[pcarfan]

Will,

That calculator leaves out the most important part, which is the change to the scene as the sensor to subject distance is changed.

Visualization of landscapes is the trickiest part of photography for me for now.

[alohadave]

The shorter the focal length the wider the field of view. A 50mm lens is more or less equivalent to the way people see things. Shorter than that perspective changes. With a 12mm lens, very close objects appear disproportionately large and the distances between things can seem much larger than normal. With a long telephoto the apparent distance between objects is flattened.

michael mckee
My Port Townsend – A City in Photographs – 365

What you are talking about is distortion and angle of view. At 12mm close objects seem disproportionately large because you are physically close to the object.

If you took two pictures from the same spot at 50mm and 200mm, and cropped the 50mm to match the 200mm picture, they would be identical besides differences in DoF.

If however, you changed your position with the 50mm to match what you see with the 200mm, your perspective will change because you had to physically move to match the angle of view of the 200mm.

[WMBP]

That calculator leaves out the most important part, which is the change to the scene as the sensor to subject distance is changed.

Well, the "scene" itself doesn't change—just what you capture in the camera. I smile as I make this pedantic comment and ask for your forgiveness, but it is a somewhat important distinction.

What you're saying is basically what I said in my previous post. Worth remembering by the way that the distance between the camera (or sensor) and the subject doesn't just impact perspective, it also has a potentially huge effect on depth of field. Beginners often think depth of field is controlled by the aperture. That's one key factor. But distance is another. (Sensor size is a third, but most of us don't carry around a couple of cameras with different size sensors, so for most of us, sensor size isn't a variable.)

Visualization of landscapes is the trickiest part of photography for me for now.

Well, that's why the camera has a viewfinder. :-)

But I think I know what you mean. I'd say that there are two ways to think about visualizing the shot. (NOTE: I avoid Ansel Adams' word "previsualization" here because it means something somewhat different from what I'm talking about.)

One way to visualize the shot is to ask yourself, what kind of shot will I get from here, where I am standing now? I am calling this visualization, because you often have to ask yourself this question without raising the camera to your eye. I've gotten pretty good at this. Especially if I'm shooting outdoors where the number of potential vantage points is nearly infinite, I tend to walk around until I "see" the shot I want to take. I confirm through the finder, then shoot.

But the other way to visualize is to ask yourself, without moving, where would the best place for this shot be? Or to put it differently, what would the shot look like if I took it from over there, or up there, etc.? I'm NOT so good at this. I sometimes think, gee, it would be nice to shoot this lake from a hot air balloon, or something like that. But that sort of thing is usually fairly easy and obvious. Actually I would not even mention this kind of visualization as something distinct, were it not for the fact that I know photographers whose work surprises and delights me precisely because they are good at this kind of thing. I see shots from time to time that make me ask, how did the photographer ever think of standing THERE to take the shot? What a brilliant and unexpected idea!

I would insist that there is great skill involved even in the first approach. Some photographers—perhaps nearly all non-serious amateur photographers—simply shoot, then figure out later if the photo is any good. The greatest satisfaction that I have from photography comes when I KNOW—absolutely—at the moment I click the shutter, that it's going to be a good shot. I confess it doesn't happen nearly as often as I wish. But the better I get at this, the fewer shots I take, and the higher my percentage of keepers.

Will

[WMBP]

By the way, in alohadave's recent post, I see that he quotes mysticcowboy, who said, "A 50mm lens is more or less equivalent to the way people see things."

Now the truth is, the camera NEVER sees the world the way we see it, because even an ultrawide angle lens has less peripheral vision than we have. We don't see the world in a frame.

But for what it's worth, I think the 50mm = normal angle of vision is correct for film cameras, and that for APS-C sensors with a 1.5x crop factor like our Pentax digitral SLRs, the "normal angle of vision" lens is somewhere around 35mm, perhaps even slightly wider. I think of my Sigma 28 f/1.7 as having a "wide normal" angle of vision. This just means that a shot taken (on a Pentax DSLR) at around 30-35mm will have pretty much the same perspectival properties as the scene viewed by the naked eye.

Of course there's no prize here for shooting with a normal angle of vision, indeed, very often, a non-naked-eye angle of vision does really interesting things. Shooting with a wide lens, from up close, makes people look fatter or at least wider. Shooting a portrait with a telephoto lens (like the Pentax 70) may lessen the impact of, oh, a prominent nose or chin. But if I were photographing Jay Leno, I'd get in tight with a wide lens.

Will

[pcarfan]

Will, I know I was repeating what you said as I was just acknowledging your pointers.

I am considering both types of visualization, but I would be happy with the visualization of different focal lengths from where I am standing, so I can pull out the correct prime from my quiver. But the two are partially related in the sense, if I know what 21 will look like from where I stand and think it would look even better if I hike for another mile, then I may set up my tripod there and not where I am currently standing. It would be nice to see what vantage point and at what focal length a scene would render ideal, but I don't want to try to run before I can walk

I started this thread after I booked my first real photography vacation. I am going to Yellowstone national park in July for a week. I am very new to landscapes, tried a few this past summer, but it was just a few days of shooting.

[pacerr]

Is there a trick or tool that will tell me, without having to mount the lens, on what a scene will look like at 15, 21, 24, 31, 43, 77, 90, 135 ...

Try this interactive tool at the Tamron web site:

Focal length comparison tool, Tamron USA

I use an extended calibrated fist (or two) to judge lens coverage. Compare the coverage of your hands, fingers, hat, lunch pail, etc., to the equivalent lens coverage in the comfort of your own back yard. Write the data sheet on the cuff of your shirt or pin it to your mittens. Just be careful how many fingers you use in public and how you point 'em!

H2

H2

[RioRico]

By the way, in alohadave's recent post, I see that he quotes mysticcowboy, who said, "A 50mm lens is more or less equivalent to the way people see things."

Now the truth is, the camera NEVER sees the world the way we see it, because even an ultrawide angle lens has less peripheral vision than we have. We don't see the world in a frame.

Exactly. Our vision, light captured with imperfect eyeballs, preprocessed by optic nerves, post-processed by brains, filled in with memory and expectation, is a combination of ultra-wide and long-tele with more-or-less infinite DOF. We filter out what we don't want or expect to see, and think we're perfect eye-witnesses. Hah. There's a famous experiment where an audience watches some gym-suited guys tossing a basketball around, and never notice the guy in the gorilla suit walking and cavorting behind them. The audience didn't expect an ape, so they didn't see him. If something unexpected is in front of us, we often just won't see it -- but the camera will.

But for what it's worth, I think the 50mm = normal angle of vision is correct for film cameras, and that for APS-C sensors with a 1.5x crop factor like our Pentax digitral SLRs, the "normal angle of vision" lens is somewhere around 35mm, perhaps even slightly wider. I think of my Sigma 28 f/1.7 as having a "wide normal" angle of vision. This just means that a shot taken (on a Pentax DSLR) at around 30-35mm will have pretty much the same perspectival properties as the scene viewed by the naked eye.

In pre-35mm days, 'normal' was calculated as the diagonal of the frame. With Ansel's 8x10" view cam, 'normal' was 13.4" (rounded to 12"). On a 6x9 cm folder, 'normal' is 101mm (often rounded to 105mm). For a 24x36mm 135 frame, 'normal' would be 43mm (often rounded to 45mm). But 50-55mm is a sweet spot for making fast lenses, so 50mm is common. 50 is actually a short tele, and many classic 135 photographers used it as such. [I think Alfred Eisenstadt was quoted, explicitly stating such, in THE EYE OF EISENSTADT.] I've read that larger formats offer more leeway for cropping; so on a 135 camera with a 50mm lens, the image is pre-cropped, without wasting valuable film real estate.

(By 135, I mean the 35mm film cannister. Using the 135 label is a bit tidier than throwing mm's around everywhere.)

For APS-C cams (most modern dSLRs), true 'normal' is 30mm (rounded to 28mm), like 43mm on a 135 cam. 35mm on APS-C is 52mm on a 135 cam, close to 50mm (and some Russian lenses are 52's). Some Pentaxians love their 40mm's, which would be 60mm on 135 -- the closest lenses actually made to that were 58mm's from Pentax, Helios, and a few others. Like using a 50 or 55 or 58mm on a 135 cam, using a 35 or 40mm on APS-C either lets you stand off more, or feel more intimate in your shooting. It's like taking your 'normal' vision and squinting, concentrating a bit more on a subject, zooming in slightly. You are pre-cropping the image, discarding the periphery.

---

Last edited by RioRico; 02-24-2010 at 05:45 PM.

[Votesh]

I think there a multiple focal lengths that can work for a scene. It mainly comes down to what you want out of the scene. What you want to capture, and how you want to capture it.

It would be impossible to create a simulation to use for the purpose of shoosing a FL in this manner. Really, you just have to try stuff out, and become accustomed to what FL you want to use to capture an element of a scene.

[newarts]

Look through your camera's viewfinder with a 50mm lens at your fist; adjust your fist location until you see something like this:


Therefore, with your fist held like this without the camera:
1) Full fist height = frame height with 50mm lens
2) Half fist height = frame height with 100mm lens
3) Quarter fist height = frame height with 200mm lens

To compose a photo, just hold your fist in front of you; your fingers will define the edges of four potential frames:


Four fingers (50mm lens) cover the whole scene, 2 fingers (100mm lens) cover half the scene, and one finger (200mm lens) covers one quarter of the scene.

[wildman]

Just to further muddy the waters...

... it's been suggested that a way to see what a particular FOV would be like is to use a zoom.

That is to set your kit lens at, say, 55mm and what the zoom looks like at 55 will be similar to what a 55mm prime will look like also.

Is that true? I can understand that FL necessarily controls magnification but also FOV as well? I would think there could be many factors in a given lens design besides FL that may determine it's FOV.

Just an honest question and I don't know the answer for sure myself.

[WMBP]

Just to further muddy the waters...

... it's been suggested that a way to see what a particular FOV would be like is to use a zoom.

That is to set your kit lens at, say, 55mm and what the zoom looks like at 55 will be similar to what a 55mm prime will look like also.

Is that true?

Yep!

The ONLY basic, absolute difference between primes and zooms is that a prime has just 1 focal length, and a zoom by definition has many.

I can understand that FL necessarily controls magnification but also FOV as well? I would think there could be many factors in a given lens design besides FL that may determine it's FOV.

It might help if you consider this: There's no such thing as "magnification." It is simply about angle of view. If you use a telephoto lens to zoom in on a bird 150 ft away, you are simply narrowing your angle of view until the only thing you can see at that distance is the bird. Now, when you take a picture of the bird, and move it back to your computer, you can "magnify" it there to some extent, at least as far as the resolution of the sensor and the sharpness of your lens allow.

If you don't like the idea of "no such thing as magnification," then you can think of it this way, too: Field of view and magnification are the same thing.

Will

[alohadave]

It might help if you consider this: There's no such thing as "magnification." It is simply about angle of view. If you use a telephoto lens to zoom in on a bird 150 ft away, you are simply narrowing your angle of view until the only thing you can see at that distance is the bird. Now, when you take a picture of the bird, and move it back to your computer, you can "magnify" it there to some extent, at least as far as the resolution of the sensor and the sharpness of your lens allow.

If you don't like the idea of "no such thing as magnification," then you can think of it this way, too: Field of view and magnification are the same thing.

Will

Not quite. A longer focal length does have more magnification than a shorter focal length. If you were simply changing the angle of view, you'd get the same level of detail from cropping a shorter focal length, and that isn't the case.

In the case of crop effect with different sensor sizes, then yes you are correct, and you are changing the angle of view, but the magnification remains the same. a 50mm lens is always 50mm, no matter the sensor size. I don't even bother to convert focal length to 'effective' focal length because it is a meaningless comparison, especially since I don't shoot full frame or film.

[omega leader]

Practice, practice, and then practice some more.

I've gotten to the point that I can raise my 105mm to my eye and shoot once I've walked to the correct point. It's almost spooky.

Sadly I can't do it with any of my other lenses, though I'm betting better with my 12-24mm.

[WMBP]

Not quite. A longer focal length does have more magnification than a shorter focal length. If you were simply changing the angle of view, you'd get the same level of detail from cropping a shorter focal length, and that isn't the case.

Dave,

I am pretty sure that you and I don't understand this differently. However, I still don't think it's useful to say "longer focal length has more magnification." For one thing, photographers don't think about focal length in terms of magnification. Well, perhaps we do in a sort of loose way—and if that's all anybody means here, then that's fine with me, too. But we don't talk about lenses the way we talk about binoculars. We don't say, "This is an 8x lens." I know that occasionally manufacturers use this shorthand to describe the extent of the zoom range of a zoom lens, but that's slightly different, I think.


wildman asked,

I can understand that FL necessarily controls magnification but also FOV as well?

The main point of my response was to suggest that he get rid of the first part of his comment ("focal length necessarily controls magnification") and that he think instead in terms of angle of view.

It's obvious that, for a given camera or sensor, if you use a longer focal length, your subject—say, a bird in the distance—will fill more of the image area. That is, it will "look bigger." And we think of that as a sort of magnification.

But consider this: A photo taken with a 70mm lens mounted on my 14.7MP K20D will capture that bird in greater detail than a 100mm lens mounted on my 6MP *ist DS. In other words, I could shoot the bird with the K20D and crop and get a better photo than I'd get from the *ist DS. This isn't a weird optical illusion sort of example. I shoot all the time with two, sometimes even three, cameras around my neck, with different focal length lenses. (NOTE: I didn't do the math here. If I'm wrong about 70/K20D vs 105/*ist DS, then make that 70mm lens at 90mm lens and I'm pretty sure I'll be right.)

So the degree to which focal length is related to magnification, or more precisely to magnification potential is dependent not just on the lens but also on the resolution of the sensor and I suppose on the acuity of the lens, as well. Sensor size and lens acuity are NOT aspects of focal length. Which to me means that focal length should not be understood in terms of magnification, but simply in terms of angle of view.

At least that's how I see it.

Will

[alohadave]

Dave,

I am pretty sure that you and I don't understand this differently. However, I still don't think it's useful to say "longer focal length has more magnification." For one thing, photographers don't think about focal length in terms of magnification. Well, perhaps we do in a sort of loose way—and if that's all anybody means here, then that's fine with me, too. But we don't talk about lenses the way we talk about binoculars. We don't say, "This is an 8x lens." I know that occasionally manufacturers use this shorthand to describe the extent of the zoom range of a zoom lens, but that's slightly different, I think.

I agree that the terminology isn't used like in binoculars. The only place I see 8x used is in P&S and bridge/superzoom cameras, and in that case it's more of a marketing term than anything really useful.

wildman asked,



The main point of my response was to suggest that he get rid of the first part of his comment ("focal length necessarily controls magnification") and that he think instead in terms of angle of view.

It's obvious that, for a given camera or sensor, if you use a longer focal length, your subject—say, a bird in the distance—will fill more of the image area. That is, it will "look bigger." And we think of that as a sort of magnification.

But consider this: A photo taken with a 70mm lens mounted on my 14.7MP K20D will capture that bird in greater detail than a 100mm lens mounted on my 6MP *ist DS. In other words, I could shoot the bird with the K20D and crop and get a better photo than I'd get from the *ist DS. This isn't a weird optical illusion sort of example. I shoot all the time with two, sometimes even three, cameras around my neck, with different focal length lenses. (NOTE: I didn't do the math here. If I'm wrong about 70/K20D vs 105/*ist DS, then make that 70mm lens at 90mm lens and I'm pretty sure I'll be right.)

So the degree to which focal length is related to magnification, or more precisely to magnification potential is dependent not just on the lens but also on the resolution of the sensor and I suppose on the acuity of the lens, as well. Sensor size and lens acuity are NOT aspects of focal length. Which to me means that focal length should not be understood in terms of magnification, but simply in terms of angle of view.

At least that's how I see it.

Will

In your example, you are mixing different focal lengths with different sensor resolutions.

With the 6mp sensor, you have fewer sensels to record the fine detail than you do with the 14.7mp sensor. Of course you are going to see more detail in the K20D file, even with the shorter lens. The K20D sensor has more sensels, so it's going to be able to resolve more detail than the *istDS.

It doesn't change the fact that the 100mm lens is going to fill more of the frame with your subject, at the same distance than the 70mm will. The magnification is based on the focal length of the lens, but the angle of view can be changed by cropping, while the magnification does not change when you crop.

Where it gets really muddy is on the computer screen. By default, most applications are going fit the picture to the window, so when you crop the picture, and the application fits the new picture to the window, it shows an apparent increase in magnification. If you were to view the cropped portion without stretching it to fill the window, you would see this demonstrated.