Applying movements

[Anupam]

I have been shooting LF for a while and am generally comfortable with all movements and focusing. I have also read the regular Simmons, Dykinga etc books although I don't have them with me right now. But I was wondering if there is a better way to use movements than I have been currently doing. Leaving aside rise/fall/shift which are relatively unproblematic this is what I do:

* compose roughly
* correct for perspective if required using back swing/tilt
* deal with plane of focus using front swing/tilt

Now recently someone mentioned to me that he uses back swing/tilt to deal with plane of focus because "then you keep within the image circle of the lens and don't risk running out." Makes sense, but then how would you deal with perspective, or wouldn't that change the perspective by default? Maybe this isn't such a problem for scenes where perspectives isn't crucial, but I would like to hear your thoughts on this. How do you apply movements in the field, tips, tricks pointers would be welcome.

Thanks,
Anupam

[Anupam]

As an addendum to that, here is an image that can be used as an example. It's a 35mm image made by someone else, but I asked permission to link it here:


How would you get the entire wire fence and the hill in focus. It seems the wire fence would need right swing and then tilt to bring the hill in focus. You'd still have to stop down a bit to get the wire fence from bottom to top sharp. This requires a lot of movement and a LARGE image circle. Could this be down with back swing and front tilt using a lesser image circle? Dividing up the movements as it were?

Thanks,
Anupam

[panchro-press]

First of all, perspective will only change if you hoist your camera and tripod to another location. No lens will change perspective.
Second, with movements, 'less is more'. With the image you show, did you try a slight front tilt?

-30-

[Anupam]

First of all, perspective will only change if you hoist your camera and tripod to another location. No lens will change perspective.

Well, yes, I know that but you know what I mean, of course. I haven't tried shooting this scene. I wasn't there - I am just trying to use it as an example to get a better understanding of things. It's the best I could find at hand - perhaps just tilt would do for this image but let's take it as an instance where such a complex movement might be needed. I am trying to get my head around the concept here rather than the practicalities.

-A

[Eric Woodbury]

There are two planes here: the ground and the fence. You can't get them both without stopping all the way down and then you will eat some diffraction or movement. You can get the fence with a swing and the ground with a tilt. A tilt will 'distort' the mountain, adding convergence, but since we don't see it as a regular force, who's to know? I'd either stop all the way down or recompose.

[Alan Davenport]

Anupam,
It is only possible to have a single plane in sharp focus, somewhere in the 3D subject space. Thus it would be impossible to have both the fence and the distant hill in perfect focus -- you would have to rely on depth of field to provide adequate focus.

Now, if I were shooting the supplied fence/hill image, and wanted everything to appear sharp, I would use swings to place the plane of focus so the foreground end of the fence was perfectly focused, then the plane of focus would extend somewhat toward the hill so that it (focus) passes halfway between the hill and the fence at that distance. Then stop down as needed to get everything within the limits of the DoF.

Just looking at the example, I doubt if any tilts would be needed or even desirable, since that would tend to defocus the fence in the foreground area.

My goal with swings and tilts, is to place the plane of focus in the optimal location, meaning a location that minimizes the focus spread (amount of refocusing needed to put the near and far subject matter into sharp focus.) Then based on that amount of focus spread I can choose an aperture that will provide the sharpest image through the subject area. See the Peterson/Hansma articles for details on this (linked elsewhere on this site.)

[Anupam]

Thanks. I know there is only one PLANE of focus, but it's about choosing the optimal plane and having to depend on DOF beyond that.

Any thoughts on my original post would be much appreciated - the picture was just an example.

-A

[Patrik Roseen]

I have been shooting LF for a while ... How do you apply movements in the field, tips, tricks pointers would be welcome.

Thanks,
Anupam

Hello Anupam,
I am in no way an expert in LF but here is how I usually do.
I use a Linhof Technika III in the field where the back is rather easily adjustable in swing/tilt planes at the same time, whereas the front is limited in tilting forward unless the whole front bed is dropped and then backward tilt is applied and finally the front is given some raise. (the Linhof back is really adjusted by pulling out the four movable corner pins holding the back the back...so it is not really a true swing I have learned)
For this reason I usually start by applying movements on the back and if necessary add tilt and swing in the front. As you have stated already a limited image circle can still be used when applying most of the movements on the back. Ofcourse there are situation where this does not work ... circular objects nearby become egg shaped and strong lights/reflections can become egg shaped too.

Regarding how to actually place the plane of focus, this is really a matter of taste and idea of the picture. There is usually some kind of trade off. The DOF will be different on either side of the plane and I have a hard time remembering/figuring out which side is which when the plane is altered. My own experience is that the DOF is usually easier accomplished if the plane is only tilted compared to when swing has been applied too, but maybe this is only my subjective feeling/mistake?

[Leonard Evens]

I'm not sure what you mean by correcting perspective by tilts/swings of the back. Such movements of the back change the plane of focus and also the relative shapes of objects. As has been mentioned, tilts/swings of the back have the advantage of not moving the section of the image circle in the frame, but they produce other effects, particularly "distortions" of image shape, which may not be desirable. Of course, if you point the camera up, you can tilt the back to keep it vertical to avoid converging verticals.

It is true that perspective, in the sense of what is along each line of sight through the lens, depends only on the position of the lens. But the term is often used more generally. For example, if the lens is in the same position, the perspective for the back vertical and the perspective for the back tilted at an angle are the same, but most people would consider converging verticals a "perspective distortion".

As to the example you gave, I agree that it is not going to be possible to get both the fence and the ground in the foreground in focus. The problem is that the two planes you want in focus are perpendicular and both very near the lens. Probably the best solution is to concentrate on the fence and just use swing.

[Donald Miller]

In situations like the example that you provided, where two planes of focus exist, I would opt to have the camera movements take care of the greatest distance front to rear. In this case that would be the distance from the nearer fence post to the distant hill. Thus the optimal movement, in this case, would be the use of swings. Now if you have front as well as rear swings, and you wished to maintain...so much as possible the size of the nearer object (the nearest fence post) I would opt for using front swings first. If you wanted more of a distortion of a near/far relationships than rear swings would be optimum.

The lens aperture would then be used to control the front to rear sharpness (represented by the front to rear distance of the fence).

I believe (I could be wrong) that everything in this scene could be in focus when exposed.