Achieving a “filmic” depth of field was a holy grail among digital shooters for a long time—encouraging the invention of a plethora of lens adapters that generated the film DoF look—but now that larger sensors are more common and affordable, the adapters are, by and large, obsolete. Digital cinematographers can mimic the look and feel of 35mm motion picture depth of field with a native large-sensor camera.
Depth of field
All of this is pretty common knowledge these days. However, in the past couple of weeks I’ve run across several instances where confusion and misconceptions about depth of field are still quite prevalent—even among professionals. This month, I’ll make a concerted effort to clear up these misunderstandings and set the record straight on depth of field.
Defining the Field
The physics of optics dictate that only one plane of a photographic image is actually in focus. Light rays pass though the lens and are focused onto the target plane (the sensor), and only that single point in space is actually “in focus.” The rest of the image falls into varying states of unfocused imagery that, depending on certain variables, can be considered “acceptable” or “unacceptable” focus. This area of “acceptable focus,” in front of and behind the plane of focus, is called the depth of field.
Before I get too deep into the technical aspects of depth of field, it’s very important to understand that this is an extremely subjective subject. The charts and calculators (thanks to modern iApps, DoF calculators are abundant) are based on mathematical calculations and theoretical optical elements. All depth of field calculators will give you is an approximation of the area of acceptable focus for a given format. The focus doesn’t just drop off sharply beyond those numbers; it’s a very soft falloff, and the DoF numbers are merely approximate guides based on certain mathematical assumptions. A lesser-quality, lower-resolution lens (one with low sharpness and low contrast) may actually appear to have more depth of field because the image is naturally soft and the area where the “acceptable” sharpness falls to “unacceptable” is harder to determine. With a very sharp, high-contrast lens, the depth of field can appear to be much less because the differences between “acceptable” and “unacceptable” are much more clearly discerned.
Circle of confusion
To mathematically determine depth of field, you must first determine the specific lens’ hyperfocal distance. The hyperfocal is an optical anomaly wherein, at a specific focal distance, for any given aperture and focal length, the acceptable focus is half that distance to infinity. Every lens, at every aperture, has a specific hyperfocal distance, but most “normal” lenses’ hyperfocals are beyond a point of practical usage. Here are a couple examples, based on 35mm motion picture lenses, at an aperture of f/4:
25mm—H = 20′ 2″
35mm—H = 39′ 7″
40mm—H = 51′ 8″
50mm—H = 80′ 9″
75mm—H = 181′ 8″
As you can see, the longer the lens, the greater the hyperfocal setting. Most lenses don’t have focal settings for beyond 20′, so in these instances you’re already focused on infinity. The wider the lens, and the smaller the aperture, the closer the hyperfocal distance will be. This leads to some misconceptions about wider lenses having extraordinary depth of field, because with wider lenses it’s often more feasible to utilize the hyperfocal function. The depth of field properties—aside from the anomaly of hyperfocal—remain consistent among focal lengths, however, as I’ll explain in further detail a little later.
The hyperfocal for any given lens and aperture is calculated by taking the square of the lens’ focal length divided by the product of the f-stop number and the circle of confusion.
H = Hyperfocal, F = Focal length of lens, f = f-stop, CoC = Circle of Confusion
The circle of confusion is sometimes, well, confusing. In the most basic terms, if we imagine that a picture is made up of millions and millions of points, those points will be photographed as sharp points within the plane of focus. Beyond the plane of focus, they no longer appear as points but as spheres (out of focus points) that become larger and larger the further away from the plane of focus they are in the image.
For this example of depth of field, I placed three playing
cards about 8″ apart and shot with my Canon EOS 7D
and 28-135mm lens.
An example: if we imagine photographing a single point of light off in the distance, when that light is in clear focus, it will be projected onto the target as a point of light. If there is a second light, closer to the lens, outside of the acceptable depth of field, it will not be seen as a point of light but, rather, a diffuse (out of focus) circle of light. Likewise, a third light, further away than the acceptable depth of field range, will also be projected as a diffuse (out of focus) circle.
These are called the “circles of confusion” because objects out of focus are no longer points but, rather, take on the circular shape of the optics of the lens. There is a range of “acceptable” focus in which the points are starting to become spheres, but they’re still so small that the eye cannot tell the difference; they still appear to be sharply focused points. The acceptable focus for a particular format is determined by the size of the circle of confusion, which represents the size of an out-of-focus sphere that can still be considered to be in focus. This number is a constant for any given format/sensor size.
It is important to note, however, that this number is subjective, and the real-world definition of what is in focus or not depends on the quality of the lens and even the quality and size of the final display. On an iPhone screen, an image might appear to have considerably more depth of field than it would if projected onto a full cinema screen. The circle of confusion number is theoretical maximum, based on the format size. Even further confusing the matter, there are variations on the base standard circles of confusion used for each format, depending on the calculator you use.
Although the circle of confusion numbers are considerably different in some cases between the two calculators, the practical results are nearly identical. For truly critical applications, it is always advisable to test on the specific equipment you’ll be using before your actual shoot.
Notice that in the circle of confusion numbers, the larger the sensor, the larger the circle of confusion. Generally speaking, the larger the circle of confusion, the less depth of field for a given scenario.
Looking at the difference between a 1/3″ video sensor (.00043″ or .0002″) and 35mm film (.001″ or .00085″), you can see that the depth of field of professional motion picture film is two to four times less than that of the smaller prosumer video format.
Once we have our given circle of confusion numbers, we can calculate the hyperfocal distance for any lens and aperture (formula above). When you have that number, you can then calculate your depth of field. Formulas for depth of field are:
H = Hyperfocal, S = Subject distance, F = Focal length of the lens
Likewise, the formula for figuring out the far distance of depth of field is:
H = Hyperfocal, S = Subject distance, F = Focal length of the lens
Don’t be scared of the math. Although the formulas are useful to get a better understanding of what is happening, with modern calculators readily available, the math isn’t something you need to deal with on a daily basis.
Chemical Wedding’s Toland ASC Digital Assistant app is one of the most accurate—and complex—calculators I’ve ever seen, but it’s not for the casual user. For most applications, I actually prefer to use David Eubank’s pCAM application—but there are dozens of others out there.
The important factor is an understanding that the most significant aspects of depth of field are: format size (circle of confusion), aperture, subject distance and lens focal length.
Longer Lenses Have Less Depth of Field
This can be a confusing statement because there is truth in it, but when you compare apples to apples, it is incorrect.
Let me elaborate: An f-stop number (one of our significant contributors to DoF) is determined mathematically as the relationship between the diameter of the aperture opening and the focal length of the lens.
f = f-stop, F = Focal length, a = aperture (diameter of the opening)
So a 50mm lens that has a 25mm diameter aperture opening has an f-stop of f/2. (50 divided by 25 = 2.)
However, if you look at a 100mm lens, an f/2 aperture is actually a 50mm diameter opening—twice as large as the f/2 on the 50mm lens. So the f/2 on the 100mm lens has less depth of field than the f/2 on the 50mm lens, simply by virtue of the fact that it is twice as large a diameter opening. This is where the misnomer actually is true: longer lenses, for a given aperture, have less depth of field.
However (and this is a big however), when you’re comparing apples to apples, things change. Many people believe—incorrectly—that they can control their depth of field by simply switching lenses. I’ve heard it almost a million times before (and even seen it in textbooks): if you want less depth of field in a shot, put on a longer lens and back the camera off.
Unfortunately, this is not true. If you are trying to maintain the same shot, keeping the subject the same relative size in the frame, then when you switch lenses (longer focal length) and back off (further subject distance) you will maintain the same depth of field properties.
If we are shooting a medium close-up of an actor on a 50mm lens from 5 feet away and we switch to a 100mm lens, then, in order to maintain the same field of view of that actor—to keep the same medium close-up—we have to now move the camera back to 10 feet away. Although the 100mm lens has less depth of field at 5′ than the 50mm lens does, when we move the 100mm lens back to 10′ away, it now has the exact same depth of field as the wider lens at a closer focus.
A 50mm lens, at an f/4, focused at 5′ has a depth of field from 4’9″ to 5’3″, for a total acceptable focus area of 5″.
A 100mm lens, at f/4, focused at 10′ has a depth of field from 9’9″ to 10’3″, for a total acceptable focus area of 5″.
When we shifted to a longer lens and moved the camera to maintain the same shot, the depth of field did not change.
So, although longer lenses do have less depth of field, in practical application, they have the same optical properties. The only way to change your depth of field (in most situations) is to increase the size of your lens aperture. The wider your aperture, the less depth of field you’ll have. Inversely, the smaller your aperture, the more depth of field you’ll have.
Proof in the Pudding
By way of example, I set up a very quick demonstration as I was writing this column. I placed three playing cards about 8″ apart, and took out my Canon EOS 7D with my 28-135mm lens to do some quick illustration. The lens is a 5.6, widest, at the long end, so I picked 5.6 as my base aperture. Starting with the 28mm lens at 2′ away from the center card, I took an exposure; then, maintaining the aperture, changed my zoom to a 50mm lens and moved the camera back to 3’6″ to approximately match the center card as the same size in the frame, and took another exposure. Then I changed the focal length to 100mm and moved back to 7’2″ and took a third exposure.
I didn’t get the distances perfect—and, doing this too quickly, I was slightly off on my critical focus on the third exposure, but the rough demonstration stands to prove the point.
Because in the first exposure the closest card is actually closer than the lens’ minimum focus range, we’re only going to look at the middle and back card for comparison. Notice that as I change focal lengths and focus distances, the relative depth of field between the two cards does not change. The focus is slightly clearer in the third exposure because my critical focus was slightly off (you can clearly see that in the focus card) and I focused slightly beyond the card, so the far card is slightly more in focus—but the relationship of depth of field between them does not change.