Depth of field Determinants

_DVA2514_2010_albany-EditWhat determines depth of field? How can one control depth of field (DOF) while shooting in the field or studio?

Though there are many resources on the web listing and explaining the factors that control DOF, most people struggle to control depth of field the moment they run out of “Aperture options”.

What I’m trying to do in this article is to give the reader some practical tips as to how the various factors that determine the depth of field could be controlled in the field to achieve the desired DOF ‘look’.

During the next few paragraphs I will try to explain some of the concepts behind the depth of field. If you are not the geeky type, or not bothered, please feel free to jump straight to the practical tips.

Let’s start by reminding ourselves one of the most important facts: A photograph is made for humans, and whenever we talk about the visual impact of a photograph we talk about how an image is assessed by the human eye – not by machines or mathematical formulae that forget the physics of human vision.

Therefore everything related to depth of field starts with the resolution power of the human eye!
Resolution power can be defined as the minimum distance between two points that allow the eye or an optical instrument to clearly identify them as two distinct objects.
While this minimal distance is quite important, what really happen is two light rays coming (or get reflected) from those two points get identified as two light rays because they form the minimal required angle for the normal human eye to identify them as two different rays. This angle, which is about one arc minute (one sixtieth of a degree), is defined as the angular resolution of the human eye.

Two points kept 0.1mm (one tenth of a millimetre) apart at 25 cm away from a normal eye can form this critical angle of 1 arc second and hence can appear as two distinct points.
If the person looking at our two points takes his or her head back just a few centimetres the angle between the rays that pass through the pupil of the eye drops below the critical number and the two points start to appear as a single point.
In other words 0.1mm or smaller circle at a 25 cm distance appears to us as a dot while anything above 0.1mm at a 25cm distance will appear as a small disc/circle.
This is the reason why we see stars that are millions of kilometres wide as mere shiny dots in the sky.

What has this got to do with depth of field? Well, everything!…here’s the reason;

When you focus on your subject and take a picture, everything on either side of the plane of focus is going to be slightly out of focus and appear blurry; but if two rays coming from a slightly out of focus point on the final print (or the monitor) are going to create an angle of one arc minute or less on the eye, that point is still going to appear as a point even though it is not on the plane of focus.

Therefore achieving acceptable depth of field is all about making sure that, a point placed away from the plane of focus still appears as a point on the final print, (or on the monitor) at the intended viewing distance.

In other words this simply means, acceptable depth is all about how much an out of focus point object is magnified on the final presented image, and whether we still identify it as a point or something bigger than a point.

Here is one important point that seems to have confused many people who have written articles about depth of field on the web: Depth of field is always dependent on the size of the final image (i.e. magnification)

Say you set your camera lens and the distance to have a particular depth of field on an 8”x10” print.
What does this mean?
A point object that was not on the plane of focus is going to fall within the critical angle of one arc minute or less when the final 8”x10” print is viewed from the intended distance. Or the point object is going to be less than 0.1mm if the intended viewing distance was 25cm.
If somebody enlarges this same image to 16”x 20” our “point” is going to be twice as large and not going to appear as a point if viewed at the previous distance. For the larger image to have the same depth, the viewer has to move back to fulfil the critical angle of one arc minute or we have to come up with a different image of the same subject with a different camera, lens and distance setting to make our point appear as a point on the larger image.

DOF-Visible-Range-Tulips-dinil

Camera
NIKON D300
Focal Length
105mm
Aperture
f/8
Exposure
1/30s
ISO
400
Camera
NIKON D300
Focal Length
105mm
Aperture
f/8
Exposure
1/30s
ISO
400

 

So what determines depth of field?

 

1. Aperture

Everyone knows how to control this and therefore I’m not going to discuss this any further.

2. Subject Distance

This is probably the most underused depth of field determinant and therefore I will discuss how to use this to get the look and the feel one is after.

If the aperture, focal length and the camera (sensor size) remain the same the effective depth of field increases with distance. So, if you want to more depth of field you move back from your subject with your camera and lens or move in to have a narrower DOF! Easy isn’t it?
What if you want to place someone or something against a background that you want to make extremely blurry?

Get close to your subject as much as you can to minimize the depth of field and separate the background from the subject as far as you can. Most beginners place their subjects very close to backgrounds even when they try to create blurry backgrounds. Placing the background close to the subject maximizes the chances of that background staying within the acceptable depth of field, which is the thing to do to create a sharp background. To make it blurry, background has to be placed away from the subject while you move close to the subject to minimize the DOF.

3. Focal Length

This could be a can of worms!
Does Focal length determine the depth of field?
Yes, for all practical purposes focal length does affect the DOF.

A hypothesis that states the focal length does not change depth of field is getting popularized on the web at the moment of writing this article. A couple of highly reputable websites I personally visit have tried to prove this point by capturing images with different focal lengths, and tied to prove the assumption “if the image size remains the same, then at any given aperture all lenses will give the same depth of field”.

While I really respect those authors, they have made basic errors by changing the subject to camera distance in order to keep the subject size constant. They have conveniently disregarded the all the basic formulae in geometric optics and the fact that focal length does change the magnification.

Longer focal lengths, provided that the aperture, the subject distance and the sensor size are kept unchanged, magnify the resultant image to a greater extent compared to shorter focal lengths. Therefore they magnify “out of focus points” away from the plane of focus more than shorter focal lengths, making them much larger in the final image.

Here are few numbers for sceptics, using a very accurate DOF calculator.

Ex 1:
Camera/Senor: 36mm x 24mm
Assumed viewing distance 25cm
Focal Length 50mm
Aperture: f/8
Distance to the subject: 10m
DOF : 24.13m – Everything from 6m to 30.12m is going to be in acceptable focus

Ex 2:

Camera/Senor: 36mm x 24mm
Assumed viewing distance 25cm
Focal Length 100mm
Aperture: f/8
Distance to the subject: 10m
DOF : 3.42m – Everything from 8.58m to 11.99m is going to be in acceptable focus

Note only the Focal length was changed in the example 2.

So if you want a narrow depth of field increased the focal length, use a wider lens for more DOF.

 

Here’s an image shot at 200mm

200mm-DOF-Visible-Range

Same scene at 70mm, only the zoom ring was rotated while keeping the aperture, camera to subject distance and the camera unchanged. Cropped to give it the similar dimensions as the above image. (compare the sharpness of trees and the grass between the bench and the trees.

70mm-DOF-Visible-Range

Camera
NIKON D800
Focal Length
70mm
Aperture
f/5.6
Exposure
1/50s
ISO
100

This is bit of cheating – as pointed out by  WWT67 below!Enlargement of the second image contributes to the apparent reduction of DOF

Here are the same two images without cropping

_DVA1011jackedder_lake201270mm Uncropped

_DVA1012jackedder_lake2012200mm uncropped

4. Sensor size

Larger sensors have less DOF and smaller Sensor have more DOF for a given aperture, subject distance and focal length, simply because smaller sensors create a smaller blur circle or a Circle of Confusion, provided both final images are going to be blown up to have same dimensions.

So if you want to create a creamy background (beautiful bokeh) for your subject;
shoot wide open, get close, use a longer focal length and a larger sensor body. Do the exact opposite if you want a larger DOF.

Just remember that you have four independently controllable (provided you have a two cameras with different sensor sizes and more than a single prime lens) variables to achieve the DOF look you are after.

1. Aperture

2. Your Feet, to move forward or backwards to change the camera to subject distance

3. Different Focal Lengths

4. Sensor Size

Use all of them without solely depending on aperture to control the depth of field.

Comments

  1. WWT67 says

    One question. You quote a website of saying “If the image size remains the same, then at any given aperture all lenses will give the same depth of field”.

    In your example you use two different focal lengths and crop the 70mm photo to be the same field of view as the 200mm. Yes, the DOF will be a different.

    What if you shoot at 200mm and then change the zoom to 70mm and move closer to your subject to get the same field of view, without cropping? Would the DOF be equal?

    • says

      What if you shoot at 200mm and then change the zoom to 70mm and move closer to your subject to get the same field of view, without cropping? Would the DOF be equal?

      No it wouldn’t be, that was the incorrect assumption made by the very reputable website.

      Changing the focal length changes the magnification, and changing the distance changes the magnification again. Assuming that the change in magnification due to focal length change was going to get nullified by the magnification change due to change in distance, simply because field of view was kept the “same” was the basic error made in the assumption I have quoted.
      Try it yourself and see, or use any of the decent depth of field calculators available online to find out the answer easy way. Let me know if you really want to get in to the mathematics of it, I will send the titles of few books I really like.

Trackbacks

  1. [...] One concern people have expressed when considering a micro four thirds cameras is about their smaller sensor size and most of the relatively slower lenses creating a comparatively larger depth of fields and less than pleasing out of focus points (bokeh). While I agree most of those concerns are valid, I think people keep on forgetting the other methods available to control depth of field. For those who are interested here’s one of my lengthy posts from the past about depth of field determinants. [...]

Leave a Reply

Your email address will not be published. Required fields are marked *


+ six = 11

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>