Crop factor : the consensual explanation ?

[yanidel]

I have never used full frame cameras with M series. Yet, I have read many explanations on the crop factor and I see a lot of diverging opinions. Goal of this thread, validate some facts, this is my current understanding based on what I have read :

- the crop factor is the ratio between full frame and the digital sensor ?
- a lens always keep the same magnification, whatever the crop ratio is ?
- it is the area framed that changes, a 50mm lens with a 1.5 crop ratio will give a picture with the same area as a 75mm on a full frame ?
- therefore it is wrong to use a 50mm for portrait thinking it is a 75mm ? The area will that of a 75mm but the proportions of the subject will be those of a 50mm lens
- sharpness in the corners should be better than a full frame camera as it only takes the central part of the lens ?
- the lenses render exactly the same on full frame and smaller sensor, except if there is a difference between corners and center? (flare, distortion ...)
- DOF is the same whatever the sensor ?

I might have missed some topics, let me know. tks.

 

[M. Valdemar]

It's really very simple and not complicated.

Your lens creates a circular image. You make a picture with a portion of that image.

Imagine you had a TV set with a round picture tube (all TV's used to have these). When you watch a movie, the picture fills the whole tube.

Now, you take a piece of paper and cover the picture tube, and then cut a small rectangle in the middle. You then watch that.

You only see the part of the picture not covered by paper, but it's still there, only you can't see if because it's covered.

Nothing else changes. The perspective of the image, the size of the image, etc etc. You just see less.

That's all that a crop factor is.

Focal length, perspective, the image thrown by the lens, nothing changes.

A 100mm lens is just a 100mm lens. It doesn't become a 150mm lens because you change the "crop factor". Take a "full frame" image and crop a rectangle out of the center of it in Photoshop. You have just created a "crop factor".

There is no mystery or anything difficult to understand.

http://en.wikipedia.org/wiki/Crop_factor

 

[rxmd]

MV is right. However, difficulties come about when you start looking at pictures of equal size, one with a 50mm lens on a 24x36 sensor, the other from the same lens on a crop factor. Suddenly you notice that both pictures show considerable differences in perspective and slightly less considerable differences in DOF.

We've had a number of threads on this at RFF and in the end most of them turned into a flame war.

 

[kshapero]

I agree with MV but with longer lens were there is little or no barrel distortion, so say with a 90mm, it really seems like a 135mm on the R-d1.

 

[ferider]

Most of the confusion comes from people mixing the terms "physical focal length" and "equivalent focal length".

Physical focal length is measured without camera. Equivalent focal length is given based on field of view with a given format when compared to 35mm format.

One of the basic principles of photography is: if you take a picture with a focal length of f1 and another one with a longer focal length f2, on the same format, the perspective visible in the second picture is _identical_ to the perspective of the first when cropped by factor f2/f1.

DOF is a term defined based on final enlargement and viewing distance. Therefore, if you take the same lens and use it on a cropped medium, the resulting DOF on the cropped medium is _less_, since it has to be enlarged more.

For example:

A lens with 80mm physical focal length on medium format will create _more_ DOF than an 80mm lens on 35mm format (same aperture and distance).

A lens with 80mm equivalent focal length (120-150mm physical focal length depending on exact format) on medium format will create _less_ DOF than an 80mm lens on 35mm format (same aperture and distance).

Best,

Roland.

 

[M. Valdemar]

It's not that there's less barrel distortion, the same distortion is there but you just don't see it because you're only viewing a small portion of the image so it seems to be less.

I agree with MV but with longer lens were there is little or no barrel distortion, so say with a 90mm, it really seems like a 135mm on the R-d1.

 

[yanidel]

So when Panasonic says its 14-150mm is a 28-300mm (among many other manufacturer), this is not correct, just misleading marketing.
People think, they are getting a huge telezoom, but they are just getting a smaller crop at maximum focal.

 

[ampguy]

It's all pretty much self correcting. Rangefinders show you +/- 30% of what you get on the negative, so a 1.3 or 1.5 crop factor is just going to go a bit tight.

I've used 28s, 35s, 40s, 50s, 52s, and 75s on both film and RD1s, and even when forgetting that the 75 was on with the 50 framelines, I rarely lost anyting on the edges that I wouldn't normally lose during the developing, negative cutting, scanning, and printing process.

Use the lens you used with film, on your digital rangefinder.

Only the Nikon F3 gives you 100% of what you see through the VF.

 

[Thardy]

Crop Factor: Knock yourself out.

 

[rolo]

I have never used full frame cameras with M series. Yet, I have read many explanations on the crop factor and I see a lot of diverging opinions. Goal of this thread, validate some facts, this is my current understanding based on what I have read :

- the crop factor is the ratio between full frame and the digital sensor ?
- a lens always keep the same magnification, whatever the crop ratio is ?
- DOF is the same whatever the sensor ?

I might have missed some topics, let me know. tks.

As above the view seen by the lens remains the same regardless of receiver size, but it isn't the same picture.

However, depth of field and bokeh effect are significantly affected by camera position. So, with any specific lens you have to move backwards to get the view in on the crop sensor. That undermines the composition/angles and the depth of field. With say, a 35 Lux on a FF you might get in close for a portrait and blow the background out with max aperture. To get the same field with that same lens and f stop, you walk backwards 1/3 extra, the relationships in the view have changed completely and the bokeh has firmed up.

If you use a shorter lens to maintain a similar view, you screw the characteristics altogether. None of that is to suggest you can make great pictures, but you can't make the same pictures.

To compensate to a degree, you need to use the lenses wider, but if you already shoot widest on FF, you can't recover that.

 

[Roger Hicks]

Only the Nikon F3 gives you 100% of what you see through the VF.

And F, and (I believe) F2 and Visoflex (minus the rounded corners) and Alpa and...

Cheers,

R.

 

[pvdhaar]

Actually, understanding this crop factor thing is very easy..

Imagine that you are peeping through a key hole.. 😱

Your eye is still the same as before, what you see on the other side isn't bigger or smaller than before, there's just less of it visible.. 🙄

You understand that? Cool! 😎

Now, digital with a crop factor is exactly like that, only what you get to see is a lot less interesting 😉

 

[ampguy]

Thanks for the clarification Roger

Thanks for the clarification Roger

I know the specs for the F and possibly F2 were 100% when sold, but are they still 100% when later lenses are attached, like say a 24mm or wider on an original F? It couldn't have tested, at the time of manufacture.

And F, and (I believe) F2 and Visoflex (minus the rounded corners) and Alpa and...

Cheers,

R.

 

[ampguy]

Hi Roland

Hi Roland

I think a better word than "perspective" is just size or viewing area. To me, perspective also includes things like distortion and compression.

So if f1 were 21mm, and f2 were 50mm, your crop by dividing 50/21 gives the same size or viewing area, but the distortion and "perspective" is likely way different with the behavior of the 21mm.

...

One of the basic principles of photography is: if you take a picture with a focal length of f1 and another one with a longer focal length f2, on the same format, the perspective visible in the second picture is _identical_ to the perspective of the first when cropped by factor f2/f1. ...

 

[rxmd]

So if f1 were 21mm, and f2 were 50mm, your crop by dividing 50/21 gives the same size or viewing area, but the distortion and "perspective" is likely way different with the behavior of the 21mm.

Try it.

I know this sounds counterintuitive to anyone who equals "wide = a particular overall perspective", but if you take a crop from the center of an ultrawideangle shot so that the angle of view equals that of a normal lens, the perspective will be that of a normal lens, too.

Just try it. It's an experiment that changes a lot of the assumptions we make about focal lengths and perspective. I think every photographer should at some point be forced to use medium and large format lenses on various formats of film, then nobody would even start this kind of discussion.

Philipp

 

[ferider]

I think a better word than "perspective" is just size or viewing area. To me, perspective also includes things like distortion and compression.

So if f1 were 21mm, and f2 were 50mm, your crop by dividing 50/21 gives the same size or viewing area, but the distortion and "perspective" is likely way different with the behavior of the 21mm.

...
One of the basic principles of photography is: if you take a picture with a focal length of f1 and another one with a longer focal length f2, on the same format, the perspective visible in the second picture is _identical_ to the perspective of the first when cropped by factor f2/f1. ...

You are wrong, Ted (except for non-rectilinear distortions); compression will be identical. Like Phillip recommended, try it.

Roland.

 

[ampguy]

how much is "the center"

how much is "the center"

If I take a 100% crop of one of my wide lenses, it still has some wide pincushion distortion that are not in any 50mm that I'm aware of.

Try it.

I know this sounds counterintuitive to anyone who equals "wide = a particular overall perspective", but if you take a crop from the center of an ultrawideangle shot so that the angle of view equals that of a normal lens, the perspective will be that of a normal lens, too.

Just try it. It's an experiment that changes a lot of the assumptions we make about focal lengths and perspective. I think every photographer should at some point be forced to use medium and large format lenses on various formats of film, then nobody would even start this kind of discussion.

Philipp

 

[ampguy]

that's the key Roland - rectilinear

that's the key Roland - rectilinear

There are no true 100% rectilinear wide lenses, except for maybe the 35 asph cron.

Here is an 18mm photo, and a 40mm photo. Note the aspect ratio of legs to body. The 18mm is not a fisheye, it shows the wide distortions of most any wide lens.

You are wrong, Ted (except for non-rectilinear distortions); compression will be identical. Like Phillip recommended, try it.

Roland.

 

[rxmd]

If I take a 100% crop of one of my wide lenses, it still has some wide pincushion distortion that are not in any 50mm that I'm aware of.

Pincushion distortion is a different matter altogether. That's an optical defect, which can be corrected or not in a particular lens. I think Roland and I are just talking about perspective. Use a good lens with rectilinear rendition and try it out, but do try it.

See attached a shot from the CV 15/f4.5 and an 1:1 crop of the center section:

This should be about the equivalent of a 40mm lens. With a better scan I could have cropped even more.

Try this for yourself with ten different pictures and if you really want to be sure take a couple of comparison shots with the 50. I bet you a beer that there will be no difference in perspective that is not due to optical defects of the lens 😎

Philipp

 

[rxmd]

There are no true 100% rectilinear wide lenses, except for maybe the 35 asph cron.

Oh well that wouldn't have been my first example, but yes there are. Generally wide angles are rectilinear if they are 100% symmetrical in construction. Then you get formulas like the Zeiss Topogon, which was meant for topographical applications where the lens absolutely has to be rectilinear. A Topogon (or a Russian Orion-15 28/f6, which is a symmetrical lens, hence slow) will probably be more rectilinear than anything Leica has to offer.

Here is an 18mm photo, and a 40mm photo. Note the aspect ratio of legs to body. The 18mm is not a fisheye, it shows the wide distortions of most any wide lens.

The aspect ratio of legs to body is because you shoot down from the top and compare things that are not in the center of the frame. That's normal with a wideangle, but when you crop you crop away these portions. You have to crop from the center, just like a digital camera sensor would.

For this kind of comparison, you mustn't move the camera between shots, and you must crop from the center. Here, you would have had to kneel down so that the boy is in the center.

Philipp