Summicron 28 ASPH bokeh - is this right?

[adan]

Advertisement (gone after registration)

On 8/22/2022 at 10:20 PM, gotium said:

with changing only the aperture (I did not re-focus) the *center* of the focal plane (which you see by choosing the midpoint between where it starts to go blurry on either side) moves all the way back to the 9-inch mark by f/2.8, and then even further back by f/4 and f/5.6.

It is normal for the depth of field range to extend farther behind the focused point than in front of it (or even "infinitely" more). Very-roughly speaking, about 1/3 of the DoF will be in front of the focus point, and 2/3rds behind the focus point.

Put more bluntly, the *center* between the blurry parts either side is NEVER where the lens was actually focused - ever, with any lens, at any aperture.

That is a conceptual error.

Look at the DoF scale itself (always recalling that DoF scales are only precise for a given print size and circle of confusion - but the proportions remain the same)

On my 28 Elmarit v.4

Lens focused at 1.5 meters and using f/16:

Predicted DoF extends to infinity behind 1.5 meters (i.e. an infinite distance behind the focus point).

But only to 0.8m in front of 1.5m

0.8m....1.5m (focus point)................................................................................................................................................................v^v^...............................∞

0.7m (1.5m-0.8m) does not equal 1.5m-plus-infinity

Or:

Using f/8, focused at 1m: The extent of the predicted DoF is from 0.8m (0.2m in front of the focus point) to 1.5m (0.5m behind the focus point.)

0.8m...(0.2m)....1m (focus point)....................(0.5m)...................1.5m

0.2m does not equal 0.5m. 0.5m = 0.2m x 2.5

Edited August 24, 2022 by adan

[gotium]

13 hours ago, adan said:

It is normal for the depth of field range to extend farther behind the focused point than in front of it (or even "infinitely" more). Very-roughly speaking, about 1/3 of the DoF will be in front of the focus point, and 2/3rds behind the focus point.

Put more bluntly, the *center* between the blurry parts either side is NEVER where the lens was actually focused - ever, with any lens, at any aperture.

That is a conceptual error.

 

Thank you for the input, well-explained and also of course matches real-world experience. I am mis-stating how to assess focus shift as the mid-point of what appears to be in focus.

Do you instead try to use the point that appears to be most in-focus? Or in which direction the edges of the depth-of-field move when aperture is adjusted (ie when stopping down I would still never expect the FRONT of the focal plane to move backward)?

[Advertisement]

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