[Wheatfield]

It isn't controversial at all. However, the reason why telephotos tend to "compress" perspective is because we tend to use telephoto lenses to bring that which is far from us closer.
The lens isn't doing the compression, it is the compressed near/far relationship that is doing it.

[audiobomber]

It isn't controversial at all. However, the reason why telephotos tend to "compress" perspective is because we tend to use telephoto lenses to bring that which is far from us closer.
The lens isn't doing the compression, it is the compressed near/far relationship that is doing it.

Good, we agree that telephoto lenses compress perspective. Not everyone seems to be on board.

People keep trying to relate "normal" to their field of view, but that's not what it's about. It doesn't matter if humans have a FOV of 180 degrees or 120 degrees or 50 degrees, because FOV doesn't enter into the definition of normal. A lens is normal if it shows similar perspective to the unaided eye. It's not a telephoto, it's not a wide-angle, it's a normal. Do we still agree?

[rparmar]

I don't have that book, but perspective is not originally a photography term anyway, it's from drawing and art. Perspective in art is uses techniques to show the relative distances between objects, through converging lines and relative sizes.

But it doesn't matter how the term is defined in a different field. In photography the term is only defined one way. It's dictated by where you stand relative to your subject and governs the appearance of spatial relationships between objects. Which also has a lot to do with perspective in art, if you think about it.

It has nothing to do with focal length, which governs field of view.

If you wish to define it some other way, I doubt you will have many supporters.

[rparmar]

Good, we agree that telephoto lenses compress perspective.

No, they do not. Go back to your art definition. When you change lenses have you changed where the horizon is? Have you changed where the vanishing point is? Have you changed how parallel lines act? If so, you are somehow changing the very geometry of the world by putting on a new lens. That is a pretty amazing feat!

Telephoto lenses compress distance, in a colloquial way of speaking. But they do not change perspective.

[audiobomber]

But it doesn't matter how the term is defined in a different field. In photography the term is only defined one way. It's dictated by where you stand relative to your subject and governs the appearance of spatial relationships between objects. Which also has a lot to do with perspective in art, if you think about it.

We agree, perspective in art and photography is the same concept, i.e. the spatial relationship between objects.

It has nothing to do with focal length, which governs field of view.

Perspective does have a relationship to focal length, as per this photography definition:

Normal Lens: A lens that does not change the perspective of the image like a telephoto or wide-angle lens. Photography Terms and Definitions for Traditional Photography

[Ben_Edict]

We agree, perspective in art and photography is the same concept, i.e. the spatial relationship between objects.



Perspective does have a relationship to focal length, as per this photography definition:

Normal Lens: A lens that does not change the perspective of the image like a telephoto or wide-angle lens. Photography Terms and Definitions for Traditional Photography

Dan - don't trust all of what you find on the web. It need not be authorative.

A lens does not change perspective. Changning the point where you stand will change the perspective. It is the same as with traditional arts. I know that in photography (also in books, nhot only on the web), the use of standard terminology has sometimes fallen victim to daily slang, but that does not indicate its validity.

Ben

[GeneV]

This is the reason, why I always saw 50mm as a compromise lens, which is nevertheless useful, not the least, because it is the fastest glass available.

The 120 deg field of view is very unsharp outside the central part, though (ofcourse the eyes scan the scenery) and the central sharp part is more like the angle of view an 80mm lens gives.

Ben

Yes, and on top of everything else, our eyes move, which further adds complexity.

[rparmar]

Perspective does have a relationship to focal length, as per this photography definition:

Normal Lens: A lens that does not change the perspective of the image like a telephoto or wide-angle lens. Photography Terms and Definitions for Traditional Photography

They are wrong. Simple as.

[audiobomber]

Dan - don't trust all of what you find on the web. It need not be authorative.

A lens does not change perspective.

But in my personal experience a lens does affect perspective. That's why I posted the photo with the trains and water tower; to demonstrate that perspective was compressed by a telephoto lens. To put it in simpler terms, the water tower looked larger, and therefore closer to the trains in my photo than it did in real life.

A normal lens doesn't affect perspective, in other words a normal lens captures a normal perspective and that's the only definition of normal that makes sense to me. No definition that includes degrees fov makes any sense to me.

[audiobomber]

They are wrong. Simple as.

What's your definition of a normal lens?

[grey goat]

I remember an old ExLax TV commercial wherein the voice-over guy says, "Doctors say what's normal is what's normal for you. However, age has changed it."

Anyway, I just wanted to add that when I got my K-7 two and a half months ago, I thought about different lenses I'd want as a "normal" lens. I wound up with an M28/3.5. Sometimes I refer to it as a "wide normal" since the standard issue on a typical 35mm SLR was in the 50mm-58mm range.

I like the M28 on my K-7 a lot. I sure would like to try an FA31/1.8, but the M28 cost a fraction of what that lens costs.

I don't have any real philosophical ax to grind here, but after reading through this thread, I just thought I'd toss in my 28 cents' worth into the mix.

(And RioRico is right: "normal" is over-rated. So, for that matter, is symmetry. . . .)

[rparmar]

What's your definition of a normal lens?

My definition is that it is a lens with the same focal length as the diagonal of the film/sensor. Thus 43mm is normal for full-frame and 28mm on APS-C.

This definition has several advantages.

1. It has nothing to do with perspective, since focal length doesn't.

2. It has nothing to do with ideas of how the human eye sees, since a camera works nothing like an eye, human or otherwise.

3. It corresponds to historical reality, eg. what camera manufacturers in all formats have always told us was normal. Yes, 43mm is off from 50mm, but the normal lens in different formats has always scaled with the film size in this way.

4. It corresponds with equivalent field of view, which is what most people really care about when they choose a focal length.

[RioRico]

My definition is that it is a lens with the same focal length as the diagonal of the film/sensor. Thus 43mm is normal for full-frame and 28mm on APS-C.

This definition has several advantages.

1. It has nothing to do with perspective, since focal length doesn't.

2. It has nothing to do with ideas of how the human eye sees, since a camera works nothing like an eye, human or otherwise.

3. It corresponds to historical reality, eg. what camera manufacturers in all formats have always told us was normal. Yes, 43mm is off from 50mm, but the normal lens in different formats has always scaled with the film size in this way.

4. It corresponds with equivalent field of view, which is what most people really care about when they choose a focal length.

Right on! And why is everyone so hung-up on normality? Or normalcy, as one authority called it. Be abnormal. Be as abnormal as possible, as abnormal as you can get away with, without being detained involuntarily for long periods. CHERISH your abnormalcy. WORK WITH your abnormalcy. And watch YOUNG FRANKENSTEIN.

But I digress. Problem is, as with so many other words, 'normal' has different meanings in different contexts. It's a right-angle, 90 degrees (from the Latin norma, a carpenter's square). IIRC, it's within one standard deviation of a mean/average (statistics). In photo marketing, it's whatever lens is supplied with the camera (unless it's marketed as a wide or tele cam). It's mainstream, not-too-weird (in uptight societies). It's being free from disease (medicine, biology). I'm sure other definitions can be found.

And in photographic optics, it's the diagonal of a frame -- but frames aren't all the same aspect, so the AOV provided by that diagonal varies. APS-C and 35/HF and 35/FF and 645cm and 6x9cm all have a 2:3 aspect ratio; their normals have the same AOV. A 4x4cm or 6x6cm or any other square frame is 1:1, with a smaller AOV. The HDTV 16:9 aspect has a greater AOV. Anything beyond that is definitely wide; but we could legitimately refer to the diagonals of aspects between 1:1 and 16:9 as 'normal'.

rparmar is absolutely correct. Can we go on to other subjects now?

My suit is pale yellow, my nationality is French,
and my normality has often been subject to question.
--T.Williams

[jct us101]

I find 28mm on a digital camera to be wide enough (the 42mm equivalent) for most of my needs, and I find that I never really keep the photos that I take when I'm using 18mm unless it's for certain situations. However, I doubt that Pentax is going to make a normal lens that doesn't start with 18mm, but hopefully when they make one it'll go to 105mm. Yes, 18-105mm, that's it.

[Marc Sabatella]

I don't have that book, but perspective is not originally a photography term anyway, it's from drawing and art. Perspective in art uses techniques to show the relative distances between objects, through converging lines and relative sizes.

True. And this fact completely disproves your hypothesis that changing focal length somehow magically alters perspective. It does not, and cannot. If it did, the last 500 years of art history would not have been possible. The science of perspective has been well-understood for 500 year and can definitely and mathematically be proven - as well as easily demonstrated - to depend on the viewer's position relative to the subject and *nothing else*. the focal length of the lens you might happen to be using is irrelevant, as perspective is perspective whether you are using a camera or not. If you stand in a given spot and looking in a given direction, there is *nothing* you can to change the perspective of what you see.

In photography, a telephoto lens compresses perspective compared to your vision, a wide angle lens expands perspective, that's all I'm saying. I don't see why it's controversial.

It's not controversial - it's just wrong. You are confusing perspective - the 500-year-old study of how parallel lines appear to converge and so forth - with what is known as "perspective distortion". It *is* true that if you take a picture with a telephoto lens, then print it and view it from a "typical" distance, the result will appear to show the sort of compression you are referring to. this isn't because the lens magically compressed anything, but is entirely due to fact that from where you view the print, the print itself does not have the same FOV as the scene depicted. If, however, you viewed that same print from further away, then it's FOV *would* match that of the scene depicted, and the illusion of compression would disappear.