since you have moved the sensor).

In fact you don't need a tele to do this. You can use the wide and point this in different directions and take several photos, and then make a crop in the center corresponding to taking these with a tele. If you stich these the result will be exactly the same as with the stitched teles (except resolution), but different to the original un-stiched wide, proving again that wides and tele gives the same perspective and that only changing position matters in terms of perspective.

So by stitching photos you are NOT emulating a wide, but you are making a new photo that can have the same angle and more or less the same content as the wide, but will have a different perspective.


My thesis is: the photos has exactly the same perspective independent of focal length as long as the position of the sensor is not changed.


The "sense of spatial relationships" is a another discussion that needs a separate thread. You will of course get different information form different pictures (such as a tele versus a wide). This has nothing to do with perspective.


P.S. Sorry to all about keeping this alive, for me this is obvious, but not all might agree that its obvious. To disagree on facts is anther issue. No-one can agree on the issue of perspective since it is inherent in our reality  ;)
« Last Edit: January 21, 2012, 11:02:09 am by AlfSollund »
Logged
-------
- If your're not telling a story with photo you're only adding noise -
http://alfsollund.com/

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: This puzzling business of "35mm lens equivalent"
« Reply #139 on: January 21, 2012, 11:28:20 am »

You get exactly the same big nose effect. Position the nose inside the tele, then change to wide without moving sensor and the nose will appear identical.

Of course it will. The nose remains the same as long as the lens is wide enough to capture it. What changes with the angle of view, as you change the zoom lens to wide, without altering your position, is the perspective, ie. the relationship between the size of the nose and other objects that come into view as you zoom to wide.

Without those other objects, ears, eyes, people standing behind the head, houses, trees etc, there is little sense of perspective. I wouldn't claim there's no sense of perspective. The nose itself, even though divorced from its background, still has a certain perspective, from tip to base.

However, when the nose is placed in a different context of smaller than expected eyes and ears and pathetically small back ground objects, as a result of the zoom moving to wide mode, the perspective of the nose as a whole changes. It then appears as an exceedingly large nose. Got it?
Logged
Pages: 1 ... 5 6 [7] 8   Go Up