Why using very small aperture for the long lens?

Pros usually recommend that for the maximum depth of filed, one should

use very small aperture. What about when using a long lens eg. 300 mm

for shooting landscape. Is this still true? When I loook at the depth

of filed chart for the long lens, aperture seems to effect very little

on DOF because long lens has very limit DOF.

 

So my question is that is there any logical reason to use very small

aperture eg F16 and smaller when using long lens (except for special

purpose like for very low shutter speed effect.) for shooting

landscape?

Most of the lens reviewer/testers show that most lens yield the

maximum sharpness at 1 or 2 stop down from the maximum F-stop (I mean

small number like F 1.4 for example.)

 

So I can't figure out when/why we need to use very small aperture for

the

long lens. Theoretically, F8 or F 11 would yield the best result for

300mm F4 lens most of the time. Am I understanding this correctly?

 

Thanks,

 

Passakorn

Hi Passakorn,

 

I would definitely stop the 300mm lens down to F/16 when I need more depth of field for landscapes. If the scene demanded it, I would even go to F/22. You will not get as much unsharpness caused by diffraction with the 300mm lens at F/22 as you would with a shorter lens like a 24mm at F/22. Given both lenses set at F/22, the aperture (size of hole) is bigger with the 300mm lens than the 24mm lens. As holes get smaller, a higher ratio of light strikes the edge of the hole and bends somewhat causing the diffraction problem resulting in unsharpness. As always, it's a tradeoff. While lenses tend to be at their sharpess around F/8, the slight loss of sharpness that you get by stopping down for more depth of field is very slight and generally of no consequence in the real world of selling images. If there is nothing to be gained by stopping down, then I would indeed shoot around F/8. Otherwise, stop down to get that extra depth of field.

Given Mr. Gerlach's response another point to consider is one might

stop down to lengthen one's shutter speed which sometimes helps to

improve the apparent sharpness of an image.

 

To clarify this a bit-sometimes a very long shutterspeed will look

sharper then a mid range one (think 4 seconds vs. 1/15th of a

second).This techique is obviously related to external factors like

wind, subject motion

Sorry to disagree slightly with John, but the effects of diffraction ON FILM are related to both the physical size of the aperture AND the distance of the aperture from the film.

 

True a 300mm lens at f22 has a much bigger physical iris aperture than a 24mm lens at f22 does, and so the angular diffraction at the iris is less. However, the light has much further to go to get to the film and though the angular extent of diffraction of the light is lower, the linear extent of diffraction on film is nominally the same in both cases. For a simple lens it's easy via simple geometry to show that the blurring of the image on film due to diffraction is determined only the the focal ratio (f-stop) of the lens. Showing this for complex (multi-element, telephoto or retrofocus) lenses is much more difficult, but the same arguements generally apply.

 

In careful tests with most good lenses you can certainly see the effects of stopping down past about f11 on image sharpness, but whether they are significant depends on what you want to do with the image. My Canon 300/4 seems to peak between f5.6 and f8. Stopped down to f22 you can certainly see a drop in sharpness and unless I needed the DOF I wouldn't stop down that far. F8 or f11 is fine though.

 

To answer the original question, Yes, there can be a point in stopping down if you want extra DOF, even with a telephoto lens shooting landscapes.

It depends how far away the scene is and what DOF you need. At f8 the HFD is 450m, while at f22 it's 160m. So if your "landscape" stretches from 300ft away to infinity, you certainly might want to stop down past f8. If everything is at "infinity", say over 1000m away, there might not be any real point in stopping down past f8 or so though. Each case is different.

Oh no, not again... Light spreading at the film plane due to diffraction is proportional to effective aperture (f-stop) and wavelength of light. So John Gerlach's statement above is false.

 

As far as image sharpness is concerned, it is best to keep long primes fairly wide open, because of the diffraction, and, more importantly, shutter speed. The gain of DOF by stopping down is meaningless if you loose sharpness to camera shake or motion blur.

It seems like my and Bob's posts crossed... But back to the original question, sometimes man got to do what a man got to do, sharpness nonwithstanding...

Andrei - have you ever seen a rigorous analysis for the diffraction spread function in the film plane for telephoto and/or retrofocus lenses. I've seen some suggestions that the answer may not be exactly the same as for a simple lens, but they were just suggestions with no real mathematical or optical basis and I don't really believe them. The simple, thin lens case is easy to prove of course, but you can't use the same proof for a complex lens without making some assumptions. It's similar to the DOF issue, which again is easy to prove for a simple, thin lens, but complex designs make for a more difficult analysis.

To ask Mr. Pananont a question; what shooting situations did you have in mind when formulating your question? I have personally obtained more desireble results using focus to dictate depth-of-field instead of aperture. An example is this; on a recent trip to Chicago, I photographed a woman walking along the west facade of Union Station. I wanted to examplify the stacked look of the columns on the facade, so I stood some distance away and shot with a Nikkor 300/4.5 at f/5.6 or f/8 (I don't recall which). Now for a 300, this doesn't provide an enormous depth-of-field, but the lens was focused somewhere between 70m and infinity, so most of the receding facade was in sharp focus.

 

A 300mm lens does not make a good near/far landscape lens, stopped down to f/32 or not. It does make a good landscape lens with good, or at least adequate depth-of-field, when focused near infinity, which is probably what you will be doing if you are trying to isolate and coompress a distant view or subject. Therefore, I don't really think that there is much benefit to stoppong the lens minimum aperture.

The situation I had in mind is when I want to take a landscape photo such as I am on the top of the mountain and want to take photo of another peak. The distance should be almost infinity (no closer foreground). Usually I use the smallest aperture all the time for the maximum DOF. It just crossed my mind recently if it's worth doing it all the time for the max. DOF. Actually, I use zoom lens (70-300ED), but I think that the general rule should apply for all lens (except fare problem, I guess).

My point is that is there much different in DOF from aperture when you shoot with the long lens focusing almost/at infinitity. For example, how much different of DOF would be at F11 and F32 when focusing to infinite. Thanks for Bob' number, but I think it's a bit difficult to use when focusing lens because it will be almost infinite. I know most of us may say why not use DOF preview. I used it a lot, but sometimes it's a bit difficult to see through the preview when you focus at infinity and use the small aperture. I am just looking for some real experience result in the field.

 

Thanks for all answers/comments

 

Passakorn

Mr. Pananont,

 

I have just performed a "field experiment." 300 on tripod, I aimed out of my window at a distant building. Focus was near 70m (close to infinity). I viewed the frame at f/4.5, noticing that tree branches falleing between me and the building, approximately 10-20m away, were out of focus. At f/8, these tree branches had sharpened a bit, but the difference was hardly noticable. At f/32, however, the branches were in reasonably sharp focus. Therefore, a considerable depth-of-field (100-120m) was gained. As for sharpness at this aperture, I don't know, as I never shoot that slow.

 

If you are shooting distant mountain peaks, close to infinity, I would reiterate Mr. Atkins comment that stopping down below f/8 or so would be somewhat pointless. However, depth-of-field with more near/far compositions would be greatly extended.

 

It is also necessary to remember that zoom lenses tend to perform their worst at tiny apertures at the end of their focal length, meaning that any unsharpness caused by an f/22 of f/32 would probabbly be exacerbated.

 

Hope that the info is helpful.

If you don't need depth of field, you should stop down two to three stops from maximum aperture for maximum quality.

 

If you do need depth of field (or want to minimize it - say, to throw a distracting background as much out of focus as possible), you should use whatever aperture you need to do the job and not worry too much about the quality.

 

For decent quality lenses, all apertures will perform pretty well. However, if depth of field isn't a concern, and you can use a tripod (or an adequate shutter speed to handhold), why not use the aperture that will give the sharpest possible results?

 

Jim

Bob, I haven't really seen any rigorous treatment of diffraction in

complicated lens designs. (Although I do physics, optics isn't my

speciality :). Anyway, I've thought about it for a bit, and it seems to

me it is possible to show that diffraction does not depend on lens

design, but on effective aperture only, assuming thin beam

approximation and diaphragm in front of the lens. I can post it if you

want to take a look. HTML isn't exactly suited for typing formulas...

There's yet another point to be considered here: Movements from the mirror. If your camera does not have a mirror lock-up feature, the mirror may cause some shaking in your camera that affects the image quality adversely. Some shutter speeds are more sensible in this regard than others; it may actually be better to shoot at 2 to 1/2 a second than on 1/15, 1/30 or maybe even 1/60. Assuming you don't carry along different grey filters, you may be better off stopping the lens down to get a longer exposure time, EVEN IF this gives you a bit more diffraction. Why not perform your own test? Shoot indoors with a powerful flash to find out how much your lens will suffer from diffraction on different f-stops, then shoot at different shutter speeds to find out if the mirror causes a slight blur.

(I would not use my own Nikkor 300 f2,8 stopped down to 22 or 32, except maybe for panned pictures. The sharpness declines from f 8.)

I have been working on some long lens 300-420mm night time photo,s for a little personal project of mine. The results have been fantastic. typical night time exposer are something like 5-20 minutes at f22-32 with 400 coler neg. I could shoot slow slide film but dont feel like standing around for hours waiting for one picture to be exposed. The deph of feild is incredible, and sharpness is fine as far as I can tell. Whats funny thing is that some times Im shooting from overpasses and when people walk by the whole bridge shakes , but since some of the exposers are 20 minutes or more there is some bluring from traffic lights but the rest of the photo appears sharp. Several years ago I took a shot of toronto,s younge street with a 300 f2.8 with a 2x converter with kodachrome 25 at night . I stopped down all the way with my old sp tamron 3002.8 so f stop was f64 and exposer was about 3 hours. Cool slide ,one of my favorite,s. The depth of feild is incredible, with hours of traffic flow in one frame. My advice is to break some rules and burn some film . Let the sharpness snobs sit at home and do math . Get out there and start shooting OO