stopping down, how much is too much?

[brian steinberger]

I have two lenses, one that stops to f45 and the other stops to f64. Alot of times I want to maximaize my depth of field without diffraction. What exaclty is diffraction? How far is too far to stop down? And in what curcumstances will I see the effect of diffraction more? Ansel Adams and others believed in f64. I'm just curious. Thanks!!

[Tony Karnezis]

http://www.largeformatphotography.info/fstop.html
http://www.largeformatphotography.info/lfforum/topic/500779.html

In general, the advice that I've been given is use the aperture that's necessary to get the proper DOF and don't worry about diffraction. I was told that for contact prints, you can't notice diffraction until you get smaller than f/128 or f/256. Could you tell us what format you shoot and how you print (ie. enlargements vs contact prints)?

[brian steinberger]

I'm shooting 4x5 Tmax 100, mainly 11x14's on 16x20 paper. I do conatct prints as reference, never to be a finished print though..

[Edward (Halifax,NS)]

If you want no effect, stop at f/32. If you stop down more it would be up to you whether the final results are acceptable. As to Adams and f/64, he was shooting 8X10 so the level of magnification is lower. Also he prized full depth of field over minimal loss of sharpness.

[Robert A. Zeichner]

The rule of thumb I've repeatedly heard is to avoid apertures smaller than f32 for 4x5, f45 for 5x7 and f64 for 8x10. When diffraction starts to become a factor, the visible result is a softening of fine detail. In certain situations where the need for depth of field is the difference between making the image you want or not, I have closed down to f45 or even a half stop smaller in 4x5 without regret in most cases. I suppose in situations where there is a tremendous amount of fine detail in the negative, the softening effect may be more obvious and objectionable.

[Donald Qualls]

Diffraction is unlikely to be noticeable on prints at around 3x enlargement (like yours) at f/64. Depending on the focal length of the lens and amount of magnification, f/90 may produce unacceptable diffraction, or you might be able (if your lens supports it, or if you were to make Waterhouse stops and install them in the shutter over the existing diaphragm) to go to 1/128 or f/180.

Worthy of note, however, is that pinhole photography has diffraction as a limiter on pinhole size, where the increased sharpness of a smaller circle of confusion starts to be overwhelmed by the decreased sharpness due to diffraction; for a pinhole distance (equivalent to focal length) comparable to focal lengths we use in large format, say 150 mm, the "optimum" pinhole size, the one with the best compromise (for sharpness) between circle of confusion and diffraction, is 0.49 mm (or about .019"), which corresponds to about f/300; this will give effectively infinite DOF with everything in field equally "semi-sharp". Any aperture similar to this (f/256, surely, but probably f/180 and possibly even f/128) will show some of the same characteristics as a pinhole -- extreme depth of field, but coupled with a visible loss of sharpness (which becomes more noticeable with any magnification) in all parts of the image.

There are mathematical formulae that suggest a minimum aperture to avoid unacceptable diffraction effects, but like DOF calculation, they're a little slippery because of the subjective definition of "unacceptable". How much image degradation you're willing to accept from a particular source (whether defocus, camera motion, diffraction, or lens aberrations) depends on how much (if any) you'll enlarge the negative in printing, what level of sharpness or fuzziness you find comfortable in the final print, and what other sources of unsharpness may be present (for instance, I've found that with a pinhole body cap on my 35 mm SLR, I can easily hand-hold at 1/4 second, because the fuzziness from the pinhole overshadows motion blur in most shots) -- all of which, in turn, may depend on the subject being photographed and the type of image sought.

No, I know that's not a clear answer -- but there *isn't* a clear answer on this one.

[N Dhananjay]

What exactly is diffraction?

When a beam of light passes an edge, it bends or spreads around it - this phenomenon is referred to as diffraction. Incidentally, diffraction occurs with all kinds of sound waves - sound, water etc. It can be understood utilizing the wave nature of light - an intuitive way to understand it is to think of ripples (i.e., waves) on the surface of a pond if you toss a stone into it. Think of what happens when these ripples reach an edge such as a wall with a slit in it through which they can pass. On the other side of the wall, you see these ripples spreading out as though the slit was the source of the waves (i.e., as though you had dropped a stone into the water exactly at the slit). In other words, you do not see the wave continuing in a straight line past the slit but it sort of propogates in all directions around the slit (i.e., it bends around the slit). The reason for this bending/spreading is the fact that the wave consists of a whole lot os waves in sync with each other. So each wave sort of pulses up and down but exactly in sync with the waves next to it, so that it is a stable wavefront. Now think of what happens if some obstruction cuts off the wavefront next to another wavefront. All of a sudden, this wavefront does not have another wavefront in sync next to it as it pulses. As a result, it spreads out.

How far is too far?

In photography, diffraction is only a function of f-stop (an oversimplification - it also depends upon the wavelength of light etc but its the only thing we have control over). The smaller the aperture, the higher the diffraction. However, diffraction is only one of the things we have concerns about. Two main reasons why we stop down. One, most lenses have aberrations that are large wide open and become progressively smaller as you stop down. So, there is some optimal point when you stop down where you have benefited from these reductions in aberrations but not suffered too much loss from diffraction. Two, depth of field. Balancing DOF and diffraction can be tricky. Diffraction results in the gradual softening of everything across the whole image. DOF is a selective softening - somethings are in crisp focus and some things are very much out of focus. You decide which one you can live with - generally, I find lack of DOF more troubling than a slight softening of the whole image. I think the reason for this is that the eye-brain system makes sharpness judgments by comparison. As long as things are not terribly and obviously unsharp, it looks to see if one part is sharper than the other.

In practical terms, you can approximate the diffrcation limited resolution with mathematical formulae (something like 1500/aperture). This means that at f/64, you would get about 24 lpmm, at f/45 you'd have about 34 lpmm, f/32 you'd have 47 lpmm etc). The figure usually quoted for the eye is the ability to resolve about 5-10 lpmm on the print (under specific viewing distances etc). So, if you are contact printing, you can stop down to f/90 and still have adequate resolution. If you are making a 3X enlargment from 4x5, at f/64, you'd have 24 lpmm on the negative and about 8 lpmm on the print (making the unrealistic assumption of no losses during the print making process), which would probably still be adequate. Some may want to play it safe and leave some more resolution for losses during print making etc. Although, from a visual perspective, I think a more logical approach is to get the DOF your visual concept needs and settle for the largest size print that gives you adequate resolution. Lack of DOF means you can't make any print at all, whereas diffraction might mean you can make only an 8x10 print instead of a 16x20 print.

Under what circumstances can I see diffraction?

If you want to see what diffraction looks like, the easiest way is under an enlarger. Place a negative in the enlarger and using a high quality focussing aid, focus on the grain in the negative. Now, stop down the enlarging lens and watch the grain turn to mush. In the field, it is hard to see diffraction. At the kinds of f-stops where you could see it (f/128 and above), the image on the ground glass is terribly dim. You can use a loupe but that also magnifies the image (the loupe may be useful to try and estimate what diffraction would be like at a particular enlargement, though). So, you are probably better served with using rules of thumb. Try using movements to reduce the amount of stopping down. Try to stay with as large an aperture as permitted by other demands such as DOF etc. Although, as the above figures have indicated, you should be fine with f/45 and probably even f/64 if you are making 11x14s from a 4x5 negative, and are a careful worker.

Cheers, DJ

[Kevin Crisp]

Brian: I would list "don't be afraid of small stops" as a lesson I wished I had learned a long time ago. I recently made a mental math error in the field converting the shown stops on my shutter for what they should have been for the lens elements I had just moved over to it. (Moved a 305 Repro Claron into a Compur 2 marked for a 240 mm lens). I ended up shooting at f:64 when I didn't intend to on 5X7 film. The print at 16X20 looks very sharp at any reasonable viewing distance and people often remark on how "sharp and clear" that print, in particular, looks. So try it, f:45 will look fine in prints this size with 4X5 film. Sometimes it is just the only way you can get what you want in focus, and I agree with DJ, you'll probably be quite satisfied with it. If I made larger prints (which I can't really do with my present sink) then I'd have to reevaluate, but within these limits, you should be fine.

[Mark Sawyer]

When I got a 480mm Apo Ronar for my 8x10, the first thing I did was shoot a negative at the smallest f/stop, f/256, just because I could. The negative is very sharp under a 4x loupe. As I tend towards contact prints for the smoother grey-scale, I now disregard the diffraction issue.

One exception is in macro work: The effective f/stop is actually determined by aperture size and the distance from the lens to the film, NOT what is marked on the aperture scale, (which is for infinity only). We compensate for this with a "bellows extension factor," but it could more accurately be called a "diminished f/stop factor."

In other words, if you have your 215mm lens cranked out to 600mm, you're shooting with a 600mm lens that just happens to be focused a whole lot closer. And f/64 on a 600mm lens is bigger that f/64 on a 215, so if you shut down to the scale-indicated f/64, you're actually way smaller than that.

But don't make me do the math... I save that for real photographs.

[Brian Ellis]

Lots of excellent answers. I'd only add that IMHO diffraction isn't a significant problem and shouldn't be a concern with 4x5 negatives until prints are larger than 16x20. I think that insuffcient depth of field will almost always be more objectionable than whatever minimal loss of sharpness results from diffraction so I use whatever aperture is needed for the desired depth of field without being concerned about diffraction.

[Leonard Evens]

You should look at the discussion of how to choose your f-stop at this same website. In particular, you might be interested in Paul Hansma's method described therein. He has produced a formula which for a given focus spread on the rail---discussed there also---gives you the f-stop which gives you the best balance between diffraction and defocus (related to depth of field). The formula involves a square root and most people would not want to try using it in the field, but it is easy enough to carry a small table with you. The website has the table. I find the f-stop recommendations too conservative in many siutations, but it at least can give you an indication of the smallest f-stop you might want to consider for a given focus spread.

[Emmanuel BIGLER]

I like this discussion very much because we try here to criss-cross
the physicist's point of view with an aesthetic point of view.
Difficult to argue against the Laws of Physics. But it is more fun to
ignore them and do whatever suits our hands and eyes in the whole LF
photographic process.



A closed-form physicist's and engineer's point of view could be :
Brian. There exist and optimum f-stop; say : f/22 for the 4"x5"
format, scale the optimum f-number like the diagonal of the format,
this yields f/45 in 8"x10", etc.... in macro at 1:1 ratio : take into
account the effective f-stop which is one stop further w/respect to the
engraved f-stop. You are "allowed" to close down 1 or 2 f-stops beyond
the optimum engraved f-stop of your preferred LF format. If you dare to close
down even more... you are entering the forbidden zone (forbidden only by theoretical considerations, of course) where so many
bad things can happen to your image that you should not do it, no,
never ;-);-)



The aesthetic point of view says : Brian; as far as LF photography
is concerned no, never listen to theoretical engineer's
considerations. LF photography means : absolute freeedom. You
can "mechanically" stop down to f/256 ? please, do it and enjoy ;-)



Not kidding now, there is an interesting article (sorry, in French) by
Jean-Marie Solichon with some real images taken with, or without
entering the "forbidden diffraction f-stop zone".

http://www.galerie-photo.com/diffraction.html

In a discussion on image "sharpness", the choice of cactus plants as a test object is probably meaningful ;-);-)
Jean-Marie prefers to stop beyond f/22 "only in case of absolute
necessity".



Une impression sur papier photo brillant des fichiers originaux en
TIFF confirme cette impression sans ambiguïté et me conforte dans mes
habitudes tendant à diaphragmer plus que f :22 uniquement en cas de
nécessité absolue.

Stopping down so much that you enter the "diffraction zone" has the
advantage of an homogeneous un-sharpness across the whole image field ;
this may, or may not be acceptable. Pinhole aficionados like very much
the absolute homogeneity of image (un)-sharpness from one inch to
infinity in the image field. My usual objection against alleged infinite depth of field in pinhole photography is : none of any image point will satisfy my preferred sharpness criterium, e.g. 100 microns of defocused spot in 4"x5" ; so to me pinhole depth of field is : zero because none of the object points are actually imaged with sufficient sharpness ; so do not transfom my beloved LF camera fitted with a beloved sharp optics into a pinhole camera ;-);-);-)



Although I accept the idea of homogeneus unsharpness from an aesthetic point, although I respect the quest for absolute freedom in pinhole photography (down with glass and optics ! down with Scheimpflug ! down with focusing !down with expensvie gera, we demand a cardboard shoe baox a s acamera ! down with flat field and flat image detectors, we demand curved image detectors ! ), this point of view this does
not actually suits my taste on the resultiong images ; I prefer sharpness and I agree with the
article by Jean-Marie Solichon, to me large format photography means
excellent sharpness, not homogenous un-sharpness ;-);-);-)



But I was trained as physicist, not an artist !

[Struan Gray]

Even physics argues for an aesthetic approach. The Airy function is quite sharply peaked, especially compared to the top hat of defocus, so 'best' is indeed in the eye of the beholder. For my money, you can get away with quite a bit more diffraction blur than the 'width' of the Airy function would suggest.

[Kirk Gittings]

"I think that insuffcient depth of field will almost always be more objectionable than whatever minimal loss of sharpness results from diffraction so I use whatever aperture is needed for the desired depth of field without being concerned about diffraction." Brian

This is the god's truth. Well said.