How to determine DOF

Started Mar 18, 2013 | Questions thread
ForumParentFirstPreviousNextNext unread
Flat view
wfektar
Contributing MemberPosts: 690
Like?
How to determine DOF
Mar 18, 2013

OK, sort of an academic question here. How do you measure depth of field?

Here's a comparison of two lenses near MFD



The one on the left is from the Cosina Voigtlander 58/1.4 shot at f/2, about 50 cm distance. The one on the left is from an OM 50/1.8 wide open, a little closer in to match magnification. To me it seems that the CV image has slightly shallower DOF, but that may be because 1) that's what I expect, and bias is a difficult thing to overcome. 2) differences in DOF are actually overwhelmed by OOF rendering differences. 3) the nominal f-number of at least one is not close to the actual.

This was stimulated by a discussion on "pixel shading", the inability of digital sensors to accept marginal rays at large angles, and it got me to do a quick and dirty exposure test:

This looks at an evenly illuminated scene at fixed ISO, setting the camera on A, and opening up the aperture and noting the shutter speed. The histograms look like they peaked in the same place but I did not quantify that. Anyway, the CV is well behaved until f/2, f/1.4 is only 1/3 stop faster. The OM behavior starts rolling off at f/2.8 -- I don't understand it but OK, it is what it is. These are on the NEX-7: details may be different on other cameras, though the general trend probably holds. Original thread here: http://forums.dpreview.com/forums/post/50791174

Anyway, I figured I'd look at DOF. Intuitively you'd expect DOF to follow the same trend as exposure and that the CV would have shallower DOF at f/2 than the OM at f/1.8 (hence the possibility of bias in the top image). In fact, the CV should have the same DOF at f/2.3 or so as the OM wide open. But how do you show that?

IOW: how do you measure DOF, without painstaking micrometric measurements on Airy discs and distances to precisions I probably can't replicate? That seems the only real way to deal with OOF rendering differences. Also, how do you measure actual f-numbers? Focal length I can manage (bellows plus target the size of the sensor, unless someone has a better way) -- is the entrance pupil measurement just a matter of taking calipers and measuring the image of the aperture?

ANSWER:
This question has not been answered yet.
Sony Alpha NEX-7
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPreviousNextNext unread
Flat view
Post (hide subjects)Posted by
ForumParentFirstPreviousNextNext unread
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow