Scheimpflug Rule when near is higher than far

[Tim Shawcross]

Every time I see an explination of the Scheimpflug Rule it includes an example like a rock in the foreground and then a mountain (which is much higher) as the far point. Using front tilt, tilting the standard forward allows the film plane, subject plane and lens plane to meet at a single point. I understand this much and have used it successfully in the field.

What to you do though when your foreground is higher than the background. See first picture attached - I shot this a week ago (one of the first exposures I made with my new Chamonix!). I simplly used the focus near, focus far, observe distance in between, focus at the middle point and stop down. Incidentally, the results using this method were not that succesful for this shot, as with my 135 stopped down to F32 I didn't get adequate sharpness throughout the image.

In this situation, can I use front tilt, except tilting the front standard backwards? See my little sketch -red lines represent film plance, lens plane and subject plane. To me this should work, but I've never tried it in the field. I have tilted the front standard backwards before (and the rear forwards) but only to get most of the image out of focus.

Tim

[timbo10ca]

I can't answer your question, but I had to say that I like your drawing :D

Tim

[Jon Shiu]

Hi, to get the foreground in focus you will need to use forward tilt on the front, or back tilt on the rear standard. It might be easier if you think of it this way: take your diagram and rotate the whole picture until the red line is horizontal. It then becomes just like a normal landscape. It's easier to just try it out to understand, rather than theory.

Jon

[Walter Calahan]

I'm just impressed anyone can spell 'Scheimpflug'.

I had to cut and paste. I second the nice drawing.

[Brian Ellis]

I would think that tilting the lens back would work but I'm not a Scheimpflug expert. In a difficult tilt situation I don't really give much thought to Scheimpflug as such, I just tilt until the camera starts working the opposite of the way it should (i.e. to the point where moving the front standard forward starts to bring the far into better focus and moving it backwards starts to bring the near into focus). At that point you know you've tilted too much and you can just back off a little to get close to the optimum tilt. Scheimpflug isn't all that easy to use in the field if you use it by trying to envision the meeting point for the three planes because that point is often several feet under ground (or somewhere up in the air above the camera in your case).

[John Hennessy]

can I use front tilt, except tilting the front standard backwards?

Yes absolutely! That seems to be solultion here.

[Sheldon N]

I think Jon Shiu has it right. Don't think of each landscape element being higher than the other in terms of pure elevation... think of it being higher relative to the optical center or line of sight of the lens. Your camera is actually pointed down, since the horizon is at the top of the frame.

If the foreground was truly higher in the frame than the distant rock, you wouldn't be able to see the distant rock. I've modified your diagram to illustrate.

I think a strong front forward tilt, or a strong back rearward tilt (tilting the back towards vertical) would have worked here. Probably the back tilt, to correct perspective and to preserve coverage.

[Daniel Geiger]

I second Jon & Sheldon's point. It does not matter what is elevation wise higher and lower, but that the image portion on the lower portion of the image (upper on ground glass) is closer and the image elements in the higher part (lower on ground glass) are further away.

Most likely your problem was not enough forward tilt, or you may have run the lens out of coverage. As a *rough* approximation, the lens plane equally divides the angle between film plane and image plane. Let's assume you pointed your camera zero-zero at an angle of 20 degrees down. The image plane looks like at an angle of approx. 70 degrees: Difference = 50 degrees, put lens plane at 25 degrees to split the 50 deg. 25 deg front tilt is quite significant. You can get away with a bit less tilt and stopping down, so that depth of field will provide acceptable sharpness in the desired areas.

[Jeff Conrad]

Tilt won't rotate the plane of focus more than 90 degrees, so you'll need to point the camera down. For starters, imagine the camera axis more or less aligned with the red line in your diagram--it's then a matter of tilting the front forward (or the back backward, or combination thereof), much as you would do in a normal situation.

[Ole Tjugen]

Tilt won't rotate the plane of focus more than 90 degrees, ...

Yes it will. Try it!

This is a quite common situation for me, when I'm up on something high shooting down. I use front drop for quite a lot of my landscapes, simply because all the interesting bits are below me!

Use front foreward tilt if you want the foreground to be sharp too. Front backward tilt puts the near bit of the wedge up in the air over the camera, which may not be what you want.

[Jeff Conrad]

I use front drop for quite a lot of my landscapes, simply because all the interesting bits are below me!

Isn't this the same as pointing the camera down and tilting the back? I think we're describing the same thing but just expressing it differently. I could have been more clear; when one talks about rotating the plane of focus, I guess an obvious question is, "Relative to what?". I meant relative to the normal position perpendicular to the line of sight, and should have said so. It's certainly possible to have an angle greater than 90 degrees between the back and the PoF, as often happens when tilting the back rather than the lens.

[Ole Tjugen]

Agreed.

So even without tilting the back, it is possible to get an angle greater than 90 degrees between the back and the PoF. You may run into coverage problems, but that has nothing to do with this question.

Some front drop is also quite common in this kind of situation, since all the landscape is "down". That helps a lot on the coverage.

You end up with something like this (Scheimpflug'ed with front tilt & drop, back vertical):
http://www.bruraholo.no/images/Senja_2007/Scan-070713-0005.jpg

[steve simmons]

The only time you would tilt the front standard backwards would be to more closely align the plane of focus with a plane of the subject when the plane of the subject was above you - such as a ceiling.

steve simmons

[Joanna Carter]

Well, After 2 1/2 hours, I finally managed to get everythng in focus on this image. I ended up with the tiniest possible bit of back-tilt on the front standard. If you look at the tables that Merklinger gives, you will see that, theoretically it should be possible to get around 85° of acceptable focus; tilting backwards by that minute amount should just about get the foreground in whilst allowing right to the horizon, although the sky might not be as sharp as required. In my shot, the sky was not relevant as I also used front drop to include the manhole cover and lose a very bright hazy sky.

[Leonard Evens]

Let me try to mediate between Jeff and Ole and perhaps add some enlightenment to the general problem. This depends on the geometry, which I will try to describe in words, but understanding it will probably require drawing some diagrams which I will leave to you.

You have to keep in mind that there are two things which set the subject plane. Let me assume for the sake of argument that the back is vertical and we are tilting the front standard. (One can do a similar analysis for other orientations.) The tilt angle sets the position of the hinge line. If you tilt downward, the hinge line will be below the lens, which I believe is what would be desired in this case to keep the foreground in focus. But the actual position of the subject plane is then set by the position of the back relative to the lens, i.e., by focusing, which to simplify the argument, I will assume is done by moving the back. As you move the back closer to the lens, the Scheimpflug line moves up. When it is level with the hinge line, the subject plane is horizontal, but as you move the back closer to the lens, it moves above the hinge line, and the subject plane tilst downward..

Now consider another important plane, that parallel to the lens plane and at a distance equal to the focal length in back of it. This plane also intersects to film plane in a line parallel to the Scheimpflug line, and no point in the film plane below that line corresponds to any real subject point. So the frame in the film plane will have to lie entirely above that line. That means that as you focus closer in an attempt to increase the angle the subject plane makes with the horizontal, you will have to raise the frame, i.e. raise the back. Thus, how far you can raise the back, or equivalently drop the front (and perhaps compensate by raising the camera), will determine how steep the angle between subject plane and the horizontal can be.

Now in practice, you would use a near, far point method to determine the tilt and the position at which to focus in order to put the subject plane where you want it. So in effect the tilt angle and the focus position, which sets the Scheimplfug line, would rule. You would then have to raise the back so the part of the subject plane you want in the picture is in the frame. Whether you can do that or not in any given situation will be determined by the details of the scene and the capability of the camera.

One last comment. I've lost track of the original question was, but if the diagram is any indication, it does appear that one might not be able to get the depression in the middle ground in good focus as well as the immediate foreground and distance rise. That would depend on whether after stopping down sufficiently, the wedge shaped DOF could encompass it.

[Ole Tjugen]

I'm sorry, Leonard, but that didn't make me any wiser.

This one did, though: http://www.trenholm.org/hmmerk/#FVC

[Joanna Carter]

I find this page http://www.trenholm.org/hmmerk/HMArtls.html with the little movies to be very useful in visualising what is going on.

Oh, and is this image anything like what you are trying to achieve?

[walter23]

While I understand how the scheimpflug principle works, I usually use a little conceptual trick to do it in practice. Instead of caring about the 3D nature of the scene, I look at the first point on my ground glass and focus on it, then see what I have to do to focus on the second point, and then arrange the tilts so that the lensboard - ground-glass distance changes appropriately for the second point. Once I've got the first tilt estimate done, I'll check the two points again and see which way I have to adjust my tilt to get it right... and iterate from there until it's exactly set.

E.g., with the typical example where you need front tilt forward, on the upside down representation on the ground glass, the bottom of the glass contains the object which is further away, and so if I've focussed on the near object (top of the glass), I have to pull the front standard back (less extension) to focus on the further object (bottom of glass). In my short hand mental notation, this means the bottom of the ground glass has to be closer to the front standard, so front standard tilt foward moves the bottom of the front standard relatively closer to the ground glass; the part of the scene that requires less bellows extension gets it by virtue of the tilt (in other words, tilting the lens forward gives more extension to the part of the scene on the top of the ground glass where the closer objects reside, and so they get in focus). I don't know if this is a strictly correct description of why the scheimpflug principle works, but it's a practical method that works for me.

A diagram would probably help... but basically, the idea is that if an object on the left of the GG requires that I focus further out, I swing the front standard to that the left side of it is further from the GG.

If you work it out, you'll see that this shorthand is actually a perfectly valid way of looking at it and is applicable to all situations. It won't fail; it's just a camera-centric way of conceptualizing things instead of a scene-focussed method. It works in the case of a planar region that you want in focus just as well as with two objects. If you wanted a tabletop in focus, you'd just focus on the close side of it, figure out what movement you need for the far side, and iteratively adjust focus & tilt until you hit the magic angle.

[Leonard Evens]

I'm sorry, Leonard, but that didn't make me any wiser.

This one did, though: http://www.trenholm.org/hmmerk/#FVC

I'm familiar with that but it doesn't show the diagram I had in mind. The diagram of the hinge rule show there would illustrate my point if you add a line for the film plane and another line one focal length to the left of the lens plane and parallel to it.

Unfortunately, it just shows the situation where the subject plane is pointed upward. Somewhere Merklinger has an animation showing how the subject plane hinges on the hinge line as the rear standard is racked back and forth, but I didn't see it at this web site.

Perhaps if I have the energy, I will post my diagram.

[Hening Bettermann]

...Somewhere Merklinger has an animation showing how the subject plane hinges on the hinge line as the rear standard is racked back and forth, but I didn't see it at this web site.....

Is this the one?
http://www.trenholm.org/hmmerk/ViewCam.mov

It's on the page that Joanna linked to.

Good light! - Hening

[Leonard Evens]

I find this page http://www.trenholm.org/hmmerk/HMArtls.html with the little movies to be very useful in visualising what is going on.


Yes, that is the animation I had in mind. Now just add a line like the front focal plane line but at the same distance on the other side of the lens. Any point below where that intersects the film plane doesn't correspond to a real subject point that the camera can image. So the frame has to lie above it.

[timparkin]

From what I understand you use tilt to get the 'hinge point' at the right distance below the camera and then use tilt to change the angle of the plane of focus..

Your focus can now rotate that plane of focus beyond the 90 degree angle, just keep on racking the bellows out and eventually the plane of focus will drop lower and lower ad then past 90 degrees... theoretically, you can focus the plan all the way back to near vertical... but that would be silly