Physics question

[maxernest]

Hey!

I want to to take a photo in which half of the final image is covered (black). I've been doing some reading and found out that, somehow counter-intuitively, if you cover half the lens with a black sheet, you don't hide what's behind it, but just make the image less bright. This is due to that even though there are some light rays covered, some of them still arrive to the lens through the non-convered part. Link for diagram and further explanation:http://www.physics.usyd.edu.au/super...alf_a_Lens.pdf

My question is, how do you actually cover part of the image with a sheet in between what need to be covered and the camera? Is this dependent on the objective's focal length? I am currently using a 50mm objective.

Thanks,

Keep shooting,

M.

 

[Ian Grant]

Cokin make a filter to do what you're asking about, but you could use black card in a filter holder.

Ian

 

[RobC]

Its more complicated than the diagram in your link makes it look.

Firstly at the edges of the lens they won't bend the light nearly as much as it shows so whoever produced that diagram doesn't know what they're doing. Secondly the diagram does not include an aperture. It is assuming that aperture is full lens diameter. That is very very rarely the case. Again, whoever drew that diagram doesn't know what they are doing.
Suggest you actually try what you're suggesting at different apertures and see what actually happens. Then you'll have a clue.

p.s. you can probably design a lens to do what the diagram shows but don't assume that a camera lens design will do it. Test it for yourself. You have a camera, you have a lens, you have a piece of card. Why do you need to ask. Just do it.

 

[Gerald C Koch]

If you use a piece of cardboard or a lens filter to cover half the lens surface the separation between the two halves will not be sharp. What you will see is a fuzzy area separating the two. Remember that all portions of a lens image all portions of a scene. As EobC points out things are never as simple as first seen.

 

[maxernest]

Thanks guys, I think you made me realize that it isn't as easy as I first thought. I will give a few trial pictures, let's see how it turns out.

 

[markbarendt]

Remember that all portions of a lens image all portions of a scene. As EobC points out things are never as simple as first seen.

The way I finally "got" that this was real was shooting through a chainlink fence. The camera can see through it better than we can as long as the subject we are focussing on is far enough on the other side, that the fence is completely out of focus.

 

[RobC]

Thanks guys, I think you made me realize that it isn't as easy as I first thought. I will give a few trial pictures, let's see how it turns out.

It is easy, its very easy. The only thing is that you will never get a sharp cutoff between the two halfs if your mask is in front of lens. The smaller the aperture the sharper it will be but not 100%.

If you can do it in printing then its simple. You just mask half the print.

If you tell us exactly what you're trying to do and why then maybe you'll get some suggestions on best way to proceed.

 

[RobC]

Remember that all portions of a lens image all portions of a scene.

Yes but not necessarily on the receptor.

 

[Stephen Benskin]

The film industry uses a matte box to mask the edges. In camera special effects also use matte boxes. The sharpness of the edge depends on the distance from the lens and the depth of field.

 

[summicron1]

Actually, this is something that used to be done quite a bit to have multiple images of the same person in one photo.

For example, they used to make an attachment to put over the lens that looked like a lens cap, but that had a half-moon shape cut out of one side of it. You would put it on the lens, take a picture of someone standing on the same side as the opening, then rotate the cap to the other half, take a second picture, and there you are: Twins! I saw caps like this advertised as long ago as the 1920s.

It's also a great way to shoot someone standing next to their guardian angel -- just double expose one half with them standing there only once.

Some folks got more creative than that -- constructing blind-type things in front of the camera with multiple openings so you could shoot the same person half a dozen times or more all on the same frame, opening and closing doors as they moved around. It took a little planning, but worked nicely. I even saw creations in LIFE magazine using this techcnique.

Photoshop has eliminated the need, but you can still play.

 

[DannL.]

You may find that covering half the lens (front) with a black card will work with some lenses, and not with others. But if you can cover the window nearest the film plane with a "blocking device" ie; tape or card, your problem is solved. In the "view-camera world" covering half a sheet of film can be accomplished with a dark-side. This can be done regardless of the type of lens used.

 

[NedL]

Or use a pinhole camera. I sometimes do in-camera dodging and burning with black card.

 

[absolutedesmond]

Its more complicated than the diagram in your link makes it look.

Firstly at the edges of the lens they won't bend the light nearly as much as it shows so whoever produced that diagram doesn't know what they're doing. Secondly the diagram does not include an aperture. It is assuming that aperture is full lens diameter. That is very very rarely the case. Again, whoever drew that diagram doesn't know what they are doing.
Suggest you actually try what you're suggesting at different apertures and see what actually happens. Then you'll have a clue.

p.s. you can probably design a lens to do what the diagram shows but don't assume that a camera lens design will do it. Test it for yourself. You have a camera, you have a lens, you have a piece of card. Why do you need to ask. Just do it.

Hello Rob,

I have to disagree. The people who drew the diagram knew mostly what they were doing. This is the representation of a thin lens in geometrical optics. What is confusing is that in geometrical optics thin lenses are actually representend by a line. Thus the size of the lens in the diagramm would not necessarily represent the physical size of the lens. Secondly the focal points are missing which is the most important thing in geometrical optics. Thirdly the optical axis is a bit to low.

Also it is perfectly possible to construct an optical array without an aperture. The defining characteristic then is the entry pupil.

What most of the people seem to be missing here is that this is a SINGLE symmetric biconcave Lens. This does not at all behave like a photographic lens.
For example in a view camera with 150 mm lens if i want to focus to infinity I have to extend the bellows to 150mm.
With a biconcave lens With a bellow extension of 150mm I would focus to 150mm in front of the lens. Or for infinity focus you'd have to extend the bellows to infinity. Not very practical!

So what you are in fact seeing in the experiment is a sort of bokeh (like with the chainlink fence). If you were to repeat the experiment and bring the object much closer to the lens (and thus also the focus plane) you would start to only see the upper half of the object and a blurry line to the where the lens has been covered.
Hope I could help

BR
Thomas

 

[RobC]

You forgot to mention scale. If that tree is 100 feet tall then that lens is bloody big. It's a brainless graphic which tells nobody anything. Its lacking an explanation. It may be an exam question but I would be complaining that the graphic is mis-representing the relationships and is therefore WRONG.

p.s. Gauss determined that a thick lens with many elements behaves the same as a single thin lens. And a thin lens would NOT bend light as much as that incorrect representation.

Note: The OP was fooled by it. It follows many others would be too unless they understood a lot more about optics than they do.

 

[Hatchetman]

why not just tape a piece of black paper between the shutter and the film

 

[ic-racer]

I've been doing some reading and found out that, somehow counter-intuitively, if you cover half the lens with a black sheet, you don't hide what's behind it, but just make the image less bright..

When you get your camera you can put a card over the lens to see that you were mislead by what you read.

 

[absolutedesmond]

You forgot to mention scale. If that tree is 100 feet tall then that lens is bloody big. It's a brainless graphic which tells nobody anything. Its lacking an explanation. It may be an exam question but I would be complaining that the graphic is mis-representing the relationships and is therefore WRONG.

You have a point. I would still argue that it is not wrong, but definitly not very explanatory. Still I believe that the confusion arose from the type of lens and not from the scale of the diagram

Edit: I am afraid I was completely wrong! The lens type doesn't matter here. Optics 101 is more in my past than i thought Thanks for an opportunity to freshen up my knowledge

 

[wiltw]

Mask AT the film plane, edge will be sharp.

 

[Jim Jones]

The experiment cited by Maxernest is more about logic than about optics. As Maxernest claims, it would result in the image being underexposed from edge to edge by a factor of 1/2. In more practical applications of blocking out half of the image, whatever masks the image will either be near the film or some distance in front of the lens. I usually use a distance of a few focal lengths in front of the lens. If the camera and background don't move, one can make several exposures of a subject at different places on the film as Summicron 1 mentioned. This requires precise placing of the mask. Purists may remind us that, even with perfect placing of the mask, there may be some slight blurring of the image at the masked edge due to the characteristics of some lenses. In my experience, one will unlikely notice this. There is also the fade-out zone of the image which increases as the aperture diameter is increases and decreases in inverse proportion to the distance to the mask. The best way to gain understanding of the whole project is to do it with variations.

 

[Nodda Duma]

Mask AT the film plane, edge will be sharp.

This is the second best way. The best way is to do it at the enlarger. Straight from the horse's (lens designer's) mouth.

 

[MattKing]

Or in the studio, with lighting.

 

[maxernest]

Actually, this is something that used to be done quite a bit to have multiple images of the same person in one photo.

For example, they used to make an attachment to put over the lens that looked like a lens cap, but that had a half-moon shape cut out of one side of it. You would put it on the lens, take a picture of someone standing on the same side as the opening, then rotate the cap to the other half, take a second picture, and there you are: Twins! I saw caps like this advertised as long ago as the 1920s.

It's also a great way to shoot someone standing next to their guardian angel -- just double expose one half with them standing there only once.

Some folks got more creative than that -- constructing blind-type things in front of the camera with multiple openings so you could shoot the same person half a dozen times or more all on the same frame, opening and closing doors as they moved around. It took a little planning, but worked nicely. I even saw creations in LIFE magazine using this techcnique.

Photoshop has eliminated the need, but you can still play.

Is this what you mean? the dual image filter? looks like a lot of things can be done with it, looks promising

http://ecx.images-amazon.com/images/I/41+tWdXdujL._SY300_.jpg

 

[BAC1967]

This is what you need.

The Splitter by Bryan Chernick, on Flickr

That's how I did this:
Cedar in the Sky by Bryan Chernick, on Flickr

It does give a soft edge though.

 

[maxernest]

This is what you need.

The Splitter by Bryan Chernick, on Flickr

That's how I did this:
Cedar in the Sky by Bryan Chernick, on Flickr

It does give a soft edge though.

Beautiful picture, I'm looking to experiment with something like this. Thanks