# Thread: Depth of focus with close subjects

1. ## Depth of focus with close subjects

In the currently hidden thread (Image quality: Normal holders vs. Quickload), someone claimed, "For close subjects such as portraits, the lense extension will increase depth of focus sufficiently to negate any slight mis alignment of the film plane."

This has some truth in it, but the question is how much: How much more depth of focus do you get when you focus close? Using Merklinger's figure 5 from his book Ins and Outs of Focus (which can be downloaded as a PDF from the web), it looks like there is a simple answer. The effective aperture is all you need to know. For example, when focusing for 1:1 macro with an 80mm lens at f/16, your extension is an extra 80mm and your effective aperture is f/32. Your depth of focus in this case is equal to what it would have been had you been using f/32 and focusing at infinity.

It appears to me that depth of focus is determined by your circle of confusion and your effective aperture, and nothing else. Your focal length does not matter, and how close you focus does not matter (except insomuch as it affects your effective aperture).

I am not sure whether this result is exact or an approximation, so any comments or analysis on this would be appreciated. If this result is only an approximation, I would appreciate hearing how far off it can be, and why.

2. ## Depth of focus with close subjects

Your/Merklinger's idea sounds good to me on a logical level. In practice, I usually just stop down a 1/3 of a stop or so more than I think I need, because all the logic in the world may go to hell when those negs come out of the developing tank.

3. ## Depth of focus with close subjects

Sorry, I made a mistake: I should have been ultra clear that my question really is about depth of focus (i.e., behind the lens) rather than depth of field.

4. ## Depth of focus with close subjects

First I thing I'd like to state that focus in itself lies in an exact plane while the rest, be it depth of field or depth of focus or cirlce of confusion, are nothing but approximations. We seem to place too much faith in all of these while having little control in how the final print is being viewed by others (unless you place your print in a glass box and instruct the viewers to touch their nose to the glass in order to see what you intended them to see (and even then, it will be off for some as not every eye resolves in the same way).

If I'm correctly connecting the alleged statement in the first line of your post to your conclusions later on:

1. in close-ups in order to focus on a subject the lens will be extended much beyond its average and it appears that's what that statement was relating to, not the focal length of the lens in use

2. you're correct that depth of focus depends on COC and effective aperature and is not related to a focal length

however, assuming I read Mr. Sidney Ray right (Applied Photographic Optics)

3. depth of focus changes with a rise in magnification ratio (for all practical purposes doField is inversely proportional to doFocus). I'll have to dig out the exact formula, but to my recollection there is one.

What that does however, is that with large depth of focus it becomes more difficult to focus exactly as there is too much of it (not easier as it's being suggested by some). I've notice this myself that in a close-up world, what I get in the end is often NOT what I thought I saw on the ground glass and have to redo it. It can somewhat be dealt with by first finding a near and far good focus (I'm relating to what I see on a ground glass) and then placing it in between.

5. ## Depth of focus with close subjects

OOps, I thing that I meant to start with an "I think". Why am I reading these things so early?

6. ## Depth of focus with close subjects

Jerry,

I urge you to download the rodenstock pre-designer software from here. (PC only)

with this software you can play with all the lens optics variables and instantly see the resulting variables including depth of focus and near and far points etc etc etc.
Its great for helping to understand what is happening without getting into the mathematics. The maths is there as well if you want it.

(use the depth of focus tab)

7. ## Depth of focus with close subjects

Your statement is correct. The depth of focus is

twice the product of the coc and the effective f-number.

So the depth of focus depends only on coc and effective f-number and is indepedent of focal length, provided the effective f-number is fixed. Note however that the effective f-number is the product of the actual f-number and the ratio of bellows extension to focal length. (That ratio is also qual to one plus the magnification.)

8. ## Depth of focus with close subjects

I neglected to say whether the formula involves any approximations. The formula arises by analyzing how far on either side of the exact image plane you can place the film plane without having circles of confusion exceed the maximum allowable circle of confusion. It is calculated by using simple geometry (i.e., similar triangles). But it does assume the lens is perfect and geometric optics applies. It also assumes no swings or tilts. Under those assumptions it is exact.

But, it ignores lens aberrations, field curvature, and of course diffraction. Moreover, you have to be careful in interpreting it. Consider what happens in a typical picture taking session even in the simplest case where the subject is confined to a plane perpendicular to the lens axis and no tilts or swings. You focus as best you can on the ground glass, but the position of the gg plane will differ somewhat from the correct exact image plane. In addition, when you put the film in, its postion may differ from both the ground glass plane and the exact image plane. These errors from the exact image plane could add or subtract. Even with the use of a loupe, the focusing error may be significant. For example, I found that my typical focusing error with a 7 X loupe could be as large as one quarter of a mm. It is important to realize that this type of error is inevitable because your eye when looking at the gg has a certain coc, and you can't focus better than that coc allows. The more powerful the loupe the smaller it is, but there is a limit to how strong a loupe can be used with any real viewing screen. Moreover, this focusing error is also magnified on closeup by the same factor. Unless your equipment is far out of adjustment, focusing error is likely to be more significant than misplacement of the film plane.

9. ## Depth of focus with close subjects

Further into the subject, as the focusing distance decreases, the depth of focus increases. For portraits this may well not be noticeable enough (although mathematically still the case). It becomes an important factor though for higher magnification ratios, as in close-up work and beyond, where many optical formulas, that have been reduced to a simplified version by equalising the focus image distance to focal length (true at infinity, less and less so as the subject gets closer to the lens), are no longer valid.

10. ## Depth of focus with close subjects

Thanks so much for each reply! And rob, thanks for the link to the cool software. I have a comment or two about the details that Leonard Evens contributed.

Of course, I should have referred to f-number rather than aperture. Leonard's language was much clearer. Leonard described a way of computing effective aperture that can be written

e = r(1+m),

where

e = effective f-number;
r = relative (marked) f-number;
m = magnification.

And my original example with m = 1 agrees with this, but we both forgot about the possibility of asymmetric lenses, and the formula can be off by more than a stop with asymmetric lenses.

For example, my Voigtlaender Telomar 36cm f/5.5 is a telephoto lens with P = 0.56 or so (by my measurement), where

P = pupillary magnification factor,

which the ratio of the exit pupil diameter (as seen from the rear) to the entrance pupil diameter (as seen from the front). My Xenotar 150mm f/2.8 has P = 0.92 or so. Most of my other large format lenses have P very close to 1, including my Super Symmar XL 80mm.

In what follows, my source is The Manual of Close-Up Photography, by Lester Lefkowitz.

The correct formula for general P is

e = r(1 + (m/P))

if the lens is front forward, but if the lens is reversed (which is wise when m > 1), the formula is

e = r(m + (1/P)).

Does this complication ever matter? Yes. For my Telomar at m = 1, I would be under exposing by almost a full stop if I used e = r(1+m) by mistake. For my Xenotar at m = 1, I could change a 10 second exposure to 11 seconds---or just ignore it. Of course, when P = 1, we get e = r(m + 1) whether the lens is reversed or not.

Lefkowitz goes on to say that the total depth of field is 2*c*e/(m*m), where c is the circle of confusion, and fortunately, this formula works for arbitrary P, provided we compute e properly. So my guess is that P has no effect on the depth of focus either.

To summarize my point, an asymmetric lens needs P to compute the effective aperture, but its depth of field---and probably depth of focus too---is unaffected by P except for how P affects the effective f-number. But if someone can disprove this, I would be delighted to see it.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•