Equivalent focal lengths and crop factors...

[Will D.]

On 2005-01-03, Jeremy Nixon <jeremy@exit109.com> wrote:

> Will D. <willd@no.spam> wrote:
>

>> In this, we note what we can and cannot see as compared to what we know
>> exists, and we note the relative perceived size of objects that we have
>> had occasion to measure objectively (if only from some standard
>> distance, etc). Artists alter our sense of orientation wrt the scene in
>> their work. And so forth.

>
> Yes, I agree; an artist using a media where the image is created from
> "scratch" can play with perspective as you say. However, a photographer
> cannot, so I don't see it as terribly applicable.
>
> However, although a photograph does only display what is actually visible,
> it doesn't include enough information for a human to reproduce the experience
> of seeing the scene in person, in terms of orienting himself and determining
> what goes where. Our eyes and brains use things that aren't included in a
> flat photograph to do that. This allows some degree of optical illusion in
> perspective, but still only in a way that doesn't change what is actually
> there. The camera records what is visible from a certain viewpoint, and
> that's it.

Well, to the extent that people learn to ignore "optical illusion in
perspective", it becomes (or can become) of less importance that perhaps
you inferred from my comments. Certainly people who know little or
nothing about photography easily ignore the issue. But my point was
that newbies trying to extent their photographic knowledge and skills
will come to ask what's going on, and to state that nothing is going on
is to suggest to them that what they see is non-existent. And that's
not the case.

>> Having argued this dual definition, what I've said is that the invariant
>> angle of view of an SLR viewfinder creates such a distortion, and that
>> newbies trying to figure all this out for themselves usually do not
>> "get" this fact. The dogma does not satisfy their experience because it
>> either ignores or refutes that experience, neither of which is
>> reasonable as far as I'm concerned.

>
> I also agree with this, but I think we differ on an underlying cause. In
> the old days of film, it was not a problem to describe perspective in terms
> of focal length, because a given focal length always meant the same thing
> to 35mm camera users. Nowadays, you simply cannot use focal length to
> describe perspective, because focal length can't describe the angle of
> view on a digital camera -- lenses of the same focal length can have very
> different views on different cameras, even the exact *same* lens used on
> different cameras. Unfortunately, lots of stuff, including written stuff
> in formerly-authoritative texts, still exists talking about perspective
> of certain lenses, and all of that stuff is now simply not true.
>
> If I put my 50mm lens on my 35mm SLR, and then put it on my digital SLR,
> it will not give me the same field of view on both cameras. On the
> digital it will behave exactly the same as a 75mm lens would on the 35mm
> camera, with regard to field of view (not depth of field, which is a
> whole other topic). If you want to consider perspective as a function
> of focal length, you must then think of it as a 75mm lens when using it
> on the digital SLR (at least a Nikon one) because that's how it will
> work. Given identical framing (and thus differing camera-to-subject
> distance) with that lens on the two cameras, the perspective will be
> different.
>
> So, I think people get confused because they have learned about perspective
> either back when you *could* describe angle of view by focal length, or
> from things written at that time. Back in the day, if you were writing
> for 35mm film users, you could say that 50mm is a "normal lens" and be
> right. You can't do that any more because it's not going to be right.

I'm talking about newbies that have never used an SLR of any kind. They
know nothing of the traditional focal length / angle of view
correspondence. That's only for people who've used an SLR enough to
know better.

>> You say the print looks exactly the same regardless of viewing distance,
>> and so it should. The question is whether or not it looks natural, or
>> does it look distorted: the relative size of objects not corresponding
>> to what might be expected, creating the impression of exaggerated or
>> compressed distances.

>
> That's what I mean. To me, a wide-angle shot with "distortion" at the
> edges (I put "distortion" in quotes because it's not really distortion,
> and I've been scolded for using that word to describe it here already,
> but I can't think of a better word to use) still looks identically
> distorted no matter where the print is relative to my eyes. But since
> everyone else says it doesn't, fine; this is something I haven't learned
> to see, like those pictures in the mall with all the dots that are
> supposed to become something if you stare at them long enough, but
> which always (to me) just look like a bunch of dots.

I use the word distortion without quotes because that's what it is.
It's not an optical illusion at all. It's what happens when one tries
to force one angle of view into another angle of view. And you're right
when you observe that the print looks equally distorted from any viewing
distance if, and probably only if, you cannot suspend your sense of
current physical environment enough to "be with/in" the scene displayed
by the print.

I've no idea what percentage of people can't do that, but I've got to
assume that a goodly percentage can, else visual art would not be the
universally appreciated commodity it has long been.

> But if that does work, it works because you're changing the angle at which
> you're seeing certain parts of the scene, which is basically the same
> thing you do when you move the camera. The wide-angle "distortion" is
> caused by seeing parts of the scene at different angles, but then
> projecting that resulting image onto a flat surface. So that doesn't
> mean that something other than viewpoint changes perspective; you're
> just changing the viewpoint.

Well, yes, of course. The scientific definition of perspective always
obtains. As far as flat surfaces are concerned, one might think of the
observed image as an image plane, but I'm not clear that this is really
relevant. In one case it's the ground glass, and the other it's a print
surface, but that doesn't mean that the viewer stops trying to extract
depth information. And it's the effect on the viewer that is primary to
my original statement.

And it's at that point that the second definition becomes relevant.

>> Here's a question for you. You've noticed the engraved message in the
>> passenger side mirrors of modern cars? It says something like "Warning!
>> Objects in mirror are closer than they appear!" Do you consider the
>> image in such mirrors to be distorted? Most people do, I think. Do you
>> think this is important? If you don't, why not? Or maybe better yet,
>> if you do, how do you explain this?

>
> That is distortion, due to the mirror being convex. So? It's basically
> the same thing; it's a "wide angle" mirror.

Exactly. It's one angle of view being forced into another angle of
view, which is the cause of the distortion. Now, here the flat surface
issue is relevant. Do you see the mirror as a flat surface? I don't.
I'm accustomed to extracting depth information from a mirror because
normally the mirror is planar. Of course there is an adjustment to be
made because of the viewer to mirror distance, but most people are able
to make that adjustment automatically, I think.

I think we've come to agree that the issue is real, whether or not it is
relevant here in general or specifically in the case of newbies
learning. We can debate the particulars and details, but it's that
agreement of the reality of the issue that I was seeking.

Will D.

[Colin D]

Ryadia wrote:
<various snips>

> Artists are more
> inclined to focus on the final result of what they do. Their cameras are
> only tools they use to achieve a photograph. They know a 100 mm lens will
> produce a portrait with better 'perspective' than your assertion that a 20mm
> one can too. Many could make the portrait with or without a camera.
>
> I still use a portrait lens for portraits because
> 80, 100 mm focal length lenses produce a better perspective to the
> photograph. That is my statement. I learned it from masters of photography,
> probably before you were born and you can't change it now with all your
> mathematics and science.
>

Yes, I agree with what you say here, Doug, those lenses will produce a
better perspective in the final image than a shorter lens - *but* the
reason has nothing to do with the lens characteristics. It is entirely
due to the position you take with respect to the subject that produces
the desirable perspective. The longer lens merely lets you compose the
image within the frame from your viewing position. A shorter lens from
the same position will produce the same perspective, but a smaller image
of the subject.

The natural thing with any lens is to reasonably fill the frame with the
subject. Short lens, closer to subject. Long lens, further from
subject. The perspective is generated by the camera postion, not the
lens.

In my photographic experience, largely before good zooms were available,
a protog. would move around to get his desired viewpoint/perspective,
and then choose a lens that would reasonably fill the frame with his
envisaged shot. Today, with passable to good zooms available, he should
still walk around to get the best position, and then zoom to frame the
shot (Most non-pro's just stand where they first saw the shot and zoom
away, ignoring perspective entirely).

I can see that it is a kind of mental shortcut to associate perspective
with given lenses, and from that to accept that the lens is the source
of perspective. I find it much better to realise that perspective is
the result of viewpoint, and then choose the optics to get the shot.
More freedom that way.

Colin.

[Will D.]

On 2005-01-03, Frank ess <frank@fshe2fs.com> wrote:

> Will D. interjected:
>

>> Whoa! I started this thread, and Jeremy was responding to my posts.
>>

>
> So, did you get your answer?
>
> If not, how many more thousands of lines do you think it will take?
>
> I suggest that everyone who has an interest in this thread now amounts
> to some negative number, that the participants have all established
> their positions (heh) and indicated the likelihood of accommodating any
> others.
>
> That's my perspective on it.

Wasn't looking for an answer.

What made you think I was looking for one? What question did you fancy
I was asking? Well, I'd ask what sort of answer you think I'm seeking,
but don't think I'm interested in anything taking thousands of lines.

Do you suppose that you actually understood what this thread was all
about? If you think I was asking a question, either you didn't read the
thread, or you're short on comprehension of what you did read. If
you're interested, I suggest you go back and find out what you didn't
get, and let me know. I'd be glad to explain what you didn't
understand.

Will D.

[Will D.]

On 2005-01-04, Colin D <ColinD@killspam.127.0.0.1> wrote:

>
>
> Ryadia wrote:
><various snips>

>> Artists are more
>> inclined to focus on the final result of what they do. Their cameras are
>> only tools they use to achieve a photograph. They know a 100 mm lens will
>> produce a portrait with better 'perspective' than your assertion that a 20mm
>> one can too. Many could make the portrait with or without a camera.
>>
>> I still use a portrait lens for portraits because
>> 80, 100 mm focal length lenses produce a better perspective to the
>> photograph. That is my statement. I learned it from masters of photography,
>> probably before you were born and you can't change it now with all your
>> mathematics and science.
>>

> Yes, I agree with what you say here, Doug, those lenses will produce a
> better perspective in the final image than a shorter lens - *but* the
> reason has nothing to do with the lens characteristics. It is entirely
> due to the position you take with respect to the subject that produces
> the desirable perspective. The longer lens merely lets you compose the
> image within the frame from your viewing position. A shorter lens from
> the same position will produce the same perspective, but a smaller image
> of the subject.
>
> The natural thing with any lens is to reasonably fill the frame with the
> subject. Short lens, closer to subject. Long lens, further from
> subject. The perspective is generated by the camera postion, not the
> lens.
>
> In my photographic experience, largely before good zooms were available,
> a protog. would move around to get his desired viewpoint/perspective,
> and then choose a lens that would reasonably fill the frame with his
> envisaged shot. Today, with passable to good zooms available, he should
> still walk around to get the best position, and then zoom to frame the
> shot (Most non-pro's just stand where they first saw the shot and zoom
> away, ignoring perspective entirely).
>
> I can see that it is a kind of mental shortcut to associate perspective
> with given lenses, and from that to accept that the lens is the source
> of perspective. I find it much better to realise that perspective is
> the result of viewpoint, and then choose the optics to get the shot.
> More freedom that way.
>
> Colin.

Possibly what is missing here is an understanding of why we do
portraiture like we do. There is a reason for it.

Consider the nature of a portrait. It is a formal image, intended for
posterity. We can ask: what is the essence, if any, of a formal image?
A reasonable answer is that it is a public image, intended for viewing
by unknown people now and in the future. The key attribute here is
"public".

A public image is intended to be what the public would see if the
subject were present *in public*. What that means is that the viewer
would be expected to see the subject as he would be seen to comport
himself in public, and one of the things that has long been the case, at
least here in the west, is that in public, one maintains one's personal
space, wherever possible. That's one's personal "comfort zone", if you
will.

We are used to perceiving faces of those with which we have only a
public relationship at the distance mandated by that personal space.
Painters and portraitists work at almost exactly that distance. It's
worth noting that the more lofty the public persona, the greater the
distance, usually but not always entailing the inclusion of more space.

Now contrast this with intimate images. Some generic image formula
comes to mind: get the eye in focus and let the rest of the face fade
out of focus, often shot from close enough so that the manipulation of
the features needs to deal with the big nose problem. Isn't it
interesting that we have no problem with the arrangement of a person's
features in this case!? There can be several reasons for this, seem's
to me. One is the obvious personal intimacy of established sexual
relationships. Another is the unwanted intimacy of those we might
consider undesireable. "In your face" shots of the poor and homeless,
and those who don't warrant respect, etc.

The fact is, I suggest, that it is convention that dictates the way we
regard the most pleasing faces, and that convention is driven by social
considerations. Fashion photography generally observes the public
distance, even when only part of the face is shown in the print. When
the features display the distortion inherent in views from an intimate
distance, people can tend to become uncomfortable unless they are
willing to accept the implications of that sort of relationship.

So what's going on here is that when we do portraiture, we establish the
courteous distance from the subject, which gives us the slightly
compressed arrangement of features that we're accustomed to seeing, and
then photographers chose the angle of view they want to fill the frame
as they like.

Yeah, irrelevant trivia, but hell, it fits this part of this thread!

Will D.

[Ryadia]

"Alan Browne" <alan.browne@freelunchVideotron.ca> wrote in message
news:crbnrr$b72$1@inews.gazeta.pl...

> Ryadia wrote:
>

> > able to even comprehend the meaning of what I originally said, much less

put

> > your argument into perspective when discussing a portrait.

>
> The only point that matters Doug, is that perspective is what perspective

is,

> and it is not anything else but what it is. > You will find that such

admission will not change your ability to make

> portraits, and it might even improve your photography in some other

context.

>
> Cheers,
> Alan

I am tempted to continue in the face if contradiction but I fear there is
no point in attempting to make the horse drink after getting it to the
water.
[url]http://209.196.177.41/07/07-06.htm[/url]

Doug