24-90 vs 90 apo

[BernardC]

Advertisement (gone after registration)

13 hours ago, Jared said:

All 75mm lenses of a given aperture on a given format have the exact same depth of field.

Sorry, no.

Not only is depth of field unevenly distributed in front-of and behind the subject (as mentioned earlier), two lenses of the same aperture and focal length can have different depth of field.

Imagine a classical diagram of how a single-element lens works. You see red light coming into focus farther from the lens, blue rays converging closer, and green in the middle. let's say that green light is focused at 10m. In that case, red light may be focused at 13m, and blue light at 9m. A typical depth-of-field scale might tell you that the scene will be "acceptably sharp" between 9.5 and 11m, however we know that points at 9m and 13m are also sharp, at least for one colour!

Now take a highly-corrected lens where the three colours focus nearly at the same point (within mm, not m). In that case, the parts of the object that are exactly at 10m will be much sharper (because all three colours are sharp there, not just one), and the points outside of the point of focus will be less sharp (no colours in focus). The depth-of-field is much less deep.

This example is simplified because it only accounts for chromatic aberration. Spherical aberration also affects the point of focus. The higher the level of correction, the more light rays will converge on the same image point, regardless of whether they travel through the optical axis, or through the periphery.

 

You don't have to take my word for it. Here's Kornelius Fleischer (then of Zeiss) explaining the same thing on photo.net 20 years ago (https://www.photo.net/discuss/threads/cfi-250-superachromat-vs-cfe-350-tele-superachromat.8109/)

Quote

Contrary to popular belief not all lenses with the same focal length produce the same depth of field at the same aperture! A typical example is the Superachromat 5,6/250: It produces extreme sharpness (250 linepairs per millimeter)at the plane of best focus. But slightly off (which may be caused by photographer's focusing error, film position error, curvature of film, humidity, registration error of mirror or focusing screen, magazine wear, to name just a few), the sharpness drops dramatically. It can drop even below the levels the sharpness of a Sonnar 5,6/250, or the one of a Tele-Tessar 4/250.

 

[Jared]

5 hours ago, BernardC said:

Sorry, no.

Not only is depth of field unevenly distributed in front-of and behind the subject (as mentioned earlier), two lenses of the same aperture and focal length can have different depth of field.

Imagine a classical diagram of how a single-element lens works. You see red light coming into focus farther from the lens, blue rays converging closer, and green in the middle. let's say that green light is focused at 10m. In that case, red light may be focused at 13m, and blue light at 9m. A typical depth-of-field scale might tell you that the scene will be "acceptably sharp" between 9.5 and 11m, however we know that points at 9m and 13m are also sharp, at least for one colour!

Now take a highly-corrected lens where the three colours focus nearly at the same point (within mm, not m). In that case, the parts of the object that are exactly at 10m will be much sharper (because all three colours are sharp there, not just one), and the points outside of the point of focus will be less sharp (no colours in focus). The depth-of-field is much less deep.

This example is simplified because it only accounts for chromatic aberration. Spherical aberration also affects the point of focus. The higher the level of correction, the more light rays will converge on the same image point, regardless of whether they travel through the optical axis, or through the periphery.

 

You don't have to take my word for it. Here's Kornelius Fleischer (then of Zeiss) explaining the same thing on photo.net 20 years ago (https://www.photo.net/discuss/threads/cfi-250-superachromat-vs-cfe-350-tele-superachromat.8109/)

 

In that link he describes the same affect that Peter Karbe refers to with the Summicron SL lenses, that is, the extremely high resolution and acutance at point of best focus creating more noticeable contrast with areas that are even slightly out of focus.  That is not really a change in depth of field.  It doesn’t change the size of the circle of confusion for a given amount of defocus. 

As to depth of field being unevenly distributed front-to-back, I acknowledged that in my initial post.

As to the contributions of chromatic aberration... It certainly changes what someone might consider “acceptably sharp” at a personal level, but no manufacturer I am aware of rates depth of field differently based on how well controlled longitudinal and lateral chromatic aberration are controlled in the lens.  Look up any of the mathematical formulae for depth of field.  They are not dependent on optical design.  Focal ratio, focal length, distance to subject, amount of magnification, and viewing distance are it.

[BernardC]

46 minutes ago, Jared said:

no manufacturer I am aware of rates depth of field differently based on how well controlled longitudinal and lateral chromatic aberration are controlled in the lens.

Zeiss is a manufacturer, see my example above. As you note, Leica also tells us that depth of field is more shallow for lenses that are more highly corrected.

I am familiar with depth-of-field tables, but those are rough approximations.

[Jared]

1 hour ago, BernardC said:

Zeiss is a manufacturer, see my example above. As you note, Leica also tells us that depth of field is more shallow for lenses that are more highly corrected.

I am familiar with depth-of-field tables, but those are rough approximations.

We may need to just agree to disagree.  Both Zeiss and Leica are, obviously, trying to sell lenses.  They both make excellent lenses.  But there are no magical optics.  There is no way to increase or decrease the size of the circle of confusion that results from a given amount of defocus short of things like apodization, playing with central obstructions, and the like. Ordinary refractive optics just don’t vary much in terms of depth of field.  There are some very subtle changes from things like amount of chromatic aberration and level of spherical Correction, as you correctly pointed out, but that’s about it.

As far as the depth of field tables... They are certainly arbitrary in that they are based on an A4 or 8x10 print size as seen from arm’s length with typical visual acuity.  One may be printing larger or cropping or viewing from closer.  One may just have higher expectations than were common when the DOF tables and scales were first generated. Still, I would not say they are approximations so much as I would say they are arbitrary.  

Edited March 3, 2020 by Jared

[Exodies]

The circle of confusion has an arbitrary radius. It is only a convention.