DOF, f-values & sharpness

[Anonomatos]

I've seen some massive, super sharp prints with a very narrow DOF (I might compare it to f2.8 when working with a 85mm 1.8D Nikkor lens on a Nikon D200 digital camera and the subject is about 2 to 3 meters away). He's a student and his equipment was under $1000,- he claimed, dunno what equipment though, might have been a Sinar. The effects and pastel-like unsharpness he achieved was far beyond anything I've ever seen (even compared to high end digital, like Nikon D3X digital camera's or Canon 5D mkII's!).

I'm reading for hours now in the internet (including this forum and tons of articles by Ken Rockwell) to figure out ways to work with DOF.

As far as I can figure out these are all variables:
-Lenses (duhh) & f-stop set
-Bellow extensions
-Camera movements

In theory I know how to calculate or review the 'plane of sharpness' (or whatever it's called :)) and I know I should lean the lens a little bit forward to get a bit of extra sharpness. But what confuses me is that most of the lenses I find (like on e-bay) have a f-stop of 5.6 as the largest opening. I know there are some big differences between lensen on digital and LF, but can't seem to figure it all out.. :confused:

I guess my question would be: are there other ways of working with DOF, except from changing the f-stop? And... how? :cool:

Hope the silly questions will get less silly soon!

[DJGainer]

The actual range that is sharp will not change, but you can use the camera movements to bring different aspects of the image into that range. Have you consulted the homepage of this forum? Lots of good articles there.

[Diane Maher]

Regarding the lenses, f/5.6 is a fast lens for LF. In fact, many common lenses have a max. aperture of f/9. You won't find lenses with wider apertures unless they are big, brass, probably heavy and 100 or so years old (okay, maybe this last is overdoing it a bit).

[Anonomatos]

The actual range that is sharp will not change, but you can use the camera movements to bring different aspects of the image into that range. Have you consulted the homepage of this forum? Lots of good articles there.

It was actually my first step towards reading the articles, so yes I did. I found a very good article about sharpness and about DOF (though a bit theoretical with diagrams I couldn't make much out of). But I couldn't find an article about really controlling the DOF and make it as small as possible.

Did I miss one of the articles maybe? :o

Thanks :)

[DJGainer]

My fault! I misunderstood you. You're not likely to find a larger apeture on modern lenses. And most interested look to increase the area of the image in focus, so you seem to be attempting the opposite. Are you sure the student was using LF? If so, it's likely he had the bellows racked out and was focusing closer in; this would surely decrease the amount of the image in focus.

[Anonomatos]

Aha, bellow out & focus closer. Those are the tips I'm looking for.
The guy was definately using LF, I might get a hold of his e-mail but I'm afraid it could get difficult.

And another question, might be a bit on the wrong bit of forum though:
Afaik in digital telephoto lenses tends to get smaller DOF. Does this also counts for LF? The "Introduction to large format lenses" doesn't say much about it.

[Robert Budding]

You might find that this DOF calculator useful:

http://dofmaster.com/doftable.html

Keep in mind that 210mm lens on 4x5 is a short telephoto and normal lenses are 135-150mm.

[r_a_feldman]

<snip>
And another question, might be a bit on the wrong bit of forum though:
Afaik in digital telephoto lenses tends to get smaller DOF. Does this also counts for LF? The "Introduction to large format lenses" doesn't say much about it.

To greatly oversimplify, telephoto lenses really don't have a smaller depth of field. The depth of field is related to the size of the image on the film, and the larger the image, the smaller the DOF. If you move in with a telephoto lens so that an object is 1:1 on the film, it will have the same DOF (for the same f-stop) as a wide angle lens when you have moved in so that the image is also 1:1.

[Brian Ellis]

Only three things affect depth of field: aperture, focal length, and camera-to-subject distance. Tilt and swing affect the plane of focus. Shift, rise and fall have no effect on depth of field. Ansel Adams' book "The Camera" has an excellent discussion of depth of field. If you read it you'll know all you need to know about depth of field.

[Leonard Evens]

To reiterate the points made above.

1) For distant subjects, with the subject at a fixed specified distance from the lens, the DOF is determined by the focal length, the f-stop, and the degree of enlargement of the image in the camera needed to produce the final print or other image. The wider the aperture, i.e., the smaller the f-number, the smaller the DOF. The longer the focal length, the smaller the DOF. The greater the degree of enlargement, the smaller the DOF. If you change more than one of these simultaneously, the answer can be a bit complicated, and you have to look at formulas (or a DOf calculator) to find out what happens.

2) For relatively close subjects, it is often more appropriate to look at the scale of reproduction or magnification (taio of subject size to image sie in the film) than the distance of the subject to the lens. If you keep the scale of reproduction fixed, the DOF is essentially independent of the focal length.

3) Tilt doesn't limit DOF significantly, it still depends on the f-stop as above. But it does change the exact plane in focus, with the DOf extending on either side of that plane. So, you may select by tilting some narrow section of a subject which lies in a plane parallel to the film plane. There will still be a lot in focus in front of and in back of that plane, but it may not contain anything visible in the photograph. The effect will be that you see only a narrow vertical slice of the subject in focus and nothing else.

[rdenney]

An example may help illustrate these points. Let's choose two extremes--the 24x36 DSLR you are used to and, say, an 8x10 view camera. And let's say we are going to make a full-face portrait. And, to maintain simplicity, we'll limit ourselves to a "normal" lens with a focal length approximating the diameter of the image frame.

Let's start with the 8x10. If I make a full-face portrait on 8x10, that face may be, say, five inches tall on the film. The actual face may be 10 inches tall in real life, so my image on the film is a 2:1 magnification. With a 12-inch lens, that means the lens will be about 24 inches from the subject's face.

That photo is nearly in the macro range, and we all know that at macro magnifications, the depth of field is really narrow. According to DOFmaster, at f/9, the depth of field will extend from 1.99 feet to 2.01 feet.

Now, I could put a 300mm lens on my Canon 5D, and the magnification would be exactly the same. So would the depth of field. But here's the problems with doing that: 1.) I'd need a bunch of extension tubes to get that 300mm lens to focus, and 2.) I'd only be able to include the tip of the nose in the picture. I could back up the camera to include the whole face, but then it wouldn't be the same picture. So, I have to use a shorter lens.

With a 50mm (2-inch) lens, I can make pretty much the same picture. With the camera in the same position, the 2-inch lens will make an image on my 24x36 frame about 0.8 inches tall. That's a magnification of 1:24--not at all close to the macro range. To get similar depth of field as what I was able to get at 1:2 with the 8x10 camera and its 300mm f/9 lens, I need something a hair faster than f/1.0, which is currently not in Canon's lineup.

And if I used a 300/5.6 lens on my 8x10 camera, there is no lens made for 35mm that is fast enough to provide a similar selective focus effect.

Now that we have established incredibly narrow depth of field using the large format, we have another advantage for the stuff that IS in focus. A given print will be enlarged only 1/8 as much. So, if an 8x10 print is my goal, the 24x36 image will require an 8x enlargement, but I can make a contact print from the 8x10 negative.

Thus, a full-face portrait with 8x10 will have considerably less depth of field and therefore much greater selective focus (if that's what I want), but in the focus plane, it will provide far greater resolution and contrast on a given size print. The results border on the mystical. People try to simulate the effect using tilt lenses on their small cameras and the like, but it's just not the same.

I have an old Ilex 8-1/2" f/4.5 lens. It's a Tessar formula--nothing at all fancy by today's standards. It's slightly longer than normal on my 4x5 camera, which I like for portraits. Despite that it's an old hunk of junk, I keep it because of that f/4.5 aperture. At f/4.5, I can get an effect impossible with small format and very similar to my 8x10 camera with the f/9 lens. No, the lens isn't that sharp when used wide open, but I may only be enlarging it by a factor of 2, so it doesn't have to be.

To me, managing what is in focus and out of focus has a greater effect on the look of a photo than just about any other decision we make. The large-format cameras give us huge control over that decision, beyond what we can do with small format. Even when we want everything sharp (which, for me, is most of the time), getting it so with a small camera is often impossible because we'd need f/32 and, even assuming our lens goes down that far, we can't live with the fuzzy diffractive mess that results. But with careful camera tilts and swings, we can usually provide the sharpness we need where we need it, even considering that we have less depth of field. And it takes a mighty small aperture to ruin a large-format image with too much diffraction.

Rick "finding cameras without movements to be pretty limiting for landscape applications" Denney

[Anonomatos]

Thanks a BUNCH, Rick! I think it's very nice of you to explain all this. Might be all common to you guys but it's mostly Abakradabra for me. But I'm starting to understand it all bit by bit!

Actually, I just ordered book 1-3 from the Ansel Adams collection (The Camera, The Negative & The Print). Think that'll help me out a bit further :-)

[Ivan J. Eberle]

What the OP is perhaps referring to is taking a large, fast aperture WWII Aero Ektar and mating it to Speed Graphic for a very narrow depth of field portraiture effect with pleasing bokeh.

[Ben Syverson]

Anonomatos, I have one tiny bit of wisdom I would like to pass on: DO NOT get caught up in trying to "understand" or measure DOF. It is probably the #1 most contentious issue in photography, ahead of even "digital vs film." Most photographers have an intuitive understanding about how DOF works (and even agree!), but when they start to try to define and talk about it, it gets ugly, especially online.

All you need to know is what Rick wrote: 300mm f/9 in 8x10 will look roughly similar to 50mm f/1.0 in 135. They both will have extremely shallow DOF.

If you're interested in portraits, that's all you need to know. Stop there! If you want to do landscapes, you will likely read about airy discs, hyperfocal distances, viewing distances, enlargement percentages and "circle of confusion" (which is most aptly named photo term ever). In other words, this way lies madness. My advice if you want to do landscapes is: f/32, tilt until everything is pretty much in focus, plug your ears and yell "lalalalalalala" as you trip the shutter. :)

[rdenney]

My advice if you want to do landscapes is: f/32, tilt until everything is pretty much in focus, plug your ears and yell "lalalalalalala" as you trip the shutter. :)

And if that doesn't work, print smaller, because probably nothing would have worked.

Rick "who might go to f/45 in a pinch, because 'out-of-focus' is worse than 'diffraction'" Denney

[Anonomatos]

Yeah, it's a bit confusion all the tables.
I talked to a guy in a shop (I guess one of the few shops in NL which sells and buys LF equipment and deliver parts from stock) and he explained it a bit more. His conclusion was also not to get too much caught up, he said: "Just try shooting wide open and try more from there on." He's prolly right :)

[March]

With respect to the other very good posts on this topic, I've sometimes found it confusing to try and understand the way DOF is effected by camera distance to subject and lens selection. If you stand farther away or closer to a subject or if you select a different lens it is true that DOF field changes. The problem is that you are also changing what is framed in the image. If you are doing a headshot with a tele lens standing far back or if you are doing that same headshot with a wide standing close in, the DOF in the frame will remain the same. So in reality DOF is effected by camera to subject distance and lens selection, but in practice it isn't always the case ... Keeping this in mind the thing that is most important to me is selecting the appropriate lens for a given job - the rest seems to fall into place.

[Ivan J. Eberle]

rdenny's example above stated a 5" image on film of a 10" face is 2:1. Wouldn't that be 1:2 instead (half life size)? Nevertheless it's still going to be a shallow DOF, but not quite as razor thin.

[rdenney]

rdenny's example above stated a 5" image on film of a 10" face is 2:1. Wouldn't that be 1:2 instead (half life size)? Nevertheless it's still going to be a shallow DOF, but not quite as razor thin.

Yes, 1:2. You knew what I meant.

Rick "razor thin = one eye in focus, everything else not, unless the swing lines up the eyes" Denney

[rdenney]

With respect to the other very good posts on this topic, I've sometimes found it confusing to try and understand the way DOF is effected by camera distance to subject and lens selection. If you stand farther away or closer to a subject or if you select a different lens it is true that DOF field changes. The problem is that you are also changing what is framed in the image. If you are doing a headshot with a tele lens standing far back or if you are doing that same headshot with a wide standing close in, the DOF in the frame will remain the same. So in reality DOF is effected by camera to subject distance and lens selection, but in practice it isn't always the case ... Keeping this in mind the thing that is most important to me is selecting the appropriate lens for a given job - the rest seems to fall into place.

To make the same picture from the same camera location with a larger format requires a longer lens and therefore higher magnifications. That's what makes the depth of field less unless you stop down to a smaller aperture to compensate.

Rick "suggesting DOFmaster for making theoretical comparisons, but preferring a loupe and a ground glass for seeing the difference in real life" Denney

[Bill_1856]

If you're using an optical viewfinder camera DOF tables are handy. If, however, you're using a camera where you focus and compose directly on the ground glass forget the tables and just use your eyes (with a loupe if necessary) to determine if something is in focus or not. Don't make it more complicated that it needs to be.

[Ben Syverson]

Don't make it more complicated that it needs to be.
Hear, hear. Excellent advice.