A proposal: a constant measure for depth of field

[rcusick]

Advertisement (gone after registration)

I was pondering how to capture a constant measure of depth of field.... Why not something akin to a "stop of light?"  I'll call it the "cu stop" after my last name.  

Like a stop of light - where the constant in the equation is the exposure to light - the cu stop constant would be a depth of field.  

Exposure has three variables - shutter speed, aperture and iso.  A cu stop would have thee variables as well - aperture, sensor size and focal length.   

Thoughts?  This would help equalize for different equipment setups - apsc vs full frame vs large format, aperture and type of lens.  

Edited February 15, 2022 by rcusick

[Advertisement]

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

[Aryel]

1 hour ago, rcusick said:

Exposure has three variables - shutter speed, aperture and iso.  A cu stop would have thee variables as well - aperture, sensor size and focal length.   

 

I think a cu stop won’t work because you are missing one major variable for depth of field: distance to subject. 
 

Edit: I also don’t really understand what would be the benefit. You have it marked on the lens. 

Edited February 15, 2022 by Aryel

[adan]

Leaves out the most important factor - ENLARGEMENT. Or MAGNIFICATION from the original image size. Including magnification based on how close we get to the picture

With DoF, that is really where the rubber meets the road. Because DoF is defined as the range or "field" of things/distances in a picture that "look" sharp, under specific conditions.

(Some will talk about "circles of confusion", but that is really just the same thing - how big must a blur be before we recognize it as a blur and no longer a sharp point?)

Traditionally, the conditions for engraving 35mm depth-of-field scales on lenses is - viewing an 8"x12" print (8x enlargement from a piece of film or a sensor 1"x 1.5", at a distance = to the diagonal of the print - about 14". For metric purists - a 20cm x 30cm print enlarged 8 times from a piece of film 24mm x 36mm and viewed from ~ 36 cm).

Pop quiz, what "looks sharp" in each of the attached versions of one image ? Where is the sharpest plane (where was the lens actually focused)?

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited February 15, 2022 by adan

[adan]

Does your opinion of what was "sharp" change with this magnification? (click for maximum image size)

(BTW, this image was made with a 21mm Elmarit at what the EXIF claims was f/5.6)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited February 15, 2022 by adan

[Aryel]

1 minute ago, adan said:

Does your opinion of what was sharp change with this magnification?

Point well taken. Thanks a lot!

[rcusick]

All good comments.  Ok should have started with my goal - measure in camera (ie at point of image creation) the depth of field across different focal lengths and sensor sizes.  I watched a video recently pointing out that medium format is a misnomer - as there is no standard defintion of sensor size.  The video then went on to show similar images taken with different sensor sizes, focal lengths and apertures could be used to achieve the same effect.  It would be great to have this standardized so one could choose apsc vs full frame vs large format equalizing with lens and aperture.  The debate should not be apsc vs full frame - it should be around the target depth of field / creative objective. 

Edited February 15, 2022 by rcusick

[adan]

Advertisement (gone after registration)

10 minutes ago, rcusick said:

Ok should have started with my goal - measure in camera (ie at point of image creation) the depth of field across different focal lengths and sensor sizes.

I think that is your mistaken point - DOF is an undefined term "at the point of image creation." No such thing. A nonsense phrase - unless and until the picture is removed from the "point of creation" and viewed elsewhere. On a computer screen or in a print at a defined final size.

Write the following 100 times on the blackboard ( ) - "Depth of Field is never an absolute that can be measured in and of itself - it is all about the appearance of sharpness in the final image size, ready for viewing. It has no meaning until then."

The term that comes closest to what you are describing (but still not very) "at the point of creation" is Depth of Focus. How much the sensor or film can be out of tolerance and still appear to capture correct focus.

And you still need to pick out of thin air some particular circle of confusion (or "size of blur") to put into the equation.

https://www.masterclass.com/articles/depth-of-focus-vs-depth-of-field

Basically, you are trying to come up with a formula for how fast several given cars can go around a racetrack - without knowing how long the racetrack is.

[jaapv]

+1!!  The definition of DOF being: the zone of acceptable unsharpness in a photographic print.
Subject matter and contrast are. important considerations as well. Which means that one does not know what car is driven and how fast the driver cares to go...
My personal definition is: DOF is in the eye of the beholder and mathematical approaches are no more than guidelines.

Another thing missing: DOF is a gradient; an image does not snap into focus at a certain point and stays there to a second point; the appearance of sharpness starts at a variable distance, builds up to the plane of focus and drops off again.

[IkarusJohn]

Oh dear.  Here we go again.

[jaapv]

It is only the 113th thread and counting, John  I did not even mention that it varies by lens design

[adan]

There is already a purported rule-of-thumb formula for comparative amount of relative blur-circle-size in the image plane (not "DoF") across different film/sensor formats/sizes.

(Never tested it myself)

Multiply the aperture by the crop factor, as well as the focal length.

E.G. a 50mm f/1.4 on APS-C (crop factor 1.5) should have the same relative amount of blurring and sharpness as a 75mm f/2.1 (= f/1.4 x 1.5) on 24x36.

And that can be extended to 30x45 (e.g. Leica S - crop factor 0.8), where the 75 f/2.5 Summarit-S should have equivalent imaging to a 60mm f/2.0 on 24x36.

Or even to real MF or large-format films - e.g. a 90mm f/3.5 on a 6x9 medium-format camera (crop factor 0.43 compared to a 35mm frame) should produce a similar overall amount of focus and defocus to a 39mm f/1.5 on the Barnack format (users of Voigtländer 40mm f/1.4s take note!)

Make prints the same size from those pairings and then the DoF in the prints should be pretty comparable. Or so the theory says.

Gets a bit trickier when the formats are different shapes or proportions (e.g. 33x44, or 6x7, or 6x6, or 2:3, etc.) - one has to determine the "crop factor" from the diagonals of the images (Pythagoras and all that). And then make prints with identical diagonal dimensions.

Edited February 15, 2022 by adan

[Gobert]

Can anyone supply me with the formula for calculating DOF? It might be helpful in the field.

Edited February 15, 2022 by Gobert

[pgk]

17 minutes ago, Gobert said:

Can anyone supply me with the formula for calculating DOF? It might be helpful in the field.

Try: https://en.wikipedia.org/wiki/Depth_of_field where there are equatons the main one of which is stated to be an approximation. It probably won't be very helpful anyway because to be of use it needs the output requirements to be pre-defined and this affects the figure for the circle of confsion which is itself tricky to be precise about. Depth of field is, in effect, a variable with a number of factors influencing it. In many ways it is easier to think of images as having a plain of maximum sharpness/clarity/efinition and accept that the aperture used affects the steepness of the transition to softness (out of focus areas).

Photography is, in the final analysis, a practical subject so the best way todetermine 'acceptable' depth of field is by getting to know your equipment and the effect of using various apertures for a given lens/sensor/output. In other words experience will help utilise the optimal aperture for the photograph being taken.

Thinking of my own photography and I would say that I very rarely ever use depth of field scales and mostly think of where I want the plane of maximum 'sharpness' to lie and whish aperture will achieve this with the most relevant transition to 'unsharpness'.

[jankap]

There are cameras, that allow to change the DOF in a picture. One picture taken, change in post only!

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Edited February 15, 2022 by jankap

[Jul]

13 hours ago, jaapv said:

I did not even mention that it varies by lens design

If I remember well, Adan used 75mm Summilux-M and Apo-Summicron somewhere on the forum to demonstrate this fact with pictures. 

[jaapv]

7 hours ago, jankap said:

There are cameras, that allow to change the DOF in a picture. One picture taken, change in post only!

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

You refer to post-focus and focus stacking, I suppose ?

[jankap]

10 hours ago, jaapv said:

You refer to post-focus and focus stacking, I suppose ?

Post-focus and post f/stop yes. But no stacking. It is one (1) photo! The first picture is set at f/8 and the second at f/1.0 in post. 

Edited February 16, 2022 by jankap

[jaapv]

Focus stacking is part of the post focus module. 

[Advertisement]

Hallo,

Look here: Leica M

[Aryel]

1 hour ago, jankap said:

Post-focus and post f/stop yes. But no stacking. It is one (1) photo! The first picture is set at f/8 and the second at f/1.0 in post. 

It is one photo from the user’s perspective. Are you sure that it did not take multiple ones at the back to achieve this?

edit: actually, the one that was shown look pretty poor (especially some of the grass on the left). So maybe it took only one picture… 

edit2: I was wrong, I did not consider sensors that can get spatial information. Thanks @jankap for pointing this out  🙂

 

Edited February 16, 2022 by Aryel

[farnz]

4 hours ago, Aryel said:

It is one photo from the user’s perspective. Are you sure that it did not take multiple ones at the back to achieve this?

I think it's a picture from a novel Lytro camera, which is a "Light-field" camera that captures all the light rays in the field (not just one plane) at one instant and uses mathematical formulae to subtract light rays afterwards to provide the image at different apertures.  (That's my unclear explanation anyway. )

Pete.