Harold merklinger

[tony hayes]

I know this thread has been done before where the question was asked how is the distance "J" or the height of the lens above the plane of sharp focus measured in practical terms after reading Merklinger's article on focusing the view camera. Following the thread didn't seem to provide the answer to me. What am I missing?
Tony Hayes

[Jeff Conrad]

Tony, I'm not quite sure what you're asking, but a couple of thoughts:


The distance J is always measured parallel to the film plane; for simplicity, let's assume the film plane is vertical.

Unless the subject is absolutely planar (e.g., your mother-in-law's living room rug), you normally want the "hinge line" below the ground plane, so there isn't a way to directly measure the distance.

I usually set tilt using Howard Bond's Focus/Check method, described and linked in QT's article How to focus the view camera (http://www.largeformatphotography.info/how-to-focus.html) on the LF home page. Merklinger's formula can still be useful as a sanity check: assuming the camera back is vertical, measure the distance from the ground to the lens axis, and use that distance to compute the maximum tilt (you seldom want the "hinge line" above the ground plane); the optimal tilt will usually be less.

[Joanna Carter]

Unless the subject is absolutely planar (e.g., your mother-in-law's living room rug), you normally want the "hinge line" below the ground plane, so there isn't a way to directly measure the distance.


I often use a tilt angle of less than 1°, which can place the hinge point way below the ground :p

[ic-racer]

If you know the tilt angle, use Table 4,page 96


Table IV tells us the distance, J, that results for a specified tilt angle. Table IVa gives us that distance in meters; Table IVb gives us the distance in feet.

[tony hayes]

Thanks to all for responses. I find Merklinger's articles amazing. It seems to me that he presents a simple formula for determining what your camera tilt should be given that you already know what value `J' is. I just couldn't find anywhere where he describes how to determine J if you don't already know what your tilt angle should be. I presume (correctly)? from Jeff Conrad's answer that the distance from lens axis to ground should be the minimum distance for J and then use this as a starting point to make up whatever distance you want J to be from here in order to place it under ground (presumably to get near plane in focus of course). Is this a correct reading of the situation?
Tony Hayes

[Joanna Carter]

Tony, as interesting as Merklinger's articles are, whatever you do, don't try to use them to take pictures.

The beauty of an LF camera is that you get what you see on the screen. Forget the formulæ and start playing with tilt and focus to see what effect they have on the image on the screen.

I teach LF workshops and can assure you that nobody who comes along wants to have to read, let alone learn, formulæ. We always start by showing someone what you get on the screen and what happens when we adjust the movements.

[Jeff Conrad]

Tony, we're all indebted to Merklinger for rediscovering the principles of adjusting tilt and focus and describing them. And the formula relating f, J, and tilt can be useful as a sanity check (you almost never want the "hinge line" above the ground plane; the worst mistake many people make is using too much tilt). But I think after that he gets a bit carried away with formulas for which I've never found a practical application. I would, however, recommend reading his description of the "Craig Bailey" problem, where he shows that when the subject has significant height, less is often more when it comes to optimal tilt.

I've looked extensively at the mathematics involved, and Leonard Evens has gone much further. But I don't think either of us uses formulas in the field. I've make extensive use of analytical methods in constructing hypothetical diagrams, and I think I've discovered a few helpful general principles. But in the field, I don't have a diagram or a handy CAD program to calculate distances and angles to feed a computer that would calculate the tilt. And I've never had (nor could I ever have afforded) a Sinar e ... I don't even use Merklinger's formula as a starting point, because I usually want the "hinge line" well below the ground plane, and placing it where I want it is usually much easier using visual methods. Again, I strongly recommend QT's article, which I think is one of the best basic references. He describes a simple iterative procedure for finding the optimal tilt; in practice, you usually get to where you want to be in 2-3 iterations.

I usually take a slightly different approach, setting the tilt by adjusting the PoF to pass through two points on the near angular limit of DoF, then adjusting focus to the midpoint of the focus spread, and determining the f-number from the focus spread (just like without tilt). For most of the pictures I take, I seem to get to where I want to be directly. But it doesn't always work quite so simply. And as always, YMMV.

I'll finish by restating a point I think both Joanna and I have stressed: a little tilt usually goes a long way.

[Joanna Carter]

I'll finish by restating a point I think both Joanna and I have stressed: a little tilt usually goes a long way.
Indeed; here is an image that has everything in focus, from the riverbed, which is about 6metres lower than the quay on which the camera was standing, to the prow of the boat, which was only about 4 metres in front of the camera, through the entire scene to the horizon.

http://grandes-images.com/fr/Paysages/Pages/France_2008_files/Media/ToulAnHeryBaliseDeChenal/ToulAnHeryBaliseDeChenal.jpg

I can only guess that the hinge point must have been around the same level as the riverbed, but I do know that the tilt required was « le basculement au poil de chien » (the thickness of a dog hair); in other words, immeasurably small.

There are some photos that can be set up "mechanically" but, in cases like this one, there is nothing to substitute for familiarity with the effects of tilt and focus, on the image on the screen.

Had it not been for Merklinger's tables, I would not have discovered the "micro-tilt" technique but; do I use the tables in the field? No, I learned the principles behind the tables, to the point that I am now able to set up and take such a shot in a matter of a few minutes.

Tony, now you've looked at the tables, put them away and get out there; experiment and play until you see how the movements work ;)

[cowanw]

Fig 41 of Deardorff's Corrective Photography shows a Architectural shot of a aircraft engine plant in which the tilt is upward to bring the ceiling in focus along it's entire length.
I expect the hinge line will be above the ground plane there.

[tony hayes]

To Jeff, Joanna, ice-racer and cowanw
I thank all greatly for your input. It's all quite fascinating to me and I love the talk back on this forum. Your practical advice its often what's missing when I get seduced by mathematics and science. It's funny when you apply a tilt thinking this is what you need and you look on the ground glass and think - that didn't work. I think Joanna I might `tilteth` too much and I've also been thrown by not realizing I often must re focus on the original point after tilting. This feed back has been great fun and much appreciated, again thanks to all
Tony

[Jeff Conrad]

Tony, one other thing to keep in mind--nothing says that you must use tilt. Unless the region you want sharp is wedge shaped, tilt won't help, and can often make things worse. And even if the region is wedge shaped, if the angle at the wedge apex is wide, you may need a larger f-number with tilt than without. Joanna's image is a good example of something right on the edge; it would have been interesting to see the results with and without tilt. How do you find out? Try it both ways, as described in QT's article. Again, this is discussed by Merklinger as part of the "Craig Bailey" problem.

[BradleyGibson]

Sorry for resurrecting such an old thread, but the question is really valid, and I was puzzled by the same question for a time.

Merklinger is making the assumption that you know, as the photographer where you *want* to place the plane of sharpest focus. Along the ground? (Then you can measure your len's height, parallel to the imaging plane to find J). Through a medallion 5m in front of the camera, and rising to meet a wall 30m beyond that at a height of 2.5m? (Below ground, but calculable, so you will again know J).

Then, he is suggesting use of his tables, knowing J, to determine proper tilt angle, so that focusing is not an iterative affair.

Whether or not you *want* to do these measurements and or calculations instead of the traditional iterative focusing approach is, of course, a matter of personal taste, but this post is intended to clarify how Merklinger's J figures were arriving "out of thin air".

It's subtle but had me puzzled for a bit as well. And I hadn't seen this question answered anywhere, so I thought I'd put it out there.

-Brad

[Emmanuel BIGLER]

Tim Parkin on the UK LF forum has offered us a nice on-line simulator of DOF with tilted planes.

http://static.timparkin.co.uk/static/focus/index.psp?lenstilt=6.1&a=98.5&focallength=90&N=32&coc=0.1

[Cor]

Nice simulator, didn't work to good on Firefox though, had to switch to IE8, and than it isn't flawless either, but I should not grumble, it's appreciated that Tim takes the time and effort to make such a thing!

Best,

Cor

Tim Parkin on the UK LF forum has offered us a nice on-line simulator of DOF with tilted planes.

http://static.timparkin.co.uk/static/focus/index.psp?lenstilt=6.1&a=98.5&focallength=90&N=32&coc=0.1

[Jeff Conrad]

This looks to be a pretty useful tool—Tim Parkin seems to pick up where Merklinger leaves off in his discussion of the “Craig Bailey” problem in Focusing the View Camera.

But it does seem to have a few glitches; in particular, the indicated angle of view (the red lines) seems to be twice the actual value. I’ve sent a few comments to the author.

Incidentally, it works as well for me using Firefox 20 as it does using IE 8 or Chrome 26.