I've laid out a clear, comprehensive, and accurate explanation of the issue. Short of typing out large excerpts of Ansel Adams "The Camera" or other photographic texts, I don't know what else to tell you.
Unfortunately, what you have said is far from being clear, comprehensive and accurate. I have tried to show you where your concepts are inaccurate, but it appears I am unable to communicate these to you effectively.
David,
Depth in DOF is a dimension, it requires a definition of its boundary. The COC is that boundary.
Cheers,
Bart
Bart, I agree. Let me try to again point out what I find preposterous. According to Sheldon's previous note above he says, "Without the print or monitor, DOF does not exist." I can't see how any sane person can believe this. If it were true, how would it even be possible to set proper exposure settings on your camera to get the desired DOF that you wish to capture? Obviously, you must have some ability to establish DOF prior to actually capturing the image, irrespective of whether or not a printer or monitor exists.
Point 1 is right. But there is a real theory behind it, based on human vision.
You may check this: http://www.betterlight.com/downloads/whitePaper/depth_of_field.pdf
So, those guys at Phase One better get their act together. What were they thinking when they created a purely digital camera with a DOF button on it?
Technically, there is nothing in the above pdf document that is inconsistent with what I have been saying. When you transfer the concept of DOF, a purely optical phenomena, from your eyes/lens to the digital sensor, you are basically substituting the CoC of the pixel for that of your retinal viewing area. This creates an effective DOF relative to the digital sensor, and is no different than what photographers have been referring to as DOF for many years with film using film grain instead of pixels to establish the CoC. Indeed, at this point in the photographic process the DOF is
fixed, whether it be film or digital. I am not disputing that the DOF is drastically different with digital as opposed to film as pointed out by the Betterlight document, since the CoC values are indeed drastically different.
However, what I am disputing is the role of print size here. Printing involves its own and
different CoC, which determines how much resolution can ultimately be seen in the final print in conjunction with a particular substrate, e.g. whether you will be able to see 200 dpi, 300 dpi, 400 dpi, etc. on your printing substrate. In no way does this printing process affect the DOF of my image that I previously captured. The printing process only establishes one particular viewing resolution, which will vary with print size. BUT, the DOF of my captured images do not change just because I print them. This is the point of my contention.
Doug and Sheldon claim that DOF depends on the print size. I simply do not find this to be true.
The ONLY case where the DOF is actually determined by the print is when the image capture process and the print process are one and the same, i.e. Polaroid. The DOF is always fixed at the point of capture.