Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

What are the best lenses when shooting fully stopped down?
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Wed Nov 23, 2011 3:32 am    Post subject: Reply with quote
buerokratiehasser wrote:
Well, no, that's not how it works, it is a hard physical limit and applies to any kind of aperture (at least with standard optical systems without weird quantum effects). It doesn't change at f/16 even if you shell out $2000 for a f/0.9 lens.


I was thinking the other way around, that f/4 macro lens can be stopped farther than f/0.9 lens before aperture diffraction affects resolution.

What I think now is the macro lens has higher optical resolution, can therefore be stopped down farther before aperture affects the higher resolution.

Aperture effect on resolution also depends on sensor density.

A good resource is Cambridge In Color: LENS DIFFRACTION & PHOTOGRAPHY

AhamB wrote:
I agree in principle (visualopsin's explanation is incorrect I think) that it doesn't change how it works in theory, but there among different lenses there definitely are differences in how the images looks. I don't haven't paid attention to resolution measurements, but it seems that some lenses retain more sharpness at f/16+ apertures than others.
There are people who are convinced that any lens with the same specs (FL and aperture) will have exactly the same DOF, but there are significant differences there too. The Zeiss MP50/2 has deeper DOF than the Planar 50/1.4 at identical apertures, across the whole range, for instance.

Anyway, I say: just stop down your lens as much as you need to get the desired DOF, but not more than needed! Learn how your equipment behaves with different settings. It can be worthwhile trading a bit of DOF or corner sharpness for better central sharpness and microcontrast.


The DOF difference is mysterious! Is there a reference I can see please? Lens focal length changes with focus distance in some (all?) lenses -- different lenses having different focal lengths at the same focus distance could account for difference in DOF, no? But subject sizes wouldn't be equal. A mystery...

Good advice imho bears repeating! :

AhamB wrote:
just stop down your lens as much as you need to get the desired DOF, but not more than needed! Learn how your equipment behaves with different settings. It can be worthwhile trading a bit of DOF or corner sharpness for better central sharpness and microcontrast.


PostPosted: Wed Nov 23, 2011 7:49 am    Post subject: Reply with quote
Yes, 50 mm really being 52 mm can be suspected (this can even change with focusing)

DOF formulae are approximations, the rule of thumb that the relative DOF stays the same with focal length (i.e. focal length does not buy you anything) is just a rule of thumb. With real lenses there may be 10% difference, too small to optimize for that imho

Also real lenses are multi-element, spread out spatially, and behave only in approximation like .. approximation of a single lens. Maybe some lenses are actually worse. Or some lenses have really bad spherochromatism that blurs slightly-out of focus images more than you like, adding to the fuzziness (remember, DOF is in itself an approximation, the closer you look at the film, the more it shrinks to a razor-sharp plane)


This knowledge is from web, I have not calculated the lens formulas


PostPosted: Thu Nov 24, 2011 12:36 am    Post subject: Re: What are the best lenses when shooting fully stopped dow Reply with quote
Pii wrote:
Most of us should know by now what are the best lenses when shooting wide open. If not, just read the forum once or twice and you will know for sure. But do we know what are the best lenses when shooting fully stopped down? Or what is the maximum f stop and how good it is comparing to the lens sweet spot?

Hope you will help me with that Very Happy


After about f/8 (speaking of 35mm format lenses) diffraction sets in and becomes the limiting factor as far as resolution is concerned. The overall quality of the lens is still a factor, and this does not affect the contrast or other properties of the lens. In other words, better lenses will still show their virtues, but resolution will suffer.


PostPosted: Thu Nov 24, 2011 6:25 pm    Post subject: Reply with quote
I think that for diffraction we must consider even the focal lenght:
If I have a 20mm for example a diaphraghm f 22 has a hole of about less of one millimeter=great diffraction
If I have a 200mm the same aperture f 22 has a hole of about less of ten millimeters=low diffraction
I think that diffraction depend on the physically little hole which compares with the wavelength of light passing through it.
So if we use a large format we have few problems with diffraction, even with wide angles(that for 24x36 format are normally telephoto lenses).
Perhaps the F64 school of Ansel Adams and his group still teaches us something...
Am I correct?


PostPosted: Thu Nov 24, 2011 6:55 pm    Post subject: Reply with quote
lucialfa wrote:
I think that for diffraction we must consider even the focal lenght:
If I have a 20mm for example a diaphraghm f 22 has a hole of about less of one millimeter=great diffraction
If I have a 200mm the same aperture f 22 has a hole of about less of ten millimeters=low diffraction
I think that diffraction depend on the physically little hole which compares with the wavelength of light passing through it.
So if we use a large format we have few problems with diffraction, even with wide angles(that for 24x36 format are normally telephoto lenses).
Perhaps the F64 school of Ansel Adams and his group still teaches us something...
Am I correct?


To some extent, yes, larger-format lenses are less affected. But within a given format (35mm for instance) the shorter the focal length the greater the effect of diffraction. None the less, a better lens will still be a better lens than another, at every aperture, even allowing for the effect of diffraction.


PostPosted: Fri Nov 25, 2011 9:25 am    Post subject: Reply with quote
lucialfa wrote:
I think that for diffraction we must consider even the focal lenght:
If I have a 20mm for example a diaphraghm f 22 has a hole of about less of one millimeter=great diffraction
If I have a 200mm the same aperture f 22 has a hole of about less of ten millimeters=low diffraction
I think that diffraction depend on the physically little hole which compares with the wavelength of light passing through it.
So if we use a large format we have few problems with diffraction, even with wide angles(that for 24x36 format are normally telephoto lenses).
Perhaps the F64 school of Ansel Adams and his group still teaches us something...
Am I correct?


No, this is not correct. Diffraction does indeed depend on the focal length as well as the absolute size of the aperture, but the f-number already includes both (after all, the “f” in it means focal length). However, one must remember that it is not the f-number marked on the lens that matters, but the actual effective f-number (which varies with focus/magnification, especially in macro).

As for larger formats having fewer problems with diffraction, that is indeed true but for an entirely different reason: to reach a given output size, larger formats need to be enlarged much less. (And, again, this is also why larger formats are otherwise less sensitive to lens quality, and also why pixel peeping high resolution digital images is quite silly because viewing at 100% crop cancels out all advantage from the sensor size or higher megapixel count, while both exist at practical viewing sizes.)


PostPosted: Fri Nov 25, 2011 10:54 am    Post subject: Reply with quote
The best way to consider this is with film. Just imagine swapping the film with another format and.. the image projected onto the film does not change (it is not practical to step-up in practice, since real lenses are optimized to illumate their formats and not much more).

From there, same f-number, same resolution on the film (reread the thread why this may not buy you much)


As for the effective aperture with macro lenses, I like to learn something new - is there any reference for that? I already knew that at 1:1 things get pretty dark, but I thought that was the T number (transmission) not the F number.


PostPosted: Fri Nov 25, 2011 11:39 am    Post subject: Reply with quote
buerokratiehasser wrote:

As for the effective aperture with macro lenses, I like to learn something new - is there any reference for that? I already knew that at 1:1 things get pretty dark, but I thought that was the T number (transmission) not the F number.


As far as I can see, the t-number refers to the transmission of lens elements, which obviously won't change with extension.

However, the effective f-number changes with extension, affecting not only exposure but diffraction as well. Usually the formula seen for this “bellows factor” is:
Code:
effective_f = marked_f * (1 + magnification)


That is, the marked f-number would only apply at infinity focus and at 1:1 magnification two stops are lost. (This makes sense if you think about the increased “focal length” with extension, e.g. a 50mm lens would have 50mm extension at 1:1 magnification so the effective f-number ends up being the same as that of a 50+50=100mm lens with the same size of physical aperture.)

(I think this formula only applies to magnification from “true” extension, macro lenses might change focal length while focusing etc… However, I don't claim to be particularly knowledgeable of high magnification macro photography.)

Last edited by Arkku on Fri Nov 25, 2011 11:16 pm; edited 1 time in total


PostPosted: Fri Nov 25, 2011 1:45 pm    Post subject: Reply with quote
A definite reference to read up would be nice,
but you are probably right.

Bellows/TC = holding the lens farther from the film/sensor. the image gets formed and then magnified because the film is farther away. It's like using a crappy m42 adaptor that is a bit too thick.

This means any diffraction effects (on the "standard" image plane) get magnified, too. So we get the worst of two worlds. I'll guess I will macro at f5,6 now


PostPosted: Fri Nov 25, 2011 1:53 pm    Post subject: Reply with quote
Diffraction patterns are also really sensitive to the shape of the hole.
A perfectly round diaphragm will leave less intense diffractogram than an hexagonal one for instance.
Round : circles of decreasing intensity as they are far from the center.
Polygon : spots in line with the polygon corners : light is more concentrated on those spots.

EDIT : i forgot the conclusion :

Choose a lens with a lot of blades. I shoot often with my jupiter-3 closed at 16, still sharp !


PostPosted: Fri Nov 25, 2011 7:34 pm    Post subject: Reply with quote
Highlight bokeh gets shaped by aperture opening, yes.

Here I think we refer to diffraction of aperture affecting resolution:

As aperture size decreases, the resolution increases to the maximum resolving power of the lens system. Smaller aperture settings degrade the achieved maximum resolution.

Film and sensor resolution limit may be reached before lens system resolution limit -- the lens system can out-resolve the film/sensor -- thus increasing the size of the smallest aperture that can be used before resolution is affected. The lens may resolve best at f/16, but film or sensor may be able to resolve lens output up to f/8 -- smaller apertures degrade film/sensor resolution.

Therefore, a lens with higher resolution can be stopped down farther than a lens with less resolution, before smaller aperture size begins to degrade lens resolution, provided film/sensor can resolve as well as the lens. If film/sensor resolution capability is less than lens system, aperture diffraction effects begin at larger aperture than setting for best lens resolution.


PostPosted: Fri Nov 25, 2011 11:31 pm    Post subject: Reply with quote
visualopsins wrote:

Therefore, a lens with higher resolution can be stopped down farther than a lens with less resolution, before smaller aperture size begins to degrade lens resolution, provided film/sensor can resolve as well as the lens. If film/sensor resolution capability is less than lens system, aperture diffraction effects begin at larger aperture than setting for best lens resolution.


Maybe I'm misunderstanding your wording, but it seems as though you are saying exactly the opposite of truth:

Let's say a hypothetical lens is really good and is able to render up to "500 megapixels" worth of detail on FF, even from maximum aperture of, say, f/2. This means that the if the lens is stopped down even to f/2.8, it is already limited by diffraction from being able to draw the 500 megapixels worth of details. Meanwhile another lens is only able to render “15 megapixels” worth of detail on FF at its best (and probably less so wide open), therefore it won't be diffraction limited until past f/11 (assuming it has reached this 15 mpix performance by them). So, the higher the resolution the lens is capable of the, less it can be stopped down before aperture size begins to limit its resolution.

(And the best possible lens would always be limited by diffraction.)

Completely independent of this, there's the maximum resolution that a sensor or film can record. If it's less than that rendered by the lens at a given aperture (including effects of diffraction, which happen independently of the sensor/film), then resolution of the whole image is limited by the sensor/film (not lens or diffraction). So again, the less resolution your sensor/film has, the smaller the aperture where diffraction starts to limit the photo's resolution.

(And, again, the best possible film/camera would always be limited by lens quality or diffraction. So the best system overall would be always be limited only by diffraction.)


PostPosted: Sat Nov 26, 2011 1:25 am    Post subject: Reply with quote
Arkku wrote:
visualopsins wrote:

Therefore, a lens with higher resolution can be stopped down farther than a lens with less resolution, before smaller aperture size begins to degrade lens resolution, provided film/sensor can resolve as well as the lens. If film/sensor resolution capability is less than lens system, aperture diffraction effects begin at larger aperture than setting for best lens resolution.


Maybe I'm misunderstanding your wording, but it seems as though you are saying exactly the opposite of truth:

Let's say a hypothetical lens is really good and is able to render up to "500 megapixels" worth of detail on FF, even from maximum aperture of, say, f/2. This means that the if the lens is stopped down even to f/2.8, it is already limited by diffraction from being able to draw the 500 megapixels worth of details. Meanwhile another lens is only able to render “15 megapixels” worth of detail on FF at its best (and probably less so wide open), therefore it won't be diffraction limited until past f/11 (assuming it has reached this 15 mpix performance by them). So, the higher the resolution the lens is capable of the, less it can be stopped down before aperture size begins to limit its resolution.

(And the best possible lens would always be limited by diffraction.)


Ah, I see, thanks, if the airy disk is already as small as it can get, stopping down isn't going to make it smaller, it will degrade. I don't think there are many f/2 500mp lenses with general photography focal lengths and/or image circles! Wink

Arkku wrote:
Completely independent of this, there's the maximum resolution that a sensor or film can record. If it's less than that rendered by the lens at a given aperture (including effects of diffraction, which happen independently of the sensor/film), then resolution of the whole image is limited by the sensor/film (not lens or diffraction). So again, the less resolution your sensor/film has, the smaller the aperture where diffraction starts to limit the photo's resolution.

(And, again, the best possible film/camera would always be limited by lens quality or diffraction. So the best system overall would be always be limited only by diffraction.)


While closing the aperture, once lens resolution matches film/sensor resolution, stopping down farther does not increase system resolution -- the system resolution limiting aperture is larger (smaller f/#) than lens resolution limiting aperture. Once airy discs get smaller than grain/pixels, resolution is lost.


PostPosted: Sat Nov 26, 2011 12:58 pm    Post subject: Reply with quote
visualopsins wrote:

Ah, I see, thanks, if the airy disk is already as small as it can get, stopping down isn't going to make it smaller, it will degrade.


The Airy disk gets larger as you stop down, that is why it limits resolution, e.g., by being larger than the circle of confusion would be in the plane of focus, or by being larger than the sensor equivalent of a single pixel (which is actually not that easy to define in CFA sensors with single-colour photosites and anti-aliasing filters, but I would say about 1.5× the diameter of a single photosite).


PostPosted: Wed Nov 30, 2011 12:15 am    Post subject: Reply with quote
Arkku wrote:
visualopsins wrote:

Ah, I see, thanks, if the airy disk is already as small as it can get, stopping down isn't going to make it smaller, it will degrade.


The Airy disk gets larger as you stop down, that is why it limits resolution, e.g., by being larger than the circle of confusion would be in the plane of focus, or by being larger than the sensor equivalent of a single pixel (which is actually not that easy to define in CFA sensors with single-colour photosites and anti-aliasing filters, but I would say about 1.5× the diameter of a single photosite).


Embarassed I knew that! (basic physics) Laughing



From Wikipedia: DOF and Diffraction

Quote:
As a lens is stopped down, the defocus blur at the DOF limits decreases but diffraction blur increases. The presence of these two opposing factors implies a point at which the combined blur spot is minimized (Gibson 1975, 64); at that point, the f-number is optimal for image sharpness.




Well now if part of my mind isn't flipped, then the sharpest lens fully stopped down would be the one sharpest wide open.

Intuitively I know it is a long focal length lens, probably a macro lens, with front element diameter optimal for greatest grinding accuracy to give best resolution. Something Voigtländer-ish Laughing


PostPosted: Thu Dec 01, 2011 12:59 pm    Post subject: Reply with quote
visualopsins wrote:

Well now if part of my mind isn't flipped, then the sharpest lens fully stopped down would be the one sharpest wide open.


With two lenses both limited by diffraction fully stopped down, there would not be a difference even if one was ridiculously soft wide open and one very sharp (and wide open could mean different things, one could be an f/0.95 lens and the other a f/3.5). I guess this was the original point leading to the whole diffraction discussion; stopped down beyond a certain point it doesn't matter which lens is used (considering only resolution).

However, if we change the poorly-defined “fully stopped down” to any given aperture, then of course you are right that the lens that starts off with a higher resolution is more likely to have reached the diffraction limit at a given stop.


visualopsins wrote:

Intuitively I know it is a long focal length lens, probably a macro lens, with front element diameter optimal for greatest grinding accuracy to give best resolution. Something Voigtländer-ish


Macro lenses are always a good bet, but I'm not that sure I follow the reasoning about long focal lengths? In my experience most long focal lengths end up with more C/A etc unless it is indeed one of Voigtländer's APO lenses. (Meanwhile several short lenses are surprisingly good, e.g., the Samyang 14mm f/2.8 has very high resolution from wide open, if I recall correctly actually better at f/2.8 than Zeiss 50mm f/2 makro-planar at f/2.8.)
Post new topic   Reply to topic
Home   Manual Focus Lenses
Goto page Previous  1, 2, 3
Page 3 of 3