I have been playing around with the DOF equations sent to me by Tom and I found something interesting: the size of the circle of confusion varies according to the distance you focus. Well, it's not really that the size varies, it's just that since DOF is a non-linear function the scale on the lens cannot accomodate these differences and so it seems that there is a different value for "c" throughout the scale.

For example: with the 90mm lens focused at about 7m, guesstimating from the markings on the barrel, you get dof from about 3. something meters to infinity at f/22. This corresponds to a c of 0.092. However, if you focus this same lens at about 2.9m, to get 2m on the closer part of the f/22 marking and 5m as the farther part of f/22, c now has a value of 0.054.

This probably isn't news to most people--but I never knew about it. This means that if you want to use 0.065mm as the circle of confusion for 6x7, the markings on the barrel are underestimating dof when you focus close, and overestimating it when you want infinity in focus.

What does everyone think of this.. ?

Regards, Marcelo

ps- I found the same thing is true for the 45mm lens, of course. C = 0.092 to get infinity in focus at f/22, and c = 0.036 if you include 0.7m to 1.5m between the f/22 markings.

-- Marcelo P. Lima (MPL4@cornell.edu), May 23, 1999

Answers

Marcelo: Interesting points here. I wish Pentax would have used their 35mm CoC for their 67 lenses. Figuring the DOF scale based on larger CoC only reduces the benefit of using a 67 over a 35mm. What is strange is the fact that different lenses have various degrees of accuracy of their DOF scales. The 75mm is fairly close, the 300mm is way off, while the 600 DOF scale really can't be used when a rear filter is in place. To take full advantage of this film size, Pentax needs to make their DOF scales more accurate on their next redesign. SR

-- Steve Rasmussen (srasmuss@flash.net), May 24, 1999.