How is shallow Depth of field (DOF) achieved?

04/11/2006 09:57:11 PM · #1

I had to admit - the more I know, the more questions I have.
Today I have a question on DOF. I have read a little, and I think I know why DOF happens and how light is focused, but how do the following items gets into the consideration, and how do they affect the final DOF?

- sensor physical size (not pixel count)
- lens size
- focal length (from lens to sensor)
- aperture size (is this tied to lens size or sensor size?)

Finally, why does some cameras have shallower DOF than others?
What are the factors?

Thanks

Message edited by author 2006-04-11 21:57:44.

Sigma 18-200mm f/3.5-6.3 DC OS for Canon

04/11/2006 10:01:51 PM · #2

The longer the focal length of the lens, the shallower the DOF at any given focusing distance.

The larger the aperture, for a given focal length lens, the shallower the DOF at a given focusing distance.

Sensor size has no bearing on DOF except inasmuch as more detailed images are perceived as sharper than less detailed ones. DOF is strictly a function of the optics of a given lens, and the lens has the same DOF no matter what camera it is mounted to.

Take my 70-200mm and mount it on Kirbic's 5D, and the DOF is identical; it just shows a wider field of view on his camera than on mine.

Robt.

Message edited by author 2006-04-11 22:02:10.

viajero

Canon EOS-20D

04/11/2006 10:26:58 PM · #3

More than you'd ever want to know about Depth Of Field:

//www.dofmaster.com/

It goes into derivations and equations, or they have an applet that will print out a handy chart for you.

Be sure to check out the part about Hyperfocal Distance. Very useful for landscape shots.

04/11/2006 10:30:20 PM · #4

Thanks Robt.
I have seen most websites also states that DOF is governed by only 2 factors:

1. aperture size
2. focal length

However, I also stumbled on this tutorial which defuncts the theory that Focal Length has anything to do with DOF. Please have anyone read this and tried it out? In summary, it says that at the same aperture, whatever focal length used will produce the same DOF. How true is this? My personal experience shows that at the same aperture, say F3.2, I get shallower DOF when at 300mm than when I'm at 38mm. Now, back to my initial post, could it have something to do with sensor's physical size (not pixel count)? Thanks

Message edited by author 2006-04-11 22:31:13.

04/11/2006 10:45:43 PM · #5

I think I'm starting to grasp the idea of how aperture size and focal length plays a role in DOF. (see image below) But I still trying to figure out if sensor's physical size matters at all?

04/11/2006 11:06:13 PM · #6

The sensor size matters _IF_ and only if, you compose the scene the same way with one size sensor versus another.

In other words, now that I have a 5D (full frame sensor) I can get shallower DOF on a full body shot with my 200mm lens than I could with my 20D -- because to fit that same person in the frame (full body, remember) I would have to back up quite a bit further away! And thus I would lose my shallow DOF by being further away.

So, the only thing that PHYSICALLY matters is focal length and aperture. But compositionally, the sensor size affects DOF as well because you are forced to move further away with a smaller sensor or shot with a wider angle lens.

paddles

Pentax K10D

Pentax FA 50mm f/1.4 AF

04/11/2006 11:06:59 PM · #7

Originally posted by crayon:

But I still trying to figure out if sensor's physical size matters at all?

No, the physical sensor makes no difference to depth-of-field.

04/11/2006 11:10:36 PM · #8

Originally posted by paddles:

Originally posted by crayon:
But I still trying to figure out if sensor's physical size matters at all?

No, the physical sensor makes no difference to depth-of-field.

When you answer with such a hard-lined "No" (without mentioning the compositional aspects of it), you leave all the P&S shooters out there scratching their heads wondering why they can't get the same shallow DOF that a digital SLR can get. They've been told they are using a lens in "35mm equivalent values" and yet the DOF they get doesn't match the same DOF they get with a real 35mm sensor using that focal length.

You're right. Of course. But I wouldn't leave off the compositional aspect of the sensor size. You simply cannot shoot the same exact scene, with the same exact focal length, if the sensor size is different. Sensor size matters. Compositionally speaking.

Message edited by author 2006-04-11 23:11:28.

Panasonic LUMIX-DMC ZS100

04/11/2006 11:20:25 PM · #9

Originally posted by dwterry:

But compositionally, the sensor size affects DOF as well because you are forced to move further away with a smaller sensor or shot with a wider angle lens.

That's very helpful! And you're right about me scratching my head as to why I cant get the same shallow DOF as most dSLR at the same camera settings... well now I know. Thanks!

p/s: I'm gonna go out and get myself a larger sensor and a faster lens (F1.0 here I come, lol)

wheeledd

Canon EF 24-105mm f/4.0L IS

04/11/2006 11:23:28 PM · #10

Originally posted by crayon:

I think I'm starting to grasp the idea of how aperture size and focal length plays a role in DOF. (see image below) But I still trying to figure out if sensor's physical size matters at all?

Yes, it does matter, but indirectly. The "normal" lens for a camera is the one that gives you an image that is about the width of what the eye sees when it looks at a scene. The normal lens for a 35 mm camera is 50 mm. If you have a camera with a 1.6 lens factor, you will need a 50/1.6 = 31.25 mm lens to give the same view. This is a SHORTER focal length lens so it will have a GREATER depth of field. People who shoot with cameras with small sensors (and therefore short focal-length lenses) have a great deal of trouble getting shallow DOF to isolate subjects from the background. People who shoot with large-format cameras (4x5s or 8x10s) have the opposite problem--they are always fighting to get enough depth of field.

The tutorial you posted was interesting. The key point is that they were moving to keep the subject the same size. Suppose you focus on something 10 feet away and everything from 9 feet to 11 feet is in focus--so the depth of field is 2 feet. If you focus on something really close, like a foot away, the depth of field will be much smaller.

Now assume that you focus on something 10 feet away with a telephoto lens. You will have a small depth of field. Now switch to a wide-angle lens and you will have a much greater depth of field. But to get the same size image, you have to move much closer! Since you are closer, the DOF will be smaller. It turns out that the DOF of the close view with the wide-angle lens will be the same as the DOF view of the telephoto lens when you move the camera to keep the image size the same.

I hope this helps. I've been doing this stuff for a long time and the intellectual understanding of these relationships did not come easily.

--DanW

04/11/2006 11:28:17 PM · #11

Originally posted by crayon:

I have seen most websites also states that DOF is governed by only 2 factors:

1. aperture size
2. focal length

So, in case the last couple of posts don't clarify it ... you should add one more item to the above list:

3. distance

The closer you are to the subject, with a given aperture and focal length, the shallower the DOF. The further you are, keeping the same aperture and focal length (remembering, too, that you've just changed the composition), the more DOF you will have.

04/11/2006 11:33:15 PM · #12

Originally posted by dwterry:

3. distance

The closer you are to the subject, with a given aperture and focal length, the shallower the DOF. The further you are, keeping the same aperture and focal length (remembering, too, that you've just changed the composition), the more DOF you will have.

EDITED: I think I got it now. Meaning, I could get the shallowest DOF if I could do all the following?

longest focal length + as near as possible to subject + largest aperture

Thanks

Message edited by author 2006-04-11 23:39:27.

04/11/2006 11:42:21 PM · #13

Originally posted by crayon:

Aperture, when defined as an f/stop. is a ratio, not an absolute. It's the ratio of the physical size of the aperture divided into the focal length of the lens. So a 25mm aperture on a 50mm lens is f/2, and a 25mm aperture on a 200mm lens is f/8.

DOF, as it relates to aperture (which indeed it does) is a function of the physical diameter of the aperture so you have to stop long lenses down further to get same DOF as shorter lenses. A lot further, obviously.

The description I gave earlier takes this into account by speaking of a "given f/stop". You have more DOF at f/8 on a 24mm than you do at f/8 on a 200mm; a whole LOT more. But this isn't true if the physical diameter of the aperture remains constant. And even beyond that, when you get past a certain physical size in the aperture you begin to have diffraction problems that lessen the sharpness of the transmitted image and lessen the perceived DOF.

Perceived DOF is subjective. Just as sharpness is subjective. The same shot looks mUCH sharper on glossy paper than on matte paper, for example. Likewise, shots look sharper at a small szie than a larger size. But as you back away from a very large image so you view it from a greater and greater distance, it tends to sharpen up, up to a point.

So it is correct, strictly speaking, to say that "DOF is independent of the focal length of the lens", because in that sense it is; given identical physical diameter of aperture, this is essentially true as far as mathematical formulas go. But people don't shoot that way. They don't think that way.

And in practical terms, it is TRUE that at a given f/stop and a given focal distance, a 200mm lens has less DOF than a 50mm lens. This is because the 50mm is using a significantly smaller physical aperture. So in general, it makes more sense to "teach" people that DOF decreases as focal length increases, and increases as the f/stop gets smaller at any given focal length.

Robt.

Message edited by author 2006-04-11 23:44:31.

04/11/2006 11:44:12 PM · #14

Originally posted by crayon:

EDITED: I think I got it now. Meaning, I could get the shallowest DOF if I could do all the following?

longest focal length + as near as possible to subject + largest aperture

Thanks

Yep.

Robt.

Panasonic LUMIX-DMC ZS100

04/11/2006 11:46:15 PM · #15

Originally posted by crayon:

Meaning, I could get the shallowest DOF if I could do all the following?

longest focal length + as near as possible to subject + largest aperture

Yup. And when you move into the world of "macro photography" (where the subject in front of the lens is the same size as the image projected onto the sensor) you'll find that you are SO close, or using such a LONG lens, that you end up having to use a much smaller aperture *just* to get enough DOF to contain the tiny subject!

For example, I had to use f/9 on this shot just to get a fraction of a centimeter in focus:

The image is of a tiny part of a small cactus plant. (if you view the image page, you'll see another post from me showing the relative size)

wheeledd

Canon EF 24-105mm f/4.0L IS

04/11/2006 11:53:10 PM · #16

Originally posted by crayon:

wait a sec... you mean the closer I am to the subject, the shallower DOF I get? I always thought it was the other way around? I mean, when I wan shallow DOF on my tiny sensor, I usually stand far away and use the longest focal length (with the largest Aperture) - am I doing it right? From your reply above, would I get even shallower DOF if I had stood close to my subject and used Macro mode? (think I'm getting confused again, lol)

But you have a zoom lens and you are changing the focal length. We are talking about the change in the DOF that you get just by moving closer. When you are in macro mode, the front of the flower may be in focus but the back (half an inch further away) may be out of focus. But if you shoot a portrait of someone (at the same zoom) and focus on the eyes, the ears (3 inches further away) will probable be in focus. You have much greater DOF in inches when you shoot a person four feet away than when you shoot a flower four inches away.

One of the reasons this is hard to think about is that there are several factors and usually you change several of them at once. If you move closer, you zoom out to get the same view, so the DOF actually remains the same. The DOF change from moving closer is cancelled out by the DOF change from changing the focal length by zooming out.

The point of the tutorial was that your strategy of standing far away and using a long focal length to get greater DOF does not work. To get shallow DOF, you need to do what David and I have done--get 5Ds.

--DanW

04/11/2006 11:57:05 PM · #17

Originally posted by wheeledd:

The point of the tutorial was that your strategy of standing far away and using a long focal length to get greater DOF does not work.

I understand where the tutorial is pointing to now, thanks!

Originally posted by wheeledd:

To get shallow DOF, you need to do what David and I have done--get 5Ds

LOL!!! (I wish I could afford one too)

04/11/2006 11:59:14 PM · #18

Originally posted by crayon:

Look at it this way; with any given lens, at any given aperture, the closer you focus the less DOF you have.

Period. With any given lens, the closer you get to your subject the less DOF you have. When you change the zoom, you are essentially changing lenses.

Robt.

Sigma 24-70mm f/2.8 EX Aspherical DF for Canon

04/12/2006 12:17:01 AM · #19

I thought this comparison might be interesting:

I wear contacts. With my contacts on I have really good long distance vision (20/15, I believe). I can see perfectly from arms distance all the way out to infinity.

But I'm getting older. My eyes don't focus very close any more - at least, not with my contacts on.

If I take my contacts off, I can focus clearly on objects that are mere inches from my nose! But move it out another foot and my ability to focus quickly drops off.

So.... up close, without my contacts, my DOF is measured in inches. But further out, with my contacts on, it seems as if my DOF is nearly infinity. :-)

hankk

Canon EOS-5D

04/12/2006 09:40:59 AM · #20

Originally posted by Bear_Music:

Sensor size has no bearing on DOF except inasmuch as more detailed images are perceived as sharper than less detailed ones. DOF is strictly a function of the optics of a given lens, and the lens has the same DOF no matter what camera it is mounted to.

Take my 70-200mm and mount it on Kirbic's 5D, and the DOF is identical; it just shows a wider field of view on his camera than on mine.

If you've got a P&S, you should compare the actual lens length, not the "35 mm equivelant".

In Robt's example, his 70-200mm has a 70-200mm length on the full frame5D and a 112-320mm "35mm equivalent" length on the 1.6 crop factor 20D.

The Canon Powershot S3 has a 6.0-72.0mm f/2.7-3.5 lens (35mm film equivalent: 36-432mm). So the DOF of both the S3 lens and the 70-200 lens will be the same if they're all set to 70mm and the same f-stop. But the "35mm equivelant" will be 70mm on the 5D, 112mm on the 20D, and somewhere around 400mm on the S3.

Message edited by author 2006-04-12 09:42:15.

jaxsond

Canon EOS-1D Mark II

Canon EF 50mm f/1.4 USM

04/12/2006 01:36:29 PM · #21

OT but important info .... my pet peeve is further vs. farther ...
"farther" think "far" measures distance, "further" measures time.

swankFoto

Fujifilm X-E1

Nikon AF-S DX NIKKOR 18-200mm f/3.5-5.6G ED VR II

04/12/2006 01:43:52 PM · #22

i like corn

Message edited by author 2006-04-12 13:45:22.