Format creep and diffraction

[Leonard Evens]

I agree with Emanuel. In actuality you get essentially the same range of f-stops with any format, if you assume the same size final print, but the range is shifted towards smaller apertures as you move to a larger format from a smaller format. If you double the linear dimensions of the format, you lose two stops at the large aperture end in terms of the DOF you get and you gain two stops at the small aperture end with respect to diffraction because you don't need to enlarge as much. But the total number of useful stops remains the same. The problem, as Emanuel points out, is that you have to shoot a slower speeds.

[N Dhananjay]

As already pointed out above, with larger formats you are not trading extra diffraction for increased DOF. You are trading speed for extra DOF. Diffraction basically increases the size of the blur circle and is a function of f stop (for a particular wavelength). Now on a larger format, you do stop down more for the same DOF, and therefore you do have more diffraction/lower resolution on the neg. However, the neg is being enlarged correspondingly less - that is, the blur circle is being enlarged less. Therefore, the size of the blur circle on the final print remains the same. So, in practical terms, what this means is that with a larger format, you trade speed for DOF - that is, you use longer exposures. Hope this helps, DJ

[Matthew Runkel]

According to Stroebel, "[t]he diffraction-limited resolution can be approximated with the formula R = 1800 / f-number, where R is the resolution in lines/millimeter and 1800 is a constant." Combining this information with your expected enlargement and an assumption about viewing distance should let you estimate the best results you could theoretically achieve with a perfect lens, perfect film, perfect film flatness, perfect enlargement, etc.

Stroebel also says that "[t]he human eye generally is considered to have a resloution of 10 to 14 lines/mm at a viewing distance of 10 in." This may be a useful generalization, but I'm sure it collapses/ignores many interesting variables.

[Emmanuel BIGLER]

Another issue about very small apertures is the precision of the shape and actual area of an iris when closed down to f/64 and more.
On older lenses with no aperture clicks and no real upper limit, at f45 and above the travel of the lever between two full stops becomes smaller and smaller and sometimes the iris shape becomes irregular. So for this very practical reason as well, small apertures with f-numbers, say, bigger than 2 stops past the best aperture are not really recommended.
But to me large format means freedom. Even freedom to play with diffraction. 35mm and medium format lens manufacturers decide for us for the maximum allowable f-stop ; upgrading to large format with old lenses allows you to play with "forbidden ultra-small apertures" ;-);-)