Visualizing Swing + Tilt together

[G Benaim]

Hello,

I have a pretty good idea of how to use and visualize each of these movements separately, but have a hard time seeing their effect together. Say you have a plane extending horizontally away ( a sidewalk, e.g.) and one vertically (e.g. a row of buildings along the sidewalk). Independently, one can easily see how tilt and swing, respectively, will help you match the plane of focus to the external plane. But what happens when they're used together? And how does one visualize it? Thanks for any help,

GB

[Leonard Metcalf]

I imagine the plane of sharp focus first (then I add in the curves of sharp focus - planes are not really all that will be in focus). Just in the same way as you can imagine a vertical receding plane as a swing, and a receding horizontal plane as a tilt. A diagonal plane will have a mixture of both. Most people would work out their movements in two steps, one for the horizontal and the other for the vertical, unless using a camera like a master technica which allows simultaneous adjustment of both planes (this is actually a joy to use in this situation).

Another way is to imagine three points in the image that you want to be in sharp focus, and use these to imagine your plane off.

Perhaps an example would help.
I firstly focus on point 2 as I am using a camera with asymmetrical movements. Point 2 is the point on the ground glass where everything hinges on (both swing and tilt) I focus on this one point.

Then I focus on point 1 by tilting the lens. Then I swing point three into focus as well. Finally I stop down the lens to ensure that my depth of field is appropriate to the image.



I hope this helps, regards,

Len

[Leonard Evens]

Keep in mind the shape of the region of adequate focus. It is always going to be a wedge centered on a hinge line. That hinge line will be the intersection of the plane of exact focus with a plane through the lens parallel to the film plane. For a tilt, it will be (usually) below the lens. For a swing, it will be a vertical line to the left or right of the lens. For a more general movement, like what you describe, it will be tilted with respect to both the vertical and the horizontal. Try to imagine where you want the plane of exact focus to be and then the resulting hinge line. When you finish with tilt and swing separately, the front standard should be parallel to this desired hinge line.

For the kind of scene you describe, it will be difficult, if not impossible, in the foreground to have both the sidewalk and the facades in adequate focus. That follows from the wedge shape and the fact that the apex of the wedge, the hinge line, will be skew with respect to both vertical and horizontal. You will probably have to decide which is more important to you, the vertical or the horizontal, get that in good focus with a tilt or swing as appropriate and then see if you can keep it in focus by using a slight perpendicular movement (swing or tilt as appropriate). The three pont method that Leonard M. recommended should enable you to do it if it is feasible, but I think there are few situations where it will actually work.

[G Benaim]

LM: Thanks for the diagram, that is very helpful.

LE: By wedge, do you mean 2-d or 3-d, i.e. a cone? I thought it wouldn't work, but it's likea puzzle that's kept going around in my mind. I can get the vertical plane w a swing, I can get the hor w a tilt, now how do I do both? I guess you get the second w dof as much as possible, right? Or does it make more sense to avoid movements and just stop down?

[N Dhananjay]

The DOF limits for a wedge - like a (very!) thick pizza slice, not like a cone.

As Leonard explained, if you use tilts and swings together, the plane of best focus is neither horizontal (parallel to the bed of the camera) nor vertical (to either side of the camera) but is angled to both planes. Say you are standing on a sand dune and want to photograph the ripples on the sand. The sand dune is rising in front of you but it is also rising from side to side such that it is lower on the left and higher on the right. Then, a combination of swings and tilts helps to get the ripples in focus. For the kind of situation you describe (sidewalk below and a building facade to be in focus), swings and tilts will not help you since they help you to get one plane into focus, whereas here you are trying to simultaneously get two planes into focus.

Cheers, DJ

[G Benaim]

DJ: That's a very clear explanation. So what's the best way to photograph two planes?

[Ole Tjugen]

So what's the best way to photograph two planes?

There isn't one.

The "best" way is to focus on the far point, and stop down as far as you can. There's a narticle by Harold merklinger somewhere on the net which explains the rationale behind this - and it really does work.

Back in the wet-plate days photographers often managed the seemingly imossible task of shooting street scenes that were sharp all over. They did this by exploiting the faults of their lenses! A simple meniscus lens has "horrible" field curvature, which means that the sides are in "near" focus when the center is focused on infinity. Some lens faults can occasionally be useful...

[Leonard Metcalf]

I have attached a diagram from p44 Stone J 1997 2nd Ed "A users guide to the view camera"... which shows the wedge that the other Leonard describes.. note that it is curved. (click on the thumbnail to see an enlarged view)

Following on from Ole Tjugen, Merkinger is worth seeing. In the article "Principles of View Camera Focus" by Harold Merkinger there are some fantastic little animations... these are fantastic for visualizing things such as the hinge rule... they can be down loaded from here.

Can I also suggest practicing at home... there is nothing like setting up and focusing the camera on various objects and see what you can do... stopping down the lens will work to a point before it gets to dark to see what is going on... then you need to take some snaps...

I particularly like the visualization of a wedge (thanks Leonard)

Regards,

Len

Attachment 200

[Leonard Evens]

Imagine a note book with rigid covers. Open it up so it has a triangular cross section. Now imagine the binding reduced to a line and extended infinitely and imagine both covers extended infinitely. That is what the region looks like. When you see diagrams in books or articles, they usually show you just a perpendicular cross section. The hinge line or apex of the wedge becomes a point, and the bounding planes become lines. That can be very misleading unless you have the three dimensional model in mind. (Mathematically, the boundaries are actually called half planes since they extend in just in one direction from the hinge line.

Note that although the DOF region is in principle infinite from side to side along the hinge line, in practice, you also have to consider the angle of coverage of the lens. This does produce a cone centered on the lens which intersects the DOF wedge. Finally, the placement of the film frame will project outward and further limit what is in the field of view.

[Leonard Evens]

I've also seen the picture Leonard M. took from Stone's book before. But I am not sure just where it comes from. It is true that, in theory, the boundaries of the DOF region depart from planes. The reason is that since the exit pupil is not parallel to the film plane, the circles of confusion become ellipses of confusion. But I haven't seen any analysis which suggests that the departure from planes is as large as this diagram suggests. Wheeler, in his notes, analyizes the departure along the midline and finds it to be very small in practical situations for large format photography. I did my own analysis over the full field and found similar results. This was the topic of a long discussion either here or in photo.net several years ago.

Merklinger also proceeds on the assumption that the boundaries are planes.

It could be that Stone's diagram comes about from experimental measurements for some specific lens, which had large deviations from what geometric optics would predict. Surely there must be such deviations. For example, due to field curvature, the plane of exact focus is not even going to be a plane. But it would be surprising if it were as large as the diagram indicates. Also, it would be surprising to find that DOF increases as you go further out in the field, as the diagram seems to suggest. I will have to track down just where that diagram comes from.