I'd value your opinions on this

String

String

Messages
730
Name
Dave
Edit My Images
Yes
For ages, I have been told that the 2x crop sensor delivers an image akin to a lens with twice it's marked focal length.

Using my 14-42 kit lens I've come to an opinion about this.

My opinion is that a 14mm lens is a 14 mm lens. The sensor crops the image to fit on it, but the perspective (converging and diverging angles etc) remains the same as if it were on a full frame sensor. Similarly a 50mm f1.8 manual lens remains a 50mm lens. The perceived image size remains the same as it did on an OM film camera, but only the width and height is restricted to fit the sensor.

It doesn't bother me, other than using the 14-42 you have to be aware of the distortions inherent in a super wide lens.
? :)
 
I don't mean to be rude, but I am not sure what you are asking? or are you stating something?

yes, a 14mm lens is a 14mm lens, I was under the impression it was only the field of view that really changed?
 
Surely it's called a crop, because the sensor only collects the data from the centre of the projected image depending on the crop factor, otherwise why would my sig 10-20 show a clear image on a crop camera, but have a huge black ring around it when used on a full frame? (lens is designed for crop camera, but a lens for use on a full frame would therefore be cropped if used on a crop sensor)
 
The use of the term 'cropped sensor' has really come about from magazine articles attempting to explain relative 'field of view' with a digital camera in relation to a 35mm film camera.

The terminology, in my opinion, is irrelevant if you've never used 35mm.

Lens manufacturers have designed some lenses to work best with the smaller sensors in the average digital camera - these will not always give a clear image on cameras with a larger sensor. Unlike the other way around.
Lenses made for 'full frame' film / digital will have their field of view changed by smaller sensors relative to the frame size of 35mm film.

Somewhat confusing agreed:shrug:

Essentially I agree with you Dave
 
Last edited:
Steeps said:
Surely it's called a crop, because the sensor only collects the data from the centre of the projected image depending on the crop factor, otherwise why would my sig 10-20 show a clear image on a crop camera, but have a huge black ring around it when used on a full frame? (lens is designed for crop camera, but a lens for use on a full frame would therefore be cropped if used on a crop sensor)
Click to expand...

4/3 was designed with lenses to match the sensor so it is not like fitting a lens from one format onto a sensor of another.

The angle of view on 4/3 for a 14mm lens is akin to a 28mm on FF. The depth of field is also different on 4/3 (will be greater) which is why wafer thin DOF is harder to get.
 
String said:
I think what I'm trying to say is that people generally believe that a 50mm lens on a 2x crop is a 100mm lens

It isn't, it's a 50mm lens with a restricted field of view.
Click to expand...

It is a 50mm lens but because of the different angle of view you (and the sensor) see the equivalent of a 100mm lens on FF. It gives the impression of greater reach, which is usually what people refer to when they quote 100mm. It certainly does not behave like a 100mm lens on FF in any other way.
 
Almost_Invisible said:
The angle of view on 4/3 for a 14mm lens is akin to a 28mm on FF. The depth of field is also different on 4/3 (will be greater) which is why wafer thin DOF is harder to get.
Click to expand...

Not sure I entirely agree with that.

Try taking a picture at 14mm and the converging/diverging verticals are much more akin to 14mm than they are to 28mm
 
Yes, it's a crop factor, NOT a magnification factor, so you don't get a bigger image as some people are (wrongly) led to believe with small sensor cameras. So a 50mm lens will still be a 50mm lens. If you put a 50mm lens (designed for a 35mm size film/sensor) on a full sensor camera then crop the image down by the "crop factor" you will get exactly the same as the small sensor would give you (I'm ignoring individual pixel size here).

As has already been suggested, the difference is in the perspective. To get the same angle of view on a 2x crop sensor as, say, a 28mm lens, you would need a 14mm lens and this will lead to a different perspective.

Sadly, several photographic magazines and many staff in high street retailers insist on the "magnification factor". They should know better!!!
 
Last edited:
Somelier said:
Yes, it's a crop factor, NOT a magnification factor, so you don't get a bigger image as some people are (wrongly) led to believe with small sensor cameras. So a 50mm lens will still be a 50mm lens. If you put a 50mm lens (designed for a 35mm size film/sensor) on a full sensor camera then crop the image down by the "crop factor" you will get exactly the same as the small sensor would give you (I'm ignoring individual pixel size here).
Click to expand...

That sounds more like what I was trying to say :)
 
Harvey_nikon said:
yes, a 14mm lens is a 14mm lens, I was under the impression it was only the field of view that really changed?
Click to expand...

Depth of field is a product of aperture and magnification rather than focal length.

Here's the proof:

dof_90_f3_5.jpg

90mm Macro set at 1:1 f3.5

dof_180_f3_5.jpg

180mm Macro set at 1:1 f3.5

dof_90_f32.jpg

90mm Macro set at 1:1 f32

dof_180_f32.jpg

180mm Macro set at 1:1 f32
 
I've always considered that since DX sensors usually have higher pixel density compared to FX there is a magnification effect on DX.
For example, if you take a image with a FX camera and then take the same photo from the same position etc using the same lens on a DX camera, the images when viewed on a PC will show features in the DX image larger than the FX. ie magnified .
(The FX will of course show a wider fov)
The same would be true if the 2 images were printed at the same dpi.
The only time when the image is the same size but cropped on the DX is when the light is falling on the 2 sensors.
Some one please explain if this is wrong.
 
Astro said:
I've always considered that since DX sensors usually have higher pixel density compared to FX there is a magnification effect on DX. (...)
Some one please explain if this is wrong.
Click to expand...

Do you really consider pixel size when composing a shot.....? What about a 24 mpix current gen FF sensor compared to a previous gen 6 mpix DX? Following your logic, that's a 4x times zoom... o.O...

No, you frame the shot based on what the viewfinder shows - any post processing cropping to "digitally zoom" thanks to high pixel resolution is just an occasional bonus, and nothing to do with the sensor size.

Sent from a touchscreen thing I hit randomly with a club, using TP Forums. As if you care.
 
yeah all the lens effects (e.g. distortion) remain the same the sensor just cuts the middle bit out of the image and uses that as the final picture. that's why a 50mm will have a field of view of 100mm on 4/3, it's not magnified it's merely cropped and luckily for crop sensors they crop out the corners so edge softness and vignetting aren't much of a problem :)
 
String said:
For ages, I have been told that the 2x crop sensor delivers an image akin to a lens with twice it's marked focal length.

Using my 14-42 kit lens I've come to an opinion about this.

My opinion is that a 14mm lens is a 14 mm lens. The sensor crops the image to fit on it, but the perspective (converging and diverging angles etc) remains the same as if it were on a full frame sensor. Similarly a 50mm f1.8 manual lens remains a 50mm lens. The perceived image size remains the same as it did on an OM film camera, but only the width and height is restricted to fit the sensor.

It doesn't bother me, other than using the 14-42 you have to be aware of the distortions inherent in a super wide lens.
? :)
Click to expand...

The perspective distortion associated with a super-wide lens arises solely because the camera is closer to the subject with the subject-to-background distance remaining the same (the ratio between the camera-to-subject and camera-to-background changes radically). Put a 14mm lens on a 2x crop camera and the photographer will have to move away from the subject, just to fit the subject into the frame, compared with using the same lens on a full-frame camera.
 
String said:
The field of view is restricted to what the sensor sees, but it's still a 14mm lens and the wide angle effects are the same. I think.
Click to expand...

String said:
I think what I'm trying to say is that people generally believe that a 50mm lens on a 2x crop is a 100mm lens

It isn't, it's a 50mm lens with a restricted field of view.
Click to expand...

But, at the end of the day, the only thing the focal length of a lens does is to restrict the field of view.

I'm not sure what else you mean by the 'wide angle effects' of a 14mm lens are when put on a 4/3 body?

A 50mm lens is wide angle on a 6x6 film camera, but you wouldn't consider its 'wide angle effects' if you mounted it on a 35mm body.

:thinking:
 
From another forum
“Perspective does not change with focal length, it changes with distance between lens and subject. With a wider angle lens you are more likely to move closer to the subject, which changes the perspective, which leads people to associate the change of perspective with the change of focal length.

If you aren't convinced, take two photographs using different focal lengths, but without moving closer to the subject. Load both pics into your image editor and crop the wider angle shot to match the coverage of the other. You will see that the perspective is exactly the same.”

Now if you take two cameras one ff the other with a crop sensor which have the same number of pixels then use any lens (that fits both)say a 50mm take a photograph of the same subject from the same place, the images loaded on the pc or printed at the same resolution will be the same size but the subject with be bigger by the crop factor.
Conclusion the focal length of a lens does not change but the effective magnification and the resulting image is as if you used a lens of xx mm times the crop factor, therefore is effectively xx mm times the crop factor ( ff 50mm = 75mm dx)
 
paulminus273 said:
Conclusion the focal length of a lens does not change but the effective magnification and the resulting image is as if you used a lens of xx mm times the crop factor, therefore is effectively xx mm times the crop factor ( ff 50mm = 75mm dx)
Click to expand...

So DX does give a magnification factor. I thought that too !
:thumbs:
 
Somelier said:
Depth of field is a product of aperture and magnification rather than focal length.
Click to expand...

sensor size and distance to subject also play a part in DoF
 
Musicman said:
But, at the end of the day, the only thing the focal length of a lens does is to restrict the field of view.

I'm not sure what else you mean by the 'wide angle effects' of a 14mm lens are when put on a 4/3 body?

A 50mm lens is wide angle on a 6x6 film camera, but you wouldn't consider its 'wide angle effects' if you mounted it on a 35mm body.

:thinking:
Click to expand...

Converging and diverting verticals mainly
 
String said:
Converging and diverting verticals mainly
Click to expand...

Caused by subject to camera distance not focal length, if you move the camera back and crop the image to give the same fov of a longer lens the image would be the same as that of a longer lens
 
Yes, I'm well aware that the view from a wide angle is the same as from a telephoto and moving back lessens the effect. So I'll rename it perceived convergence and divergence.

The fact is, the effect is more pronounced on a 14mm lens than it is on a 28mm. Notice I said effect.

This is described in Andreas Feiningers' book "The complete photographer" First published in 1966, and my bible for many years.
 
Last edited:
String said:
Yes, I'm well aware that the view from a wide angle is the same as from a telephoto and moving back lessens the effect. So I'll rename it perceived convergence and divergence.

The fact is, the effect is more pronounced on a 14mm lens than it is on a 28mm. Notice I said effect.

This is described in Andreas Feiningers' book "The complete photographer" First published in 1966, and my bible for many years.
Click to expand...

I have not read this gentleman’s book and have to rely on the little bit of knowledge I gained from my work in the design and development of optics.
If I understand you correctly should you set a your camera on a tripod X feet from a subject take a photograph with your lens set at 14mm then another set at 42mm, crop so the subject is the same size there would be a perceivable difference in the convergence and divergence of vertical lines. I don’t believe so but would be happy to be proved wrong.
 
paulminus273 said:
I have not read this gentleman’s book and have to rely on the little bit of knowledge I gained from my work in the design and development of optics.
If I understand you correctly should you set a your camera on a tripod X feet from a subject take a photograph with your lens set at 14mm then another set at 42mm, crop so the subject is the same size there would be a perceivable difference in the convergence and divergence of vertical lines. I don’t believe so but would be happy to be proved wrong.
Click to expand...

If you look at a landscape taken on say a 10mm lens the skyline has a pronounced curve. Is that apparent on other lenses too?

I did say perceived effect.
 
String said:
If you look at a landscape taken on say a 10mm lens the skyline has a pronounced curve. Is that apparent on other lenses too?

I did say perceived effect.
Click to expand...

That's linear distortion, a lens aberration. Only shows up with straight lines near the edges of the frame, like buildings, or the horizon above the sea.

You shouldn't get it, and with the best lenses you don't. Easily corrected in post though, and a few cameras, mainly compacts, actually do it digitally in-camera.
 
yeah, I'm aware of that. The point I'm trying to make is that we cant all afford tilt and shift lenses, and so have to put up with reality. Convergence, divergence actually exist, but our eyes compensate for it. Want proof? look at the sky scraper through a square hole cut in a piece of cardboard and you will actually see that the lines aren't parallel. Same with railway lines. My original post was that I believed that a 14mm lens on a 4.3 camera is still a 14mm lens and the distortion you get from a 14mm lens is still the same, but the field of view is restricted by the sensor.
 
Within the limitations of manufacture, the radius of the curve would be the same; the linear deviation from the tangent is the function of X2 therefore the longer the tangent in the fov the more pronounced deviation increase
 
String said:
yeah, I'm aware of that. The point I'm trying to make is that we cant all afford tilt and shift lenses, and so have to put up with reality. Convergence, divergence actually exist, but our eyes compensate for it. Want proof? look at the sky scraper through a square hole cut in a piece of cardboard and you will actually see that the lines aren't parallel. Same with railway lines. My original post was that I believed that a 14mm lens on a 4.3 camera is still a 14mm lens and the distortion you get from a 14mm lens is still the same, but the field of view is restricted by the sensor.
Click to expand...

No, I'm talking about linear distortion, straight lines curved near the edges of the frame, such as the skyline mentioned above.

What you're referring to is not really distortion at all, it's exagerated perspective as a result of a close shooting distance, though it is often called distortion (incorrectly).

That is nothing to do with focal length, and if you shoot the same picture from the same place with a 15mm lens on a Canon cropper and then a 24mm lens on full frame, the image will be framed exactly the same, and perspective will be the the same too. The only thing that will change there is depth of field, which will be about one stop deeper with the crop camera, at the same f/number.
 
String said:
yeah, I'm aware of that. The point I'm trying to make is that we cant all afford tilt and shift lenses, and so have to put up with reality. Convergence, divergence actually exist, but our eyes compensate for it. Want proof? look at the sky scraper through a square hole cut in a piece of cardboard and you will actually see that the lines aren't parallel. Same with railway lines. My original post was that I believed that a 14mm lens on a 4.3 camera is still a 14mm lens and the distortion you get from a 14mm lens is still the same, but the field of view is restricted by the sensor.
Click to expand...

Move further away with your piece of cardboard and the Convergence, divergence decreases ergo distance not lens .
 
Yes. I called it distortion because it's how people see it.

As for the image frame of a 15mm on a cropper and full frame, I'm not sure I agree. On my camera a 24mm lens allows me to see much much more in the viewfinder of an om than it does through the viewfinder of my e510
 
paulminus273 said:
Move further away with your piece of cardboard and the Convergence, divergence decreases ergo distance not lens .
Click to expand...

Geez! I know. I'm not a mathematician I'm a simple guy with a camera. I understand what you are saying. Obviously all you want to do is beat me over the head with your superior knowledge.

Read the OP
 
String said:
Yes. I called it distortion because it's how people see it.

As for the image frame of a 15mm on a cropper and full frame, I'm not sure I agree. On my camera a 24mm lens allows me to see much much more in the viewfinder of an om than it does through the viewfinder of my e510
Click to expand...

You're not taking account of the crop factor.

Your Olympus has a 2x crop factor, so in terms of field of view compared to full frame, you would need a 12mm lens to match 24mm, which is a big difference. Canon has a 1.6x crop factor, so 15mm = 25mm; on a Nikon with a 1.5x crop factor, it's 16mm = 24mm.

And the exagerated perspective you can get with wide-angles used close is not distortion, because that's how the view actually appears from that viewpoint. The lens just records it. Move back, and it goes away.
 
HoppyUK said:
You're not taking account of the crop factor.

Your Olympus has a 2x crop factor, so in terms of field of view compared to full frame, you would need a 12mm lens to match 24mm, which is a big difference. Canon has a 1.6x crop factor, so 15mm = 25mm; on a Nikon with a 1.5x crop factor, it's 16mm = 24mm.

And the exagerated perspective you can get with wide-angles used close is not distortion, because that's how the view actually appears from that viewpoint. The lens just records it. Move back, and it goes away.
Click to expand...

That was exactly my point. Thank you
 
String said:
Geez! I know. I'm not a mathematician I'm a simple guy with a camera. I understand what you are saying. Obviously all you want to do is beat me over the head with your superior knowledge.

Read the OP
Click to expand...

You may well be better at using a camera than I, but having read the OP you seem to have a misconception about focal length and cropped sensors.
If you use a lens of xmm on crop sensor with the same number of mp as a ff sensor you will get the same image as xmm times crop facture on the ff, there may be differences in IQ , dynamic range and tonal quality but one will fit exactly on top of the other.
 
You must log in or register to reply here.
Back
Top