Finding the Focus Plane
When there is not a focal plane that is clear to identify, how do others go about estimating it? The composition that I am referring to is leaves from a yellow-poplar sapling where no leaf is existing on the same plane, even considering the entire width of one leaf, there simply is no plane to focus on. Now the task is to estimate the best focal plane. I estimated the Schiemflug principle from outside the camera and used just the slightest front tilt, the camera was pointed downward slightly, lens to subject distance was about 3 feet with a 120mm lens on the 4x5. I then used the rear standard to bring a section of one of the leaves in focus that was not the closest leaf to the lens. I wanted achieve focus while being able to use a somewhat fast-ish shutter speed, but it was impossible (for me, as this is where I struggle a lot with LF). I wound up stopping down a good bit and having to use speeds from 1-3 seconds. Does my description here reveal a glaring mistake to anyone, or what? I'll develop these sheets tonight, if it is successful, it will be due to simple stopping down. I wonder in this situation, would it be better to just leave the lens plane un-tilted and just focus the way I did and stop down, without messing with Schiemflug---this being equivalent to pointing my 35mm down at the leaves and doing the same thing.
So how do you estimate the focus plane when there is no plane to focus on---I have a feeling the answer is just practice, practice, practice, like most other things.
Yes practice. But Sinar LF cameras can help you calculate optimal plane of focus and calculate the correct f stop for desired depth of field.
I went online and found a depth of field calculator that you could plug in the lens size, camera format, and distance to come up with a near and far focus for all my lenses. I then wrote it down in a notebook I keep with the camera outfit. I did calculations for 3', 5', and 10' with various f stops. A good friend of mine uses a string with knots in it to set up the distances. It helps a lot to know if you have 6" or 6' of depth of field for any combination you use. Another trick, once you know the near and far, is to focus on the nearest point you want in sharp focus, mark the rail of the camera, focus on the furtherest point, mark the rail, set the front lens plane to the middle somewhere, and then stop the lens down to achieve the greatest DOF. There is a great article in the newest issue of View Camera Magazine. That's my quick down and dirty method, hope it helps.
I have colored gaffer's tape all over my gear usually for cine work, I will most likely end up using it for marking the rails on the 4x5 too. Someone ought to come up with an iPhone app that works like that online DOF calculator.
In trying to understand the limits of DOF vs Hyperfocal distance vs diffraction and center of confusion, I am wondering what the limits are in using the Scheimpflug theory in landscape photography where the closest object in my frame is typically about 5-7 feet with a 135mm and no closer than 4 feet with a 90mm or wider.
For example, The attached image is my very first day using 4x5. I shot the scene as a lens test of my 135mm. I first tagged the peaks in focus, then tilted until I saw the rocks in the bottom 1/4 of the frame hit focus. I then re-focused to get the peaks back in and did these steps until I saw both zones come into focus. I then stopped down to F/22 and shot the exposure.
So the lake shore in the background is soft, not ideal...
Is there another way to go about this or am I simply pushing the limits of tilt and need to stop down more, which I generally try to avoid if it gets within a stop or two of minimum aperture due to diffraction.
"I'm the freak that shoots film. God bless the freaks!" ~ Mainecoonmaniac ~
I was more interested in finding a plane of focus when a an obvious plane is not present---for example, if I am focusing on the ground, the ground is the obvious plane, achieving sharp focus at all four corners of the ground glass with front tilt is simple at wide open aperture. But in my example, there is no real plane by which to aid in focusing and I was just wondering how others approach that situation---I guess it's mostly lots of aperture.
Sponsored Ad. (Subscribers to APUG have the option to remove this ad.)
Focus on the near object, then focus on the far object. Set the focus geometrically between the two points. On a view camera it would be 1/2 the distance on whichever standard you were using for the focus. The aperture needed can be estimated based on your focus spread (distance between the near and far focus point on your focusing rail in millimeters):
So, I'm guessing (and that's all I can do with this of course) that your tilt was still off when you moved to setting aperture. My thinking is that you got the bottom of your tilted wedge of focus (top and bottom of the focus region when tilted won't be parallel anymore, I don't believe) covering the mountains, but not low enough to cover the far shore.
Originally Posted by PKM-25
I'm no master of this either, and recently made a shot similar to yours, where I missed getting the foot of the distant mountain in sharp focus. For me, this wan't a big problem in the image, because that was a part of the scene I didn't want the eye stopping on anyway, there were power lines... so it wasn't the plan, but it worked out.
In your case, if you had continued slight tilts you might have gotten the bottom (far shore) in focus, and then added a smaller amount of aperture to raise the wedge to include the top of the mountain, but I can't really tell you.
Regarding the focus process itself, I usually start with the near field focus, then start tilting and tweaking from there to find a tilt and front standard distance (focus knob) that captures all the image objects I want in focus. I don't think I get ic-racer's comment above - he seems to be advising without regard to tilt. Of course you could just stop down to f/64, but then you'd be in diffraction land, which you are obviously trying to avoid by using tilt.
In your image, given the nature of the tilted area of focus, your challenge would probably be the near field tall plants, more than the far shore, but it seems not impossible to accomplish. You do probably have to stop down a bit, but not to f/64, to get the wedge to cover what you need.
One thing I think you'll like about your new Chamonix, PKM, is the axial tilt on the front standard. At least for me, it is so much easier to wiggle around than my Wista's base tilt for trying to find the right angle. I keep one hand on my loupe, and one hand on the standard moving the tilt, watching the effect. With the Wista, it took two hands, and all the focus changed due to the base tilt moving the whole lens forward and back. Annoying. With axial tilt you see the edges of the focus area moving across the image. Very nice.
I saw a video on youtube about practical focusing using tilt. The instructor started by focusing on the near field, and noting the focus position. Then he moved the focus to capture the far field of focus. He checked the amount and direction of the move, and suggested that the tilt (or in his case swing) had to move in the same direction when focused on the near field, and that the distance moved suggested the tilt distance (grossly, meaning either a lot or a little). So he would then focus back on the near field, apply tilt in the indicated direction (which is certainly going to be forward for your type of scene), evaluate, and repeat the process. Once tilted, the aperture is going to control the "up" and "down" of the size of the focus wedge, not the "in" and "out" of depth of field.
One lesson I learned recently is to pay attention to the corners of the ground glass while tilting the front and adjusting aperture, to make sure you don't start vignetting. My Schneider 135mm has a lot less coverage than I thought when I started doing tilts.
Originally Posted by CPorter
I think in your case, you have to think in terms not of flat planes as defined by a set of surfaces (eg, leaf faces), but a plane defined by your near and far objects. You don't describe your scene enough for me to visualize it, but it doesn't sound like you are focusing on a whole tree. If you are trying to focus on a tree, then your plane is vertical, and you shouldn't need tilt.
But if you are talking about a bunch of leaves, perhaps along a branch, or scattered on a table, or cascading down a wall on a vine, your question should be something like, what is the near thing I want in focus, and what is the far thing? Imagine a line through both of those objects. Then, consider that line as expanded into a plane (in other words, that line is the central axis coursing along the plane - the line is parallel to the plane, not perpendicular to it). That thin plane should capture as many other objects in the scene as you want to be in focus. (You can imagine an infinite number of planes rotating around that line, you choose one.) Your controls are then to tilt/swing to get the focus plane to match that imagined plane, and aperture to add shape (thickness) to the plane (makes it not a 2-D plane but a wedge - the smaller the aperture, the thicker the wedge).
It's really sort of difficult to describe this in the abstract...
Last edited by chuck94022; 05-14-2012 at 11:28 PM. Click to view previous post history.
Reason: clarify relationship between aperture and focus wedge in penultimate paragraph.
Just my opinion, but I think this photo looks pretty good as it is. The far trees are a bit soft, but that just makes them more background, and puts more emphasis on the foreground. Since the foreground seems to be where the interesting dynamic part of the image is, that doesn't seem a drawback.
Originally Posted by PKM-25
"People get bumped off." -- Weegee
ic-racer uses the same method I do and would recommend for you.
In your complex plane-finding scenario I would suggest this approach: First, try to imagine a plane in your scene that minimizes the distance from the elements in the scene to a point on your imaginary plane. Then find a few things that actually are located on the plane to use as focusing points. Some need to be near, some far. Use camera movements to get these in focus wide-open.
Next, check your focus spread by focusing on the objects farthest from your imaginary plane (I have millimeter scales on the beds/rails of all my cameras to make this easier). Once you find the nearest and farthest focusing point, set your focus halfway between and use the appropriate aperture for the spread (ic-racer has given you the table in his post above).
Now, you only have to get good at imagining in 3D to place your plane of focus and finding the near and far focus points (which can be rather tricky if you've used lots of movements; e.g., the base of a mountain may end up being the "far point" if you have used tilts).
To check if you have your movements right, or have chosen the correct plane of focus, try a different plane and see if the focus spread increases or decreases. The plane that has the least focus spread is the one you want.
There is a learning curve to all this, so don't get frustrated; practice and be patient.