depth of field and film plane proximity

[WoolenMammoth]

wondering if anyone has any thoughts on the rear lens' proximity to the film plane and the depth of field it renders. Surely there is no mathematical explanation as DOF should be DOF however I feel like Im getting greater DOF with a F2 nikkor SLR than with a F2 Summicron M mount. Granted, I havent shot pictures of a measuring tape with both but got me thinking about the differences between the two systems and the large difference the lack of that reflex mirror can or can not account for. Has anyone done a proper study of this, I searched ye olde information superhighway with no results.

thanks

 

[manfromh]

Maybe you were closer to the subject with the SLR. As far as I know the lens-to-film distance should'nt affect DOF, but I cant be sure of it.

 

[Ronald M]

Debth of field depends on F stop and distance to subject. Nothing else.

 

[Finder]

Depth of field is related to the numeric aperture (angular size of the entrance pupil) of the system. I am not sure if it is possible to have different entrance pupil sizes while maintaining focal length and exit pupil angular size. Naturally, the lens designs are not going to be the same, but I am not sure of the impact, if any, on DOF.

It could be related to focus errors/precision of the two cameras. Another issue would be contrast as "sharpness" is related to that. Perhaps the lens contrast is not the same and giving the "feeling" you have.

You would need to test the lenses. If the difference is real, then you need to figure out why.

 

[ferider]

Debth of field depends on F stop and distance to subject. Nothing else.

Incorrect.

1) it does depend on format and enlargement.
2) it is a qualitative measure depending on the size of the COC, created on an enlarged print, looked at from a given distance.
2) Just like the thread starter, I have observed different 50mm lenses creating different sized COC, at the same effective aperture (i.e. including transmission loss). Anybody who owns a Canon 50/1.2 and a CV 50/1.5 should compare the CV wide open to the Canon at f1.5-1.6. You'll see (information about the Canon's transmission loss is available on Dante Stella's web site).

Roland.

 

[MichaelM7]

Debth of field depends on F stop and distance to subject. Nothing else.

Hi Ronald,

this is not correct. Depth of field depends on numerical aperture (a number related to the F stop number), the focal length of the lens, the focusing distance the lens is set to and the radius of the so called "circle of confusion", a sensor / film dependant constant that is related to the resolution of the film.

A very in depth and easy to understand introduction into these concepts have been given in the Zeiss Camera Lens News, Issues 1 to 4, which are available online in the Camera Lens News Archives at

http://www.zeiss.com/C12567A8003B58B9/Contents-Frame/15C75F926592E5C1C1256CED0054968D

That said, it could mean that the original poster used different kinds of film on SLR and RF camera which result in different circle of confusions. Films with higher resolution (low ASA numbers) can resolve higher frequencies (black / white crossings) that films with high speeds.

Also, the resolution of the lens is also a factor to consider: if you have a very good lens, the distortion introduced by the lens will not infer with the depth of field, i.e. the limiting factor stays the film. If you have a very bad lens, the resolution of the film might be better than that of the lens. Of course, the depth of field will be influenced by this. All in all, it is really true, that the "experienced" depth of field depends on the actual lens / film combination.

Hope that helps.

Michael

 

[rxmd]

2) Just like the thread starter, I have observed different 50mm lenses creating different sized COC, at the same effective aperture (i.e. including transmission loss).

There are two possible reasons for this.

Firstly, what's relevant for DOF is not the T-stop (aperture including transmission losses), but the f-stop (without them) - the f-stop is really just the effective aperture as a function of physical size of the iris and the "effective entry pupil", i.e. of the placement of the iris in relation to the focal length of the lens and the placement of the lens elements. Transmission losses just make things a little darker, but don't influence the projected "circles of confusion". So if you have two lenses with identical f-stops, but different T-stops, and you stop down one a bit so that their T-stops match, you are photographing at different f-stops and will get different DOF results.

Secondly, in my impression it is quite difficult to observe the "circles of confusion" directly - they are more of a conceptual aid to describe sharpness in terms of the resolution limits of our sensory system. When looking at extreme enlargements, comparison is also difficult when comparing across lenses with different bokeh; it is highly dependent on the scene we are observing, as a lens which, for example, has a harsher, more contrasty rendering of highlights might visually leave the impression of smaller "smudgy discs" and hence more DOF than a Sonnar-type with a soft, less deliminated rendering. These subjective things also influence our perception of DOF, making it difficult to make exact statements.

With the usual caution about DOF (i.e. assuming we compare between prints at the same magnification etc. pp.), across lenses with the same f-stop and focal length and when photographing objects at the same distance, DOF should be identical, regardless of the placement of the individual lens elements.

 

[oftheherd]

Incorrect.


...

2) it is a qualitative measure depending on the size of the COC, created on an enlarged print, looked at from a given distance.

...

Roland.

It is approximately 1/3 in front of and 2/3 behind the plane of focus, which is where the actual focus is set to. It is subjective in that it is that area that is in acceptable focus. That may vary by individual. It has been in all the reading I have done, considered to be measured by the COC on the negative. If a given print is viewed at the "correct" distance, the apparent sharpness will always be the same, regardless of the degree of enlargement.

I think I have read that RF lenses are a different design that SLR lenses, having to do with the greater distance between the rear of the lens needed in SLR lenses due to the mirror. I don't know if that can account for any differences or not. Also there is the question of DOF, as in depth of focus. How much that might play into it I don't know.

Of course I think it is all a moot point if you are getting acceptable photos based on your experience with a given lens. And then just becomes a curiosity, which seems to be how the original poster looked at it.

If I am wrong, have at me. I no longer have most of the books I used to have for reference, and my memory may be quite faulty.

 

[rxmd]

Hi Michael,

Depth of field depends on numerical aperture (a number related to the F stop number), the focal length of the lens, the focusing distance the lens is set to and the radius of the so called "circle of confusion", a sensor / film dependant constant that is related to the resolution of the film.

this is also not correct, with respect to the radius of the circle of confusion. The lens projects its image on any medium; the properties of the projection are independent of whether there is a sensor, a film or nothing behind it. The "circle of confusion" is just a term to describe the acceptable level of blurriness in the negative up to which a picture is perceived as being sharp. It's a way of describing the way our visual system works. Incidentally, this is also what the Zeiss paper #1 says that you quoted (http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN01e/$File/CLN1.pdf):

A certain amount of blur is supposed to be tolerable. According to international standards the degree of blur tolerable is defined as 1/1000th of the camera format diagonal, as the normally satisfactory value. With 35 mm format and its 43 mm diagonal only 1/1500th is deemed tolerable, resulting in
43 mm/1500 » 0.030 mm = 30 μm of blur.

This value of 0.03 mm as acceptable circle of confusion is, of course, set completely arbitrary. It is not a constant that corresponds to anything in the real world (especially not to properties of the film, sensor etc.); it is just a convention that refers to the visual impression of sharpness in the average viewer when viewing prints from a distance which is roughly correlated to the print's size. (If you blow up a picture to the size of a billboard and then scrutinise it from 10cm away, you will notice 0.03 mm "acceptable circle of confusion" is actually quite unacceptable under these circumstances; DOF will be a lot thinner.)

DOF, therefore, depends very much on the enlargement of images and the relative values the size and the distance of the print we look at. This is, for example, why Soviet lenses tend to have generous DOF markings; because in the Soviet Union prints tended to be smaller, but still viewed at arm's length, they could be a little more generous because in the smaller print things still looked sharp.

Philipp