Diffraction and f stops

[gma]

Regarding the loss of quality caused by diffraction when using very small f stops:
Is there a specific diaphragm diameter (in mm) that is common to all lens focal lengths at which diffraction degrades the optics? Typically 50 mm lenses are best around f/8 and longer lenses at smaller f stops. I would think that somewhere in the range of 5 or 6 mm diameter any lens would achieve the best performance, whatever that f stop is for any particular lens. Somewhere I read that most lenses are best at 2 or 3 stops from the maximum aperture. Another factor might be ths shape of the iris. Most older leses have a perfect circle and some newer lenses have a definite geometric opening.

gma

 

[Ed Sukach]

Lenses do not go abruptly from "best" to total disaster. I realize that a few photographers are going to be super-critical and consider the "best" aperture to be "the only one acceptable" ... but then ... why have an aperture selection at all.
I personally will not hesitate to use any aperture I want to.

The "best" aperture will depend on the lens design criteria. Enlarging lenses are typically "best" in the center of the aperture range ... but the change from there to the extremes is ... let's face it ... not great.

Camera lenses - *may* be best in the "center", but high-speed lenses *tend* to be better closer to the larger aperture openings - under the theory of "that's where they will be used most often".

Guys... I would suggest that more confidence be placed in the Lens Designers - most of those I've worked with were *very* knowledgable and bright. They understand "diffraction" and a whole lot of other things - very well.

 

[David A. Goldfarb]

If you want to find the optimal aperture for a particular lens, it's usually best to test that lens, since as Ed says, design characteristics vary and there is no universal rule.

There are certain lenses that I have that I try to use at certain apertures, because I know they have a nice look at one aperture or another (e.g., the Voigtlander 50/2.0 is beautiful at f:8, and Heliars look most like Heliars at wider apertures, and the Color-Skopar on my Voigtlander Perkeo II likes to be at f:11-16 or the corners will be unsharp), but in general, if you need the DOF, it is better to stop down as much as you need to get the DOF you want and not worry about diffraction. If you are using relatively modern lenses, they are usually designed not to stop down past the point where diffraction is a significant worry.

Inadequate DOF is always a bigger visual distraction, in my opinion, than the general softness that may result from diffraction. If you are getting noticable diffraction at small apertures, then the solution is often just not to print as big. If you shoot large format and contact print, you can pretty much ignore diffraction effects if you don't go past, say, f:256.

 

[Donald Miller]

I personally find diffraction to be a consideration in my photography. Like many other photographers, my early large format efforts amounted to stopping down the lens to eliminate the need to use the camera movements. That was true because the use of camera movements was daunting to me at the time.

I think that it is worth noting at this time that I am using relatively modern lenses and my goal in their use is to obtain maximum print sharpness and local contrast. I understand that if one is using older lenses or soft focus lenses that what I am about to say will not be applicable.

I think that a great deal depends on the film format, how it is going to be printed, and to what enlargement. If that is involved. Obviously if one is using medium format then movements are not normally available.

Diffraction is a factor of lens opening size to focal length size coupled with the film format size. It is caused by light rays bending at the aperture opening. If the opening is allowed to be of sufficient size the bending of these rays does not become the consideration that it does when the lens is stopped down excessively.

In my photography today, I try to never stop 4X5 down below F32 and ideally I try to keep it at F16-22., On 8X10 and 12X20 I will stop down to the limits of the lens if indicated. In my 4X5 I enlarge the negative typically to 11X14 and occasionally to 16X20 maximum. My 8X10 and 12X20 are contact printed. If I were enlarging 8X10 my fstop limit would be F64. That is because the enlargement factor would not be as great as 4X5. As degree of enlargement increases the effects of diffraction become increasingly apparent.

Since I began living to the parameters I indicated for 4X5, my print sharpness and local contrast have definitely shown an improvement.

The effects of diffraction are one of the links in the chain that will either contribute to or detract from the rendering of a fine print.

I didn't arrive at this criteria by myself. This is information that Howard Bond and others indicate as being a condition of their photography.

 

[David A. Goldfarb]

Interestingly, even though it seems like Don and I are disagreeing in theory, I follow almost exactly the same practice. Since I enlarge 4x5", I try for around f:22 usually, unless I want very short DOF or high shutter speed or I'm in low light, and try not to go past f:32, though occasionally I'll go to f:45. For 8x10" and 11x14", I'll also go as far as the lens goes if I have to, which is usually f:45 or f:64, since I contact print these formats.

Michael A. Smith has written that he'll go as far as f:256 (I suspect his only lenses that go that far are longer Artars), and he contact prints exclusively.

 

[Thilo Schmid]

Diffraction can be quite limiting for macro work, because it depends on the effective f-stop. A lens at f22 will have an effective f45 at 1:1, which is already limiting at 4x5 and "normal" print sizes.

Sharpness is always a compromize between DOF, diffraction and shutter speed.

I, personally, try to avoid f-stops below
f32 for 4x5
f22 for MF
f16 for 35mm
and one f-stop less for macro work.

 

[Jimmy Peguet]

Michael A. Smith has written that he'll go as far as f:256 (I suspect his only lenses that go that far are longer Artars), and he contact prints exclusively.

I think Michael was speaking about Weston's photographs, not about his own pictures. He once wrote to me that EW used apertures up to f64 and that there was only once, he thought, that EW made a smaller aperture. The system used by Weston on some of his lenses may have been US stops, which is a different system from the current system. In this system, 256 must be equal to 64 in our system. This confirms other discussions with some friends good connoisseurs of vintage photography.

 

[gma]

Thanks, Jimmy for the reminder. I have some of those lenses of 1900 vintage that are engraved with the old US stop system. I knew that they are not f/ stops and found a conversion chart some time back. By that time I had measured to determine what f/ they were. In case there are others who are puzzled by the numbers or are trying to figure out why their exposures are so far off here they are:

US 4= f/8, US 8=f/11, US 16=f/16, US32=f/22, US64=f/32, US128=f/45, US256=f/64

gma

 

[Ed Sukach]

Diffraction is not linearly consistent with "f/stop" ratios. It is a function of aperture diameter and distance-to-film plane, but not in the *same* mathematical proportion.

I've reviewed one of my old textbooks - "What is Light", by A.C.S. van Heel and C.H.F. Velzel (translated from the Dutch). Here they offer one of the simplest explanations of the diffraction phenomenon I've ever found. At that - `WAY too involved to be coherently condensed enough to be posted here.

An extract:
"We now notice that a light beam cannot indefinitely be made thinner and thinner. The following experiment will show that phenomenon of Airy's disc is indeed caused by the bounding of the light beams. We allow a wide parallel beam of light to fall on a screen with a small hole ( figure 34). A good ways off we put a second screen on which we observe Airy's disc, the diameter of which is seen to be inversely proportional to the diameter of the hole (and proportional to the wavelength of the light used), ...
... We will be able to explain the so-called diffraction phenomenon when we have studied the effects which appear as a result of the interference between several waves - ... "

The next ten or twelve pages are fascinating... but require a *great* deal of focused attention.

The net result, at least in my humble interpretation - applied to camera lenses - is, "Don't worry - be happy".
The lens designers are taking care of you.

Then again - there is the Holga...