stopping down for focus

[Los]

so, stopping down for focus (depth of field) seems like a pretty popular technique. so i tried it. using some aperture recommendations for numerous focus spreads. and it worked, sort of. while the near and far points are in focus, they have a fuzziness around the edges. i'm guessing diffraction is the cause of this. so my questions are:

Does anyone else see this phenomenon in their own deep stop photos, and are there any ways of improving it (unsharpmask, contrast, focusing somewhere other than halfway between the focus spread)?

[C. D. Keth]

Well, you don't focus halfway between. It's generally a third of the way from the closest to the farthest. The split gets closer to half and half as you get closer to the lens.

That doesn't sound right to me. It sounds like you lens either has some aberrations that are going wonky or you're estimating the spread wrong.

[David Karp]

Well, you don't focus halfway between. . . .

Actually, you do. Check this out: http://www.largeformatphotography.info/fstop.html.

And this article mentioned in the aformentioned page: http://www.englander-workshops.com/documents/depth.pdf

[Brian Ellis]

I can't really envision the effect you're getting. Fuzziness around only the areas that are "in focus" doesn't sound like diffraction since diffraction shows up as a slight softness of the entire image, not just parts of it. Diffraction is a potential problem with smaller formats, especially 35mm, because of the big enlargement factors involved (e.g. roughly 8x for an 8x10 print from 35mm film, which translates to a 32x40 print for 4x5 film). But it isn't usually a realistic problem with 4x5 and larger film because of the relatively small enlargement factors normally involved.

I assume you're using the focusing technique described in the first article David cited. If you're using that technique correctly any "fuzziness around the edges" is almost certainly the result of something other than diffraction but without knowing more details it's hard to figure out what that might be. In addition to providing more details (e.g. are you looking at a print or a negative, what film format are you using, is this b&w or color film, exactly what focusing method are you using) it would also help if you could post a copy of whatever you're looking at or at least a more detailed description of one of the images and the effect you're seeing).

[JPlomley]

I worry a bit about focus shift in the corners with wide angles, so if the light allows, I will determine focus spread with the lens stopped down to f/8 (even though maximum aperture is f/4.5) and use a Silvestri loupe instead of my Schneider (Silvestri does a better job at getting to the corners on my Arca Swiss GG and has a magnification of 6x which is beneficial for wide angle lenses). With all my other lenses I determine focus spread with the lens wide open, set the focus point to the median, and have never had a problem. Have never noticed any "fuzziness" when I stop down for the purpose of positioning a GND.

[Leonard Evens]

It is hard to respond because we don't know how you are choosing the f-stop. If you are using any plausible method to determine the f-stop, the image should be sharp at the limits of DOF over the entire image, provided you are viewing the image at a reasonable distance. Of course, almost any lens will perform better in the center than at the edges. Lens aberrations and curvature of field will generally degrade performace as you leave the center of the image circle. This will be clear if you view the image under significant magnification. So if you are using a loupe to look at the image, you should expect what you see. The basic method for choosing the f-stop based on focus spread, considering just defocus and ignoring diffraction, assumes some specification of sharpness in terms of a chosen maximum circle of confusion (coc). That in turn assumes the image is being viewed in a certain way. Usually you are advised also to stop down one or more additional stops to take account of less than perfect lens performance. If you use high magnification to examine the image, you are in effect choosing a very small coc, and lens defects are bound to play a significant role. Also, diffraction will be important even at moderate apertures. If you use Hansma's method, you will get, in principle, the best you can expect balancing defocus and diffraction. But, again, if you magnify the image enough, you will detect degradation of the image due to field curvature and other factors.

The proper placement of the standard is the harmonic mean of the lens to film distances for the near and far points. But in almost all practical situations this is very close to the midpoint between the two positions, and you should use that, unless you have some reason to favor either the foreground or the background. Some people have argued that you should always favor the background, but I don't find such arguments convincing. There are various one third rules, but they are generally wrong except in limited cirucstances.

[steve simmons]

The method I teach, and it worked in my recent Monterey workshop, is to have the student focus so that the closest objuct of interest and the farthest object of interest are equally out of focus. Then, as they close down, the dof lines spread equally near and far so that these two objects are brought into the dof area at the same f-stop. This is after all best efforts with movements are done.

One of the mantras I teach is "The groundglass is truth." If you really learn to study and look at the gg you won't be fooled.

steve simmons

[Robert Fisher]

Steve, what is "plan B" when the GG goes dark after stopping down?

[jetcode]

Steve, what is "plan B" when the GG goes dark after stopping down?

particularly in a situation such as a macro shot where the bellows extension just took 2 stops of light and when you stop down to f/16 you can barely see the image let alone guage it's sharpness

[steve simmons]

Have a big darkcloth that really blocks the light, use a 4x loupe, stay under the cloth to let your eyes adjust. The students actually found this to be surprisingly easy.

In the case of closeup work you can do a check with Polaroid. Type 55, where you check the neg, can be a useful tool. Outdoors, in daylight, no one had a problem.

I am also a fan of the gg brighteners as most of them will brighten the image by 2 stops.


steve