Back To Listings RSS Print

What Wired Didn’t Understand about Depth of Field

It's not about analog vs. digital, it's frame size vs. focal length

By Art Adams | August 16, 2008

In the Sept. '08 issue of Wired Magazine, in an article about the RED camera called "A Star is Born," the author states that the difference in depth of field between film cameras and HD cameras is due to "analog versus digital," with no other explanation. I've written this for him and for anyone who wonders why 35mm film and large sensor HD cameras have so little depth of field.




Here we see a hypothetical 25mm lens mounted on a 35mm film camera or an HD camera with a 35mm-sized sensor. The lens casts a circle of light, and the sensor/film gate "samples" the center of the image. The depth of field remains roughly the same for both cameras because the film gate and the sensor are roughly the same size.



Here's the same lens, shooting the same subject at the same distance, but with a Super 16mm film gate or 2/3" sensor. All that's changed is that we're now sampling a smaller section of the image projected by the lens. We've gone from a wide shot to a close up just by changing film or sensor size, but the depth of field remains the same for a subject that distance from a 25mm lens. And while we'd expect a lot of depth of field for a wide shot, we don't expect it on a close up--but try to throw the background out of focus on a 25mm lens in Super 16mm or on a 2/3" HD camera. It's hard to do.

The difference in viewing angle isn't strictly 2:1 between 35mm and 16mm (35mm-sized sensor and 2/3" sensor) but it's close enough (around 2.4:1) to give you an idea of what's going on. And if you're wondering why you can't ever get the background out of focus on a 1/3" chip camera, it's because you're sampling an even smaller section of what the lens sees, so even shorter focal lengths are required. For example, while the 25mm lens above started out as a wide shot and became a head-and-shoulders shot simply by stepping down in sensor size, another step down in sensor size will make it a tight closeup--but still with a lot of depth of field.

To shoot the same shot on all the cameras we've spoken about (35mm, 2/3" and 1/3") we'd be looking at using (approximately) the following lenses to keep the same angle of view as in the top diagram (a wide waist-to-head shot):

35mm film = 25mm lens (a lot of depth of field)
2/3" = 12mm lens (tremendous depth of field)
1/3" = 6mm lens (near infinite depth of field)

All will yield the same size shot, with the camera at the same distance to the subject, but with drastically different depths of field.

If we shot a close up of this same person with an 85mm lens, here's what would happen:

35mm film = 85mm lens (not much depth of field)
2/3" = 42mm lens (twice the depth of field as 35mm film)
1/3" = 21mm lens (four times the depth of field of 35mm film, twice that of 2/3" sensor)

With each step we are doubling the depth of field... which is why you'll never get much out of focus when shooting with an HVX-200 or DVX-100.

Go to page two for some handy depth of field and angle of view rules...

Page 1 of 2 pages 1 2 Next »

Share This

Back To Listings RSS Print

Get articles like this in your inbox: Sign Up

Comments

billS: | August, 18, 2008

the so-called ‘normal’ lens is the measured diameter of the image…
i.e. on a 35 mm still camera its 24mmx35mm= 50 mm diagonal distance across the image..
on a hasselblad, its 80mm
on a 4x5 view camera, its 150mm…
which is assuming a “pinhole” lens.. ie simple lens…
but a modern lens is multi-element,—
so a 85 mm movie camera lens does not cast a 85mm circle of light… they concentrate the light coming through the lens, down to a smaller circle , to improve the f-stop/T-stop,  which can screw with the whole depth of field thing also…

david_winters: | August, 22, 2008

Here is a link to a very detailed article on DOF.  The “Background blur” section is my favorite.

“Much of the confusion in DOF discussions arises because people base their judgment on out-of-focus parts of the image. A shallow DOF is not synonymous with a generously blurred background. A shallow DOF implies that there is a shallow region in object space that is rendered acceptably sharp, regardless of whether the background is just not sharp or completely blurred.”

http://www.vanwalree.com/optics/dof.html

David S: | August, 28, 2008

Sorry to repeat myself from the Red One article, but you are getting a lot of this wrong:

Depth-of-field actually has to do with magnification, subject-to-camera distance and aperture opening.

“The smaller size of the sensor means that lenses must be approximately two times shorter in focal length to capture the same image size as seen on a 35mm sensor.”

BUT - the smaller size of the sensor means that the image must be magnified more to present the same-size projected (or printed) image. This increases the size of the “circles of confusion”, thereby decreasing the depth of field.

In actual fact, the size of the sensor has little (directly) to do with depth-of-field, if equivalent (for the sensor size) focal lengths and apertures are used, and the image is projected to the same size, you’ll get similar depth-of-field.

In practice it’s a little different, because you’d need correspondingly larger apertures on the wider lenses to get the equivalent shallow depth-of field.

You’re confusing focal length and sensor size with magnification. The focal length and sensor size have no direct bearing on depth of field. The magnification, subject distance and aperture do.

billS: | August, 28, 2008

DOF vs. format size
http://en.wikipedia.org/wiki/Depth_of_field#DOF_vs._format_size
To a first approximation, DOF is inversely proportional to format size. More precisely, if photographs with the same final-image size are taken in two different camera formats at the same subject distance with the same field of view and f-number, the DOF is, to a first approximation, inversely proportional to the format size. Strictly speaking, this is true only when the subject distance is large in comparison with the focal length and small in comparison with the hyperfocal distance, for both formats, but it nonetheless is generally useful for comparing results obtained from different formats.
To maintain the same field of view, the lens focal lengths must be in proportion to the format sizes. Assuming, for purposes of comparison, that the 4

David S: | August, 29, 2008

You use wikipedia as a source? Wow. This wouldn’t be the first time Wikipedia was wrong.

The Wikipedia article doesn’t take into account the magnification of the negative or sensor for viewing, either by projection, or printing.

You know what happens when you enlarge something? It gets softer. So, the smaller format needs more enlargement to reach the final viewing size.

In reality, images are magnified to the same viewing size, regardless of the source. So, if your target is an 8x10 print, the 35mm frame needs more enlargement than a medium or large-format frame.

This theory is only true if you are looking at the unmagnified negative or slide - which doesn’t happen in reality.

Another thing that people seem to be missing, is that there’s no such thing as “sharp”. There’s only varying sizes of “circles of confusion”. Enlargement affects the size of these circles. Even your “in-focus” parts of the picture won’t look sharp if you enlarge them enough.

Please login or register to comment