Real-world "digital" DOF...for LCT, Ho-Co et al

[adan]

Advertisement (gone after registration)

Thought this might be of interest - real-world pictures of just how untrustworthy DOF scales are if one wants big images from cropped sensors (Hasselblad, in this case) - along with a bit of history and theory of DOF.

 

http://luminous-landscape.com/tutorials/Digital%20Focusing.shtml

[ho_co]

Thanks for the link, Andy!

 

As we know, the standard DoF markings are already recognized as out-of-date for 135, despite what Leica says in the M8 booklet. And that's what G Ferguson shows.

 

Here LuLa approaches the matter not by looking at graphs of the math involved, but by showing the results. A very good addition to the thread to which you allude!

 

I would simply stress that we've recognized all along with film that the depth of field engravings aren't to be trusted for great enlargements, and accepted the results. In many cases now we've got smaller sensors than previously and need to consider the effects of the crop factor as well.

 

There may also be something in the nature of digital capture itself that we need to consider--we already recognize that in many cases, digital 'looks sharper.' Then any decrease in depth of field would be even more noticeable.

 

--HC

[rosuna]

Ferguson's analysis is very interesting, but it shows and obvious thing: the focus plane is just a plane, not a gross wall.

The 135 marks show the limits for perceived sharpness in the print, and this depends on the print size. 100% crops can not show this. The sharpest crop and the blurest crop will not be so different in the print if it is sufficiently small. Of course, they are different, but the human eye cannot perceive the difference. However, 100% crops show the differences. The same was the case with film. There is nothing special about digital cameras here.

[john_r_smith]

Ohmygod. And that 100mm Hassy lens is one of the best and sharpest optics Zeiss has ever made. But of course, the 100% crop shown would be a tiny part of the final print. Still, so much for the standard circles of confusion

 

This is very interesting to me, because I have been running full-tilt into severe DOF problems with my MF work recently. Much worse than I ever remember before. Now it occurs to me that this also coincides with my having the film scanned and then printing digitally. Is it that the in-focus bits now look sharper, and so the transition of the DOF seems more abrupt?

 

John

[john_r_smith]

Incidentally, one of the things that drives me absolutely crazy with the LC-1 (Digilux 2) is the lack of any DOF indicator on the lens. I know that at the wide end there is so much DOF you hardly have to worry, but at the long end the only way to be sure is to carry a printed table with you.

[lct]

Lens markings did work in 135 for me and for APS-C i always use those of the nearest faster f stop without any problem.

Theory and practice happen to say the same fortunately, at least for me.

[rosuna]

Advertisement (gone after registration)

Lens markings did work in 135 for me and for APS-C i always use those of the nearest faster f stop without any problem.

Theory and practice happen to say the same fortunately, at least for me.

 

Dear LCT,

 

then, if you set f/8 you use f/5.6 marks for focusing, isn't?

 

It is a safe rule. In that way, you get a security margin.

[lct]

Just my good old formulas, Ruben.

Based on usual CoC values (0.03mm for 135 and 0.02mm for APS-C), any lens on a 135 camera has about the same DoF at f/5.6 as the same lens at f/8 on APS-C bodies.

Then given that DoF markings of lenses are based on the 0.03mm CoC as well, i use the f/5.6 ones when i choose f/8 on my R-D1.

Don't know what modern theories may say about this but it works for me.