Jump to content


Photo

Long lens, little DOF question?


  • Please log in to reply
19 replies to this topic

#1 Aaron Solomon

Aaron Solomon
  • Basic Members
  • PipPip
  • 84 posts
  • Camera Operator

Posted 26 July 2011 - 10:21 AM

Why do long focal length lenses have less DOF then short lenses? I would think that since the F-stop = Focal Length/Lens Diameter, the long focal length lens will have a correspondingly larger F-stop and therefore smaller circle of confusion? Is it because the F-stop applies to each lens individually and can't be compared between lenses, and the short focal length lenses will be closed down (because they pass more light than the long focal length lenses) to give it more DOF?

Sorry, I just never bothered to find out the reason behind the long lenses having little DOF, and it's been annoying me for too long.

Aaron
  • 0

#2 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 26 July 2011 - 10:59 AM

There are two factors at play, one is the affect of focal length on DOF and the other is the affect of apparent DOF. Within a single format, changing focal length doesn't change DOF much if you compensate for the change of view by moving the camera closer or farther to maintain the same subject size. However, telephoto images "look" more shallow-focus because the same out-of-focus background that the wider-angle shot had is being cropped and enlarged relative to the subject, so the enlargement of the background makes the actual amount of out-of-focus look more visible, more obvious.

But remember that depth of field doesn't really refer to how out-of-focus a background looks, it refers to the range that the subject's distance can be to the lens and look in focus. Which is why I used the term apparent depth of field to describe the shallower-focus look of telephoto lenses due to the enlargement of the background behind the subject.

Now when comparing different format sizes where you need to change focal lengths when changing formats in order to maintain the same field of view and same distance to subject, then you find that shorter focal lengths have more depth of field than longer focal lengths when both are creating identical images due to the cropping of the different sized sensors or film formats.

If you compare the DOF figures for a 25mm and a 50mm lens, for example, at the same distance and f-stop, you see that a 25mm lens has the equivalent of 4-stops of increased DOF. But on the same format, the figure on the 25mm lens will be smaller because you didn't change distance. But put the 25mm lens on a format that is half the size so that the format is cropping the lens image by half, then you get the same FOV as the 50mm lens on the format that is twice as large. But because the smaller image has to be enlarged by twice to be shown on the same sized screen, your CoC figure is now twice as critical, so you lose 2-stops of DOF on the 25mm. But since it had 4-stops more DOF than the 50mm, the net difference is a 2-stop increase in DOF.
  • 0

#3 Aaron Solomon

Aaron Solomon
  • Basic Members
  • PipPip
  • 84 posts
  • Camera Operator

Posted 27 July 2011 - 11:34 AM

Thanks David! :)
  • 0

#4 Peter Ellner

Peter Ellner
  • Basic Members
  • PipPip
  • 30 posts
  • Student
  • California

Posted 12 August 2011 - 07:59 PM

There are two factors at play, one is the affect of focal length on DOF and the other is the affect of apparent DOF. Within a single format, changing focal length doesn't change DOF much if you compensate for the change of view by moving the camera closer or farther to maintain the same subject size. However, telephoto images "look" more shallow-focus because the same out-of-focus background that the wider-angle shot had is being cropped and enlarged relative to the subject, so the enlargement of the background makes the actual amount of out-of-focus look more visible, more obvious.

But remember that depth of field doesn't really refer to how out-of-focus a background looks, it refers to the range that the subject's distance can be to the lens and look in focus. Which is why I used the term apparent depth of field to describe the shallower-focus look of telephoto lenses due to the enlargement of the background behind the subject.

Now when comparing different format sizes where you need to change focal lengths when changing formats in order to maintain the same field of view and same distance to subject, then you find that shorter focal lengths have more depth of field than longer focal lengths when both are creating identical images due to the cropping of the different sized sensors or film formats.

If you compare the DOF figures for a 25mm and a 50mm lens, for example, at the same distance and f-stop, you see that a 25mm lens has the equivalent of 4-stops of increased DOF. But on the same format, the figure on the 25mm lens will be smaller because you didn't change distance. But put the 25mm lens on a format that is half the size so that the format is cropping the lens image by half, then you get the same FOV as the 50mm lens on the format that is twice as large. But because the smaller image has to be enlarged by twice to be shown on the same sized screen, your CoC figure is now twice as critical, so you lose 2-stops of DOF on the 25mm. But since it had 4-stops more DOF than the 50mm, the net difference is a 2-stop increase in DOF.


Quick follow-up question: correct me if I'm wrong, but as far as I know, the size of the aperture determines the depth of field because it determines the range of angles that light rays in a scene are accepted to the image plane. A small aperture gives a deep depth of field because it acts like a pinhole camera and only allows a small portion of light rays to pass through, thus preventing all of the light rays from an object overlapping and creating an out-of-focus effect.

But here's my question: in order for a zoom lens to maintain a constant f-stop throughout the zoom range, the f-stop formula tells me that the size of the entrance pupil (apparent size of the aperture as viewed magnified through the lens) would actually have to increase. So that means that the size of the entrance pupil at the telephoto end of a lens is actually larger than at the wide end. Thus the depth of field should become shallower when a zoom lens is at the telephoto setting than at the wide setting...yet this doesn't seem to be the case, as depth of field doesn't change when a lens with a constant f-stop zooms. So why isn't there a shallower depth of field if the size of the entrance pupil is bigger?

Thanks, a really appreciate it!
  • 0

#5 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 12 August 2011 - 10:24 PM

Quick follow-up question: correct me if I'm wrong, but as far as I know, the size of the aperture determines the depth of field because it determines the range of angles that light rays in a scene are accepted to the image plane. A small aperture gives a deep depth of field because it acts like a pinhole camera and only allows a small portion of light rays to pass through, thus preventing all of the light rays from an object overlapping and creating an out-of-focus effect.

But here's my question: in order for a zoom lens to maintain a constant f-stop throughout the zoom range, the f-stop formula tells me that the size of the entrance pupil (apparent size of the aperture as viewed magnified through the lens) would actually have to increase. So that means that the size of the entrance pupil at the telephoto end of a lens is actually larger than at the wide end. Thus the depth of field should become shallower when a zoom lens is at the telephoto setting than at the wide setting...yet this doesn't seem to be the case, as depth of field doesn't change when a lens with a constant f-stop zooms. So why isn't there a shallower depth of field if the size of the entrance pupil is bigger?

Thanks, a really appreciate it!


The f-stop determines the depth of field, not the size of the aperture -- but the size of the aperture relative to the focal length of the lens, i.e. the f-stop.
  • 0

#6 Tom Jensen

Tom Jensen
  • Basic Members
  • PipPipPipPip
  • 1234 posts
  • Camera Operator
  • Dallas, TX

Posted 13 August 2011 - 01:45 AM

Zoom lenses don't maintain a constant f stop through out the range. As the lens gets wider, it becomes faster. The reason that the zoom lenses are slower is so that there is a dedicated stop where the stop won't change. Say if the lens was a 5.6 and on the wide end it is a 2.8 and you could open it up to a 2.8 and you zoom all the way in, the image will go dark because the fastest the lens is on the long end is a 5.6. It has to be dedicated to the slowest stop throughout the range. I hope that made sense.

the subject distance or footage also determines depth of field.

Edited by Tom Jensen, 13 August 2011 - 01:46 AM.

  • 0

#7 Brian Drysdale

Brian Drysdale
  • Basic Members
  • PipPipPipPip
  • 5070 posts
  • Cinematographer

Posted 13 August 2011 - 03:25 AM

Zoom lenses don't maintain a constant f stop through out the range.


You can get constant f stop zoom lenses, the difference being that they are much larger than the zoom lenses that ramp the stop. My understanding of the reason is that the front diameter of the lens begins to restrict the amount of light passing though at the larger focal lengths, even though the stop on the iris remains the same.
  • 0

#8 Peter Ellner

Peter Ellner
  • Basic Members
  • PipPip
  • 30 posts
  • Student
  • California

Posted 15 August 2011 - 10:50 PM

The f-stop determines the depth of field, not the size of the aperture -- but the size of the aperture relative to the focal length of the lens, i.e. the f-stop.


Wait a minute, now I'm really confused! In your first post you said that focal length doesn't affect depth of field, it only magnifies or minimizes the size of the out of focus regions (and I assume the extent of the in focus region is what you define as depth of field since this is what I go by). So focal length doesn't affect depth of field, this was my thought prior to reading your post, which you only backed up.

But now you're saying the size of the aperture doesn't affect the depth of field, but rather the f-stop. How can this be? If, as you say, the size of the aperture doesn't affect the depth of field, and the focal length doesn't affect the depth of field... then how can the f-stop (focal length divided by aperture diameter) affect it?

Either the size of the aperture is the only thing that affects the depth of field, or both the aperture and the focal length can affect the depth of field individually and together, but I don't get how neither the aperture now the focal length can affect it while the f-stop can. Since the f-stop is a function of both, I don't see how two things that don't affect the depth of field can be taken together to create something that does.

I trust your expertise, but I must be really confused because it seems that you're contradicting yourself.
  • 0

#9 Dom Jaeger

Dom Jaeger
  • Basic Members
  • PipPipPipPip
  • 1599 posts
  • Other
  • Melbourne, Australia

Posted 16 August 2011 - 01:37 AM

It can be a very confusing subject.

Mathematically speaking, the factors affecting depth of field are the distance to subject, focal length and aperture.

(There's also the circle of confusion value which changes depending on the viewing criterion and format and how much loss of focus is still considered acceptably sharp, but that's outside the lens factors.)

Now because of the way subject distance and focal length are related, if the subject size remains the same across different focal lengths (by altering the distance), those 2 factors cancel each other out leaving only aperture as the determining factor of the depth of field. This is what David was explaining.

The relative aperture or f-stop is is a measurement of how much light is transmitted through a perfect lens and is defined, as you said, by the ratio of the focal length to the the diameter of the entrance pupil. In a zoom the entrance pupil diameter (basically the image of the aperture as seen from the subject space) will contract as the focal length increases, maintaining a constant f-stop.

So yes, fundamentally the entrance pupil diameter, the focal length, the distance to subject and the aperture all affect depth of field, but because these factors are all inter-related, the simplest way to understand it is: for a given format, with a constant subject size (and a chosen circle of confusion), the only factor determining depth of field is the f-stop.
  • 0

#10 Dom Jaeger

Dom Jaeger
  • Basic Members
  • PipPipPipPip
  • 1599 posts
  • Other
  • Melbourne, Australia

Posted 16 August 2011 - 01:59 AM

Oops :rolleyes:

That was meant to be "In a zoom the entrance pupil diameter will expand as the focal length increases".
  • 0

#11 Dom Jaeger

Dom Jaeger
  • Basic Members
  • PipPipPipPip
  • 1599 posts
  • Other
  • Melbourne, Australia

Posted 16 August 2011 - 03:06 AM

Zoom lenses don't maintain a constant f stop through out the range. As the lens gets wider, it becomes faster. The reason that the zoom lenses are slower is so that there is a dedicated stop where the stop won't change.


Yes, depending on their design and range, some zooms will have their fastest f-stop restricted as they approach the long end, as the size of the optics limits the entrance pupil diameter.

Some older zooms allowed for a faster f-stop at their wide end, but the image would visibly darken if you zoomed in. To maintain a constant aperture the iris needed to be stopped down to the long end limit or beyond.
  • 0

#12 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 16 August 2011 - 07:19 PM

Wait a minute, now I'm really confused! In your first post you said that focal length doesn't affect depth of field, it only magnifies or minimizes the size of the out of focus regions (and I assume the extent of the in focus region is what you define as depth of field since this is what I go by). So focal length doesn't affect depth of field, this was my thought prior to reading your post, which you only backed up.

But now you're saying the size of the aperture doesn't affect the depth of field, but rather the f-stop. How can this be? If, as you say, the size of the aperture doesn't affect the depth of field, and the focal length doesn't affect the depth of field... then how can the f-stop (focal length divided by aperture diameter) affect it?

Either the size of the aperture is the only thing that affects the depth of field, or both the aperture and the focal length can affect the depth of field individually and together, but I don't get how neither the aperture now the focal length can affect it while the f-stop can. Since the f-stop is a function of both, I don't see how two things that don't affect the depth of field can be taken together to create something that does.

I trust your expertise, but I must be really confused because it seems that you're contradicting yourself.


I've probably caused confusion by using the terms aperture and f-stop interchangeably. F-stop affects depth of field and aperture size a part of the f-stop but relative to focal length by definition. But that's not the same thing as saying that since f-stop affects depth of field, the focal length affects the depth of field. It's just that the f-stop is more than the physical size of the aperture.

Look, if you want to visualize this, just take a photo with a wide-angle lens, print it, and then crop that photo and enlarge it and print that -- in essence with the cropped photo you'd get the same image as if you had zoomed in on a zoom lens or switched to a more telephoto image. In other words, the telephoto image is just a cropped version of the wide angle image -- the amount of soft focus in the background is the same because it's the same image, it's just in the cropped and enlarged version, the background behind the subject is going to be enlarged, making the amount of softness in the background more visible. Zooming into the image didn't decrease the depth of field, it just enlarged the image and made the depth of field more obvious.
  • 0

#13 Adrian Sierkowski

Adrian Sierkowski
  • Sustaining Members
  • 7117 posts
  • Cinematographer
  • Los Angeles, Ca

Posted 16 August 2011 - 07:49 PM

David, I think that's the best/easiest way I've heard of explaining it. I think I'll have to steal it (with proper credit of course).
  • 0

#14 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 16 August 2011 - 07:58 PM

I think one thing people should keep reminding themselves of is that depth of field describes the focused distance range when the subject looks acceptably in focus -- not how much the background behind the subject looks out of focus.
  • 0

#15 Aaron Solomon

Aaron Solomon
  • Basic Members
  • PipPip
  • 84 posts
  • Camera Operator

Posted 24 August 2011 - 03:09 PM

So even if the aperture is changed during the zoom, there's no actual change in the depth of field?
  • 0

#16 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 24 August 2011 - 03:27 PM

So even if the aperture is changed during the zoom, there's no actual change in the depth of field?


As a "constant aperture" zoom lens changes focal length, it actually changes the aperture size to maintain the same f-stop, so the exposure doesn't shift as you zoom.

Sure, there is a little change in depth of field as you change focal length so it's not exactly the same thing as just cropping the shot, partly because the Circle of Confusion figure changes if you compared a cropped & enlarged image to one shot on a longer lens, where essentially the projected lens image is larger compared to magnifying the final recorded image to achieve the same image.
  • 0

#17 Tom Jensen

Tom Jensen
  • Basic Members
  • PipPipPipPip
  • 1234 posts
  • Camera Operator
  • Dallas, TX

Posted 24 August 2011 - 03:43 PM

So even if the aperture is changed during the zoom, there's no actual change in the depth of field?


In would would depend on how far you zoomed and how much the aperture changed and the direction of the zoom. For all intents and purposes, these are the factors that affect DOF (1) Aperture, (2) Focal Length, (3) Subject to focal plane distance, (4) Format size.
  • 0

#18 Aaron Solomon

Aaron Solomon
  • Basic Members
  • PipPip
  • 84 posts
  • Camera Operator

Posted 24 August 2011 - 08:18 PM

Thanks!
  • 0

#19 Brian Drysdale

Brian Drysdale
  • Basic Members
  • PipPipPipPip
  • 5070 posts
  • Cinematographer

Posted 25 August 2011 - 01:54 AM

As a "constant aperture" zoom lens changes focal length, it actually changes the aperture size to maintain the same f-stop, so the exposure doesn't shift as you zoom.


First I heard that this happened, although I'm not saying that it doesn't. My understanding of the ramping effect is that the front diameter of the lens begins to restrict the light coming though at the longer focal lengths (effectively acting as an iris at the front of the lens). This sort of makes sense when comparing the constant aperture versions of a similar zoom range lens (eg the Fujinon Cine lenses), which are physically much larger than the compact ENG ramping aperture versions.
  • 0

#20 william koon

william koon
  • Basic Members
  • PipPip
  • 82 posts
  • Student

Posted 29 August 2011 - 10:16 AM

As a "constant aperture" zoom lens changes focal length, it actually changes the aperture size to maintain the same f-stop, so the exposure doesn't shift as you zoom.

Sure, there is a little change in depth of field as you change focal length so it's not exactly the same thing as just cropping the shot, partly because the Circle of Confusion figure changes if you compared a cropped & enlarged image to one shot on a longer lens, where essentially the projected lens image is larger compared to magnifying the final recorded image to achieve the same image.

This is news to me indeed. All this while my understanding is, as has been explained in almost all photography books, DOF is influenced by subject's distance to the lens, focal length, iris and arguable film format. David, your explanation is logical but I am really confused.
  • 0


Metropolis Post

Aerial Filmworks

CineTape

Visual Products

Tai Audio

CineLab

Media Blackout - Custom Cables and AKS

Technodolly

Rig Wheels Passport

Broadcast Solutions Inc

Abel Cine

Wooden Camera

Gamma Ray Digital Inc

Ritter Battery

rebotnix Technologies

Opal

Willys Widgets

FJS International, LLC

The Slider

Paralinx LLC

Glidecam

Paralinx LLC

Aerial Filmworks

Visual Products

Willys Widgets

Tai Audio

CineTape

Technodolly

Wooden Camera

Rig Wheels Passport

Abel Cine

Media Blackout - Custom Cables and AKS

Gamma Ray Digital Inc

Opal

Metropolis Post

rebotnix Technologies

The Slider

FJS International, LLC

CineLab

Ritter Battery

Broadcast Solutions Inc

Glidecam