DOF-I'm Confused?

[Stu W]

Ok, I'm always hearing people talking about stopping down their 1.5 Summicron to give them the same DOF as an Elmar 3.5 My question is this. Is DOF governed by the len's light transmitting design or by the fact that stopping down a fast lens gives it the pinhole effect? A small enough pinhole should theoretically have an infinite DOF. My 500mm f8 SLR lens has about 6 inches of DOF but my f1.2 stopped down to f8 doesn't need to be focused. Please unconfuse me. Or unconfuse "them". Stu

 

[danielnorton]

my take

my take

3 factors make up DOF

1- focal length of the lens, the longer the lens the less DOF
2- Aperture, the wider the aperture the less DOF
3- distance to subject, the closer you are to the subject you are focusing on, the less DOF



Not sure if this is what you mean, but if you take 2 lenses of the same focal length and set them both at the same aperture (3.5 in this case) and focus them at the same distance, they should have the same DOF.

 

[Socke]

Ok, I'm always hearing people talking about stopping down their 1.5 Summicron to give them the same DOF as an Elmar 3.5 My question is this. Is DOF governed by the len's light transmitting design or by the fact that stopping down a fast lens gives it the pinhole effect? A small enough pinhole should theoretically have an infinite DOF. My 500mm f8 SLR lens has about 6 inches of DOF but my f1.2 stopped down to f8 doesn't need to be focused. Please unconfuse me. Or unconfuse "them". Stu

The shorter the lens and the smaller the aperture the bigger DoF.

You're right with the pinhole effect but you have to factor in the enlargement by the lens.

The formula for DoF as it appears in the "American Cinematographers Manual," 8th edition, pages 698-699.

Hyperfocal Distance: h = f^2 / ac

f = focal length of lens
a = aperture diameter (f/stop number)
c = circle of confusion

Depth of Field, Near limit: hs / h + (s - f)

Depth of Field, Far limit: hs / h - (s - f)

h = hyperfocal distance
s = distance from camera to object
f = focal length of lens

 

[Huck Finn]

www.vanwalree.com/optics/dof.html

 

[beethamd]

Stu,
When light is refracted in the lens, it exits at cones of light, arriving at the image plane. With a f/1.2 lens, the aperture is wide and so the cones of light are wide when passing through the aperture arriving in focus at the pointy end of the cone at the image plane. As this cone of light is so wide at the aperture and a point at the film plane you see how at only a very small distance away from the film plane the image is not in focus. When the lens is stopped down, the cone of light is smaller at the aperture and gradually becomes a point at the image plane, thus giving more depth of field.

The focal length of the lens does not affect DOF.

 

[Gabriel M.A.]

I think part of the confusion may be that no such thing as a f/1.5 Summicron exists 😉

And, btw, "focal length of the lens does not affect DOF" is true. DOF is determined by the focal length, aperture, and "format" (i.e. 35mm, 6x6, etc.) however.

 

[VinceC]

>>The focal length of the lens does not affect DOF<<

In actual, nontheoretical usage, it does. A 50mm lens focused close and set to f/3.5 or f/4 has a depth of field measured in inches whereas a 28mm has a DOF measureable in feet and a 21mm has a DOF measured in meters.

beethamd's explanation is dead-on from a scientific/engineering point of view. However, another variable in depth-of-field calculations is the resolution of the film, as well as the resolution of the final print size. Depth of field is, more or less, that point where the cone of confusion/apparent sharpness intersects with the resolution of the film and/or a printed image held at viewing distance.

Old film had lower resolution, so depth of field scales from the 1940s and 1950s tend to be a little too generous.

 

[Poptart]

Lenses have the same DOF at a particular distance; the difference is in perspective & image size. A 50mm lens at 10' will produce a very different photo of a field mouse than will a 500mm lens, but if you enlarge the mouse in the 50mm pic, he'll have the same depth of field. The difference is in foreshortening, which will be more pronounced with the shorter lens.

 

[Socke]

Reichman has a nice demonstration of the effects on his site

DoF with different focal lenghtes

 

[wdenies]

Be aware that the principle of DOF is based on a optical illusion or a shortcoming of the human eye as you want: simply stated up to a certain diameter the eye or the brain can not distinguish between a sharp point and an unsharp circle. We see both as sharp!
BUT:
What happens when enlarging a picture? Yes, the diameter of the "unsharp" points increase and at a cerain enlargment our brain will tell this is not sharp anymore. Thus enlargment decreases DOF.

The DOF formulas are based on a well defined print size (arround A4 I think).
Make the experiment a postcard print that looks razor sharp up to infinity will show a well noticable DOF when enlarged to 30x40

Wim

 

[richard_l]

Lenses have the same DOF at a particular distance; the difference is in perspective & image size. A 50mm lens at 10' will produce a very different photo of a field mouse than will a 500mm lens, but if you enlarge the mouse in the 50mm pic, he'll have the same depth of field. The difference is in foreshortening, which will be more pronounced with the shorter lens.

At a particular distance, different focal lengths will have different DOFs but the same perspective (foreshortening). Geometric perspective depends only on the camera to subject distance.

The reason long lenses are thought to flatten perspective and wide lenses appear to exaggerate perspective is because longer lenses are generally used when the camera to subject distance is greater than when wider lenses are used.

http://pw2.netcom.com/~rlsaylor/misc/per1.jpg

 

[Poptart]

Sorry, I was drunk. You're right, depth is the same at the same image size, but perspective is different. Perspective is the same at the same subject to film distance.

 

[jaapv]

What I missed up to now,although Socke implied this in his formula: The DOF is not equal front or back of the focal plane. The ratio is 1/3 to 2/3. This effect becomes more noticable as the subject distance decreases.

 

[wdenies]

As I already mentioned, the perception of DOF is subjective.
The DOF indication on most lenses are based on a printsize of 8x10 inch, viewing distance 25 cm, circle of confusion 0.01 inch.
This definition implies that DOF is influenced by print size and viewing distance.
Examples:
50 mm lens on a 35mm camera, f8, focus distance 3m

Printsize: 25 cm
viewing distance 25 cm DOF: 2.28-4.4 m
50 cm 1.84-8.18 m

Printsize: 40 cm
viewing distance 25 cm DOF: 2.5-3.75 m
50 cm 2.15-4.96 m
1 m 1.68-14 m

In the past this viewing distance trick was used by Minox. On shows (Photokina) they showed large razor sharp enlargments of their mini format, but.... stands were arranged in such a way that spectators had to stay 4-5 m from those pictures.

 

[jaapv]

As I already mentioned, the perception of DOF is subjective.
The DOF indication on most lenses are based on a printsize of 8x10 inch, viewing distance 25 cm, circle of confusion 0.01 inch.
This definition implies that DOF is influenced by print size and viewing distance.

This is very true, and is actually one of the discussion points in the digital world
regarding sensor crop. Of course, photographers working in different formats have known this forever. The confusion there is that the relationship between apparent focal length/format and DOF is not a linear one. in other words, an effective 180 mm 35 mm equivalent on a 1.3 sensor crop camera, being a 135 mm in the real photographic world gives the DOF of a 160 mm lens.
This was one other Minox trick, their photographs had the perspective of a 50 mm lens and the DOF of a 21 mm lens.

 

[VinceC]

Quantum mechanics tells us it's impossible to reliably photograph anything because our subject can be in more than one place at a time and that the act of recording them will alter their existance. Nonetheless, these quantum errors are generally not meaningful in real-world applications.

I'm not sure if this thread has helped clear up Stu W's confusion. Reichman's example, linked above, is interesting. But how useful is to someone learning how to handle a camera? This optical illusion is achieved by dramatically altering the camera-to-subject distance for each photograph, thereby dramatically changing perspective. The 17mm view in his example would be unachievable with a rangefinder because the lens is just a few inches from the foreground subject.

If you pick up a camera and start using it, the difference in depth of field from one lens to another becomes very important. Elsewise, we wouldn't have other forum threads discussing whether or not a rangefiinder can be accurately focused beyond the 90mm focal length and why the base length on the Bessas should have been longer.

In common usuage, depth of field becomes very important when combined with changing perspectives, and those depth-of-field markings on lenses are critical to learning good photography. For example, if you sit down and READ your depth of field scales, you'll see what JAAPV missed up until now -- that the zone of acceptible focus is 2/3 longer behind the subject than in front of it. Those scales are calibrated for distance from the camera's film plane to the subject, so changing lenses is an act of changing perspectives. (So is walking around.) If I'm taking a picture of kids in my living room, my 21mm lens focused at six feet and shot at f/4.5 will give me a depth of field of measurable in meters -- about 3.5 feet to 50 feet; a 28mm lens focused to six feet away at f/4 will give a depth of field of about 4-feet-9-inches to 8 feet. DOF for my 50mm lens focused at six feet at f/4 is 5.5 feet to 6.5 feet. With the 85mm lens, DOF is 4 inches. With the 135mm lens, DOF is 1.5 inches.

 

[wdenies]

VinceC
I think you missed the point

 

[jaapv]

The point being that DOF is not an absolute value, but dependent on the film format, film resolution, film grain, subject distance,aperture, print size, viewing distance and perception of the photographer. The basis of the mathematical part of this subject is the Circle of Confusion, which is just that -as it is a variable. Adding to the general misunderstanding are lens makers defining the accuracy of their autofocus as "within 1/2 DOF" (for instance Canon),which is meaningless without defining the parameters. If one doesn't understand these issues, it is pretty hard to understand why a photograph does or does not turn out to expectation - which is exactly what beginning -or indeed all- photographers need to know.

 

[Stu W]

OK, I meant to say Summarit, not Summicron as was pointed out. My initial confusion was that why should an inefficent lens design, a slow lens, have the same dof at a given distance at a given aperture simply because of it's poor design. If I put a neutral density filter on a lens it wouldn't increase it's depth of field? But then we get back to the formula Aperture Diameter=Focal Length ÷ f Number. No mention of lanthium glass, multi coating, aspherical elements, computer designed. I did however check on the Leica lens specs, and yes you guys are right. The dof for a summilux or an elmar of the same focal length at the same distance and f stop are just about the same.
I always used the "f" number as a measurement of how much light was being delivered to the film, and assumed that a fast lens would need more "slowing down", hence a smaller aperture diameter than a slow lens to deliver equal amounts of light to the film. I thought the smaller aperture diameter should increase depth of field, thinking back to pinhole cameras. I'm convinced by the math but still confused by the concept. Stu

 

[jaapv]

A slower lens is not an inefficient design, it is simply different. There is no relationship between the larger max aperture and quality. Rather the reverse, really.