The View Camera Focus

8/15/2019 The View Camera Focus

1/4

In the previous installment (Depthof Field for View Cameras—Part I,Shutterbug, November 1993), we sawtwo ways in which tables could beused to outline depth of field for viewcameras. The limits of depth of fieldwere defined either in terms of anglesor as fractions of the lens-to-subjectdistance. In this final article on thetopic, we will look at two ways to es-timate depth of field without resort tospecial tables. It is indeed possible touse standard depth of field tables, asprepared for ordinary cameras. (Byordinary cameras I mean those havingthe lens axis fixed perpendicular tothe film plane.) It is also possible toapply to view cameras the object-fieldmethod of estimating out of focus ef-fects as described in Adjusting Depthof Field, Part III (Shutterbug, June1992).

I will also to explain, as promised,the circumstances under which fo-cusing farther away makes some nearobjects sharper.

Let’s start with how to use stan-dard depth of field tables. Figure 1 il-lustrates. It’s very simple, but my ex-perience is that one often has to think twice when applying it. A ray isdrawn from the lens to some point onthe plane of sharp focus near whichwe wish to know the depth of field.We determine the distance, R, fromthe lens to the plane of sharp focusalong the ray. We look up the nearand far limits of depth of field for thatdistance and for the aperture and fo-cal length of the lens we are using.The near and far limits so determinedare just that: the distances to the nearand far limits of depth of field alongthat ray. To avoid unnecessary com-plexity, Figure 1 shows only the nearlimit, so calculated. Notice here, how-ever, that the depth of field de-termined this way does not give youthe depth of field measured per-pendicular to the plane of sharp focus.The numbers one gets can be de-ceiving when the ray makes a small

angle with the plane of sharp focus.The depth of field can be numericallylarge and still be almost non-existentfor some purposes. I have found thatmy brain seems to assume that thedepth of field will be given in a direc-tion perpendicular to the plane of sharp focus. I have to keep remindingmyself that this is not so for ‘alongthe ray’ estimates. One also needs tobe careful not to use “infinity” as thedistance from lens to plane of sharpfocus, when the distance is really be-yond infinity.

This adaptation of standard depthof field tables is appealing becausewe need nothing new. The tables pro-vided with our lens, or available inbooks, already apply. The difficultiesare two-fold. First of all, the answerone gets applies for those few raysalong which the lens to plane of sharpfocus distance is R. For any otherrays we must measure the new valueof R and do the calculations (or usethe table) again. A more serious po-

Depth of Field For View Cameras—Part II by Harold M. Merklingeras published in Shutterbug, December 1993.

Figure 1: We can use standard depth of field tables to estimate depth of field along any ray that passes through the plane of sharp focus. If the lens-to-plane of sharp focus is R,we simply look up the depth of field

for that distance and for the appropriate f-number in a standard table. The table should tell us the distance, X, to the near limit of depth of field, as well as the distance to the far limit, which is not shown here. L is the depth of field on the lens side of the plane of sharp focus.

LENSFILM

FILMPLANE

PLANE OF SHARP FOCUS

L

NEAR LIMIT OFDEPTH OF FIELD

LENSPLANE

?

RAY THROUGH

QUESTION MARK

Figure 2: The simplified method described in Figure 1 cannot tell us directly whether or not an object

located at the question mark is in acceptable focus or not. We must establish the inner limit of depth of field using at least one ray which passes through the plane of sharp focus. As shown, the question mark would not be in acceptable focus. If it were moved far enough to the right,however, it would be. Yet a ray from the lens to the question mark would still not pass through the plane of sharp focus.

LENSFILM

FILMPLANE


RX

L

NEAR LIMIT OF

DEPTH OF FIELD

LENSPLANE


2/4

tential problem is that this methodcannot tell us directly whether an object will be in acceptable focus ornot, if the ray from the lens to thatobject does not pass through theplane of sharp focus. This is related

to the “beyond infinity” problem Imentioned. The question mark in Fig-ure 2 is an example of an object forwhich the lens-to-object ray does notpass through the plane of sharp focus.In order to establish whether or notthe “?” is in acceptable focus, wehave to draw the limits of depth of field. We could use two rays whichdo intersect the plane of sharp focus,and draw lines between the results.Or we can use one ray and ourknowledge that the depth of field iszero at the hinge line, as shown in

Figure 2. Notice that the “?” asshown in the figure is not in accept-able focus, but if it were moved suf-ficiently far to the right, in a directionparallel to the plane of sharp focus, itwould be.

It might be tempting to reasonthat the lens-to-plane of sharp focusdistance for a ray passing through thequestion mark is infinity. In fact, thetrue optical distance is beyond in-finity. If we were to use tables or cal-culations based upon infinity we

would not get the right answers. Wecan use the table values for infinityfocus only for those rays that are par-allel to the plane of sharp focus.

A question I often used to ask my-self is: “How is depth of field affected

by using lens tilt to tilt the plane of sharp focus?” As this simplified de-scription helps to explain, depth of field along any given ray is not af-fected by tilting the plane of focus.But one must be careful not to takethis as saying that depth of fieldmeasured perpendicular to the planeof sharp focus does not change. If wemeasure depth of field perpendicularto the plane of sharp focus, tilting theplane of sharp focus will always re-duce depth of field. Depth of fieldmeasured perpendicular to the plane

of sharp focus will become very smallwhenever the plane of sharp focuspasses close to the lens.

The second simplified method of evaluating depth of field utilizes the“object field” method. With thisscheme we calculate the resolution‘spot size’ in front of the camera. If an object is larger than that spot size,the object will be recorded in the im-age. If the object is smaller than thecalculated spot size, it will be missed:contrast will be too low for the object

to be seen distinctly in the image.This object field method was de-scribed for ordinary cameras in “Ad- justing Depth of Field—Part III”(Shutterbug, June 1992). Applyingthis method to tilted planes of focus,

we can again measure the lens-to-plane of sharp focus distance, R, asshown in Figure 3. Then, for a lens of (stopped down) aperture diameter, d,and an object at distance, X (meas-ured along the ray), an object must beof a size equal to or greater thanS=dL/R in order to be resolved. L issimply the difference between dis-tances X and R. Alternatively, in or-der to resolve objects of diameter S,they must lie within a distance L(measured along the ray) of the planeof sharp focus, where L=RS/d.

A somewhat more general resultis obtained for the object field methodif we measure all distances in a direc-tion perpendicular to the plane of sharp focus. The distance from lens toplane of sharp focus is denoted as D.And the depth of field, L, is simplyequal to DS/d (on either side of theplane of sharp focus) everywhere.This is the same formula we obtainedfor ordinary cameras. Figure 4 il-lustrates this scheme. This methodhas the advantages that no tables are

Figure 3: The ‘Object Field Method’ of estimating depth of field described in an earlier article (June 1992) can also be adapted for view cameras.Here we use that method to estimate the resolution possible at distance, X,along a ray which passes through the plane of sharp focus at distance R. S is the estimated size of the smallest object that will be registered at full contrast in the image. The diameter of the lens aperture is d. L is simply the difference between R and X.

LENSFILM

FILMPLANE


RX

L

POINT OF INTEREST

LENSPLANE

d

RAYL = R

S

d

or

S = dL

R

LENS

FILM

FILMPLANE


NEAR LIMIT OFDEPTH OF FIELD

LENSPLANE

D

L

PLANE THROUGH LENS,PARALLEL TO


d

L = DS

d

or

S = dL

D

Figure 4: The Object Field Method can be generalized by measuring all lens-to-object and lens-to-plane distances in a direction perpendicular to the plane of sharp focus. For a constant resolution spot size, S, the inner and outer limits of depth of field are parallel to the plane of sharp focus. Only the inner limit is shown here. The outer limit will be the same distance, L, on the lower side of the plane of sharp focus.


3/4

DIAPHRAGM

FILM

FILMPLANE

PLANE OF SHARP FOCUS#1

LENSPLANE

#1

5'0.25"

10'

'BEHINDTHE

CAMERA'

20'

25'

PLANE OF SHARP FOCUS#2

LENSPLANE

#2

S1 = 2d = 0.5"

S2 = 1.25 d = 0.3125"

?

OBJECT 'BEHINDTHE CAMERA'

LINE THROUGH LENS,PARALLEL TO


5'

necessary and that it valid every-where.

You may recall that in Part IV of “The Scheimpflug Principle” (Shut-terbug March 1993) I provided a for-mula for determining the distance D.Often it is easy to estimate D, but notalways. Sometimes it is actually easi-er to use that formula for D.

The object field description for

depth of field will help us understandhow it happens that a subject can beimaged more sharply by focusing far-ther away. For the ordinary camera,the depth of field distance, L, cannever be greater than D for objects onthe camera side of the plane of sharpfocus. This limitation is not nec-essarily true for view cameras.

Let’s suppose that an object lies adistance 10 feet from the plane of sharp focus (as for the “?” shown inFigure 5), and that the lens aperturediameter is one-quarter of an inch.

Using the formula S=dL/D, we findthat if D is equal to 5 feet, the resolu-tion spot size, S, is equal to 2d, twicethe diameter of the lens, or one-half inch. If we now focus farther away,so that D is now 20 feet, the distance,

L, becomes 25 feet. And calculatingd, we find the resolution spot size is25/20 times the lens diameter, or0.3125 inches. Focusing farther awayhas improved our resolution, eventhough the lens-to-object distance hasstayed exactly the same. Under whatcircumstances does this phenomenonoccur? The answer is whenever L, onthe lens side of the plane of sharp fo-

cus, is greater than D. For ordinarycameras, this situation would put theobject behind the camera where itcould not be seen. With the viewcamera, the situation can arise easily.The effect is most pronounced whenD is very small, that is, when one isusing significant amounts of lens tilt.

My tongue-in-cheek use of theterm “behind the camera” perhapsneeds a little explanation. Imagine aplane passing through the lens andoriented parallel to the plane of sharpfocus. Any object on the opposite

side of this plane from the plane of sharp focus is ‘behind the camera’.The definition is illustrated by theshaded area in Figure 5.

I had thought that this ‘behind thecamera’ effect would be a minor

one—a technical curiosity. But whenI was taking pictures of that cementplant (used for illustrations in Part Iof “Depth of Field for View Cam-eras” and Part IV of “The Scheimp-flug Principle”) I quickly learned thatthe effect is very real and that it canmake a pictorial difference. The topsof the trees in the foreground got sig-nificantly sharper when I reduced the

lens tilt from 12° to 6°, pushing theplane of sharp focus away from thelens by a factor of about 2.

Having an object lie ‘behind thecamera’ is not a sufficient conditionto ensure that it will be “acceptablywell rendered” in the traditional depthof field sense. By a strange quirk of the mathematics, a second necessary(but still not sufficient) condition isthat the object must also lie fartherfrom the camera, measured per-pendicular to the film plane, than onehyperfocal distance! Thus, that tree I

used as an example in the previous ar-ticle, had to be more than one hyper-focal distance from the camera toeven stand a chance of falling withinthe conventional depth of field. Ac-ceptably sharp or not, all objects lo-

Figure 5: Here we illustrate how increasing the lens-to-plane of sharp focus distance from 5 feet to 20 feet improves resolution by a significant degree for an object 5 feet ‘behind the camera’.The resolution possible at the “?” is S1 (0.5")when the plane of sharp focus is 5 feet from the lens and S2 (0.3125") when the plane of sharp focus is at 20 feet. A bonus is that with less lens tilt being necessary, the “?” should lie more comfortably within the coverage circle of the lens.


4/4

cated ‘behind the camera’ will bemore sharply rendered by focusingfarther away.

The traditional depth of fieldmethod and the object field methodare complementary to one another. Ioften find that I use both. First I usethe traditional methods to obtain arough estimate of depth of field, thenwork out whether critical elements of

the image will be resolved or not us-ing the object field method.

Before concluding, I should prob-ably comment on the depth of fieldcalculators one finds on some viewcameras. Usually the calculator con-sists of a scale (marked in f-numbers)that rotates with the knob that movesthe camera back. One typically fo-cuses on the most distant point to beacceptably sharp, sets the scale tozero, and then focuses on the closestobject. The scale then indicates therecommended f-number. One con-

tinues by turning the knob back toone-half the recommended f-number,thereby putting the film half-way be-tween the two extreme focus posi-tions. These calculators will give anindication of the conventional depthof field so long as the focusing mo-tion moves the film in a direction per-

pendicular to the film plane. If thefilm plane and direction of focusingmotion are less than 25° or so off per-pendicular, the results will be ac-curate enough. If the angle should be-come 45° or more, the resultsrecommended by the calculatorshould be considered suspect. This allassumes, of course, that you knowwhat criteria the manufacturer used

for the acceptable diameter of the cir-cle of confusion, and that you agreewith it. One of my biggest ‘beefs’concerning depth of field tables andscales is that manufactures do not al-ways tell us what criteria were used todetermine the scales or tables. Tomake matters worse, manufacturersare not always consistent from lens tolens. I have been told by collectorsthat different depth of field scales aresometimes found on otherwise iden-tical lenses.

I hope you have found these ar-

ticles useful in gaining some insightinto the behavior of depth of field andfocus for view cameras. Informationon the topic is scarce, and it is im-portant to understand these things inorder to make best use of tilts andswings. I have not attempted to ex-plain everything; emphasis has been

on those topics not described else-where. As I observed earlier, the mathand the calculations can get a bit tedi-ous, but the underlying principles areeasy to understand. Even a basicknowledge of the principles shouldhelp you improve your photographicimages. Since I made up the tables forlens tilt and depth of field, I havefound that I use them extensively—

more than I would have guessed be-forehand. For me, an occasional viewcamera user, they shorten the set-uptime appreciably.

Someday, tables will be a thing of the past. The view camera will senseall its settings and send the results toa computer. The computer will drawfor the photographer a ‘map’ of situa-tion, and give precise numbers as andwhere requested. The photographerwill then have a choice: change thesettings on the camera and watchwhat happens to the map, or change

the map and listen as little micro-motors in the camera adjust its move-ments to provide the suggestedchanges.

© Harold M. Merklinger 1992.

Documents

The View Camera Focus