DOF Question

[ECohen]

Advertisement (gone after registration)

DOF on three large cubes (modern art, for arguments' sake)  will exhibit a very different DOF from a leafy forest. So will a grey misty scene do from a sunny beach scene.  (to take the it into the hypothetical extreme: an uniformly  grey image will have no structure, no contrast, but an infinite DOF)  It is, in the end, all about the impression of sharpness in the eye of the beholder.

 

Interesting. I've always used a tape measure or rangefinder.  Measure what I want in and where I want the DOF  to fall. On less critical subject i guesstimate using the same principles...... I never thought of DOF and anything but  measurements of actual distance....using the numbers marked on a lens....a mathematical formula

Thank You for something new and interesting to think about.  ......."impression of sharpness"

 

Not to change the subject.....

Its what has held me back from playing with a 50 Noctilux  f/.95 at 3' you've got only 7"

If you look at in terms of the "impression of sharpness" and not an actual number ....7"

It puts that lens in a new light........and all lenses for that matter.

 

Thank You for the explanation.

Edited December 2, 2017 by ECohen

[pico]

It is, in the end, all about the impression of sharpness in the eye of the beholder.

 

That impression is embraced by the term 'accutance'.

DOF is a scientific metric that exists outside impressionism.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[jaapv]

Up to a point, yes, Pico:

 

 

 

"acutance" describes a subjective perception of sharpness that is related to the edge contrast of an image

 

 

However, it has to be taken one step further: Somebody has to look at the print - from a given distance- and decide which parts are sharp and which are unsharp - and that will depend on the subject, the rendering, the surface structure of the print and personal preference. Even on the light that the print is viewed under.

[david strachan]

That's such an important consideration. Be it an oil or a photograph.

 

I have some oil paintings which look better in a "darker"  hang.

 

It really makes such a huge difference to one's hung pictures...light is everything.

 

...

Edited December 2, 2017 by david strachan

[david strachan]

Ooops, a f'up post.

 

Talking about exhibiting prints..

 

...

[pop]

Doesn’t an ND filter affect the contrast?

No, it does not.

 

An ND filter passes only a fraction of the light - say, one fourth. Contrast is the ratio by which two different patches in the image differ. 

 

When one spot in the unfiltered image is three times as bright as another spot, it will still be three times as bright if both are darkened by 75%.

[jaapv]

Advertisement (gone after registration)

 

Interesting. I've always used a tape measure or rangefinder.  Measure what I want in and where I want the DOF  to fall. On less critical subject i guesstimate using the same principles...... I never thought of DOF and anything but  measurements of actual distance....using the numbers marked on a lens....a mathematical formula

Thank You for something new and interesting to think about.  ......."impression of sharpness"

 

Not to change the subject.....

Its what has held me back from playing with a 50 Noctilux  f/.95 at 3' you've got only 7"

If you look at in terms of the "impression of sharpness" and not an actual number ....7"

It puts that lens in a new light........and all lenses for that matter.

 

Thank You for the explanation.

 

Thanks. It has always puzzled me that photographers take the mathematical approach (which is useful as an indication) and treat it as if it were an absolute value, without even taking the intended size and use of the final print into consideration.

Visualisation is essential in images that depend on the use of DOF. In that sense the markings on lenses are only a crutch and completely useless in many cases. They date from a time that everybody used fairly thick film and enlarged to 6x9 cm.

[pgk]

Somebody has to look at the print - from a given distance- and decide which parts are sharp and which are unsharp - and that will depend on the subject, the rendering, the surface structure of the print and personal preference. Even on the light that the print is viewed under.

 

You are talking about a subjective impression of sharpness not a definition of Depth of field. Depth of field for a given set of parameters is a mathematical calculation because the whole concept is based on a mathematical hypothesis. In point of fact its not quite as absolute as we might all like to think but for all practical intents and purposes it is. Change the shape of the aperture blade though and there will be subtle shifts in the way a lens images, and lens design can be adjusted to shape oof areas (bokeh) and thus again, very subtly change the image transitions. DOF calculations are a fairly accurate and practical way of determining what will appear to be 'sharp' within a photograph and what will not - given that you need to define all the parameters required for the equation; including viewing requirements (which is what many people don't seem to take into account).

 

But to go back to the question, no, the use of any flat filter including an ND should not shift the depth of field perceptibly.

Edited December 2, 2017 by pgk

[Exodies]

No, it does not.

 

An ND filter passes only a fraction of the light - say, one fourth. Contrast is the ratio by which two different patches in the image differ. 

 

When one spot in the unfiltered image is three times as bright as another spot, it will still be three times as bright if both are darkened by 75%.

Except the eye does not have a linear response.

[pgk]

Except the eye does not have a linear response.

 

Which is only relevant if you put an ND (sunglasses) in front of your eye. An ND filter on a camera is supposed to do what it is designed to do - reduces the light hitting the sensor by a given factor. That said this isn't entirely true and all ND filters are not the same as their degree of light transmission can vary across the visible spectrum so they can introduce variance in response and thus colour shifts - some relevant info here: https://schneiderkreuznach.com/en/photo-optics/b-w-filters/filtertypes/uv-clear-1/nd-100-series. tried a few and found that they all seem to tend to create colour shifting to some degree, some only marginally but other substantially.

Edited December 2, 2017 by pgk

[jaapv]

You are talking about a subjective impression of sharpness not a definition of Depth of field. Depth of field for a given set of parameters is a mathematical calculation because the whole concept is based on a mathematical hypothesis. In point of fact its not quite as absolute as we might all like to think but for all practical intents and purposes it is. Change the shape of the aperture blade though and there will be subtle shifts in the way a lens images, and lens design can be adjusted to shape oof areas (bokeh) and thus again, very subtly change the image transitions. DOF calculations are a fairly accurate and practical way of determining what will appear to be 'sharp' within a photograph and what will not - given that you need to define all the parameters required for the equation; including viewing requirements (which is what many people don't seem to take into account).

 

But to go back to the question, no, the use of any flat filter including an ND should not shift the depth of field perceptibly.

I think that the subjective impression is the only thing that matters. The precise calculation -is it that precise? there are a number of different formula to calculate DOF ranging from simple to complicated- is only an aid to obtain the final result the photographer wants to present to the viewer.

 

Mathematical calculation that is as subjective as any such a calculation in biology or medicine as it is based on a value that is no more than a subjective premise, i.e. the circle of confusion. That is the average size of the spot that eye can resolve and reproduce in order for the brain to interpret it as a point. As every eye and brain -and spot! (contrast, edge sharpness?)-  is different that is as subjective as it gets. Other disciplines such as biology, medicine and psychology have long since recognized the weakness of such calculations and will at the very least attach a statistical analysis to the figure obtained.

To which may be added that the assumption of the spot was made on 1920ies film, with a thick emulsion that had far more aberrations like diffusion and refraction than present-day ones and on a 6x9 cm print ar 30 cm. And is rendered far more precise on a sensor, which is a flat plane.

 

The only conclusion can be that DOF cannot be caught in calculations that may be precise in themselves, but only approximate the phenomenon as it is seen by the people that look at our photographs.

By all means, use the DOF scales and DOF-masters to give a rough indication but be aware that they cannot replace the individual optical illusion that DOF is.

[jaapv]

Except the eye does not have a linear response.

But the eye has not come into play yet. We are still talking about recording the image on the sensor. The amount of light falling on the sensor will be reduced, but time of recording -shutter speed- will compensate for that, keeping the exposure in the same range as far as the eye is concerned.

But you do introduce another aspect of DOF - it will be different on a bright image as opposed to a dark one.

[pgk]

I think that the subjective impression is the only thing that matters.

 

Its not. Think of all the scientific applications to which photography is put. Yes, even if the calculation is subject to variables, all of which need to be defined, there are times when it is essential to determine whether detail is relevant or not and this can come down to calculated figures which can indicate whether the detail is spurious or viable (and of course as you say statistical analysis may now be added into the equation too - I remember seeing some image handling forensic software that used statistical analysis to amend digital files). Its also of use in underwater photography when using a dome port because the virtual image is spherical so dof needs to cover the closest part of the subject matter - this works in reverse in as much as its important not to have essential corner detail too close because it cannot be imaged accurately. At the end of the day though much photography does not require precise depth of field and that which does starts to drift into the realm of specialists, but that does not mean that dof calculations are irrelevant but suggests that they are relevant to those who actually need them. In 'normal' photography the most relevant use of dof calculations should be hyperfocal distance settings but these have become less appreciated it seems to me with photographers relying less on them than actually shifting the focus point to the bit they most want to be in focus (infinity is now infinity ).

[jaapv]

Those scientific applications mostly rely a digital interpretation of the data. That will take out the biological factor, i.e. the human observer and thus the subjectivity.

I don't think we are talking about scientific and technical applications of photographic techniques on this forum. I think I omitted the word "here" in that sentence .

[pgk]

I don't think we are talking about scientific and technical applications of photographic techniques on this forum.

 

But we all too often muddle up detailed theory and practical photography - some is relevant but most isn't .

[jaapv]

Exactly - That is why we should stick to sharp-is-sharp and forget about DOF.

[pgk]

Exactly - That is why we should stick to sharp-is-sharp and forget about DOF.

 

If only .

[ECohen]

Thanks. It has always puzzled me that photographers take the mathematical approach (which is useful as an indication) and treat it as if it were an absolute value, without even taking the intended size and use of the final print into consideration.

Visualisation is essential in images that depend on the use of DOF. In that sense the markings on lenses are only a crutch and completely useless in many cases. They date from a time that everybody used fairly thick film and enlarged to 6x9 cm.

 

My excuse ...I was a Studio Advertising photographer.  I lived day in and day out by those scales and measurements . And those scales are darn accurate....not just because of the thickness of film. Shoot a group of people you cant rely on perceived sharpness to get the back row in focus ......start with math.

 

I can certainly see that you are correct. With DOF the math should  only part of the "equation"

There so many choices to making a photograph many we make unconsciously, from experience.

 

I always rely on "how does it feel"  when looking at or creating, Art/Photography. Which I guess is a kin to "perceived sharpness"

 

I have only been on this Form a few years ...this really is a great form of ideas, experience and insight .

Edited December 2, 2017 by ECohen

[mikemgb]

Exactly - That is why we should stick to sharp-is-sharp and forget about DOF.

 

Try saying that to the Noctilux crowd. 

[Advertisement]

Hallo,

Look here: Leica M

[ECohen]

Try saying that to the Noctilux crowd. 

 

Serious question ....Isn't perceived sharpness what Noctilux is all about?

The reality is at 3' f/.95  DOF is 7" ........the rest is "perceived sharpness".