Sensor size and diffraction.

Tom Axford wrote:

BobORama wrote:

Erik Ohlson wrote:

Cute cartoon, but you are obviously not considering the SMALL sensor, diffraction-limited cameras under discussion.

Your f/3.3 rule is bunk as it does not take into account the size of the sensels, which is about the only thing you don't mention, and the only thing that matters other than f-stop.

Your rather blunt response suggests that you are one of the small minority who believe that depth of field should be defined in terms of the number of pixels in the image rather than just the overall size of the image in the camera (i.e. the sensor size).

If the conventional definition of depth of field is accepted, then Erik is entirely correct. By conventional definition, I mean that used for the past hundred years; e.g. the definition used when depth of field scales are engraved on manual focus lenses and also the definition used by popular online DoF calculators.

If you wish to start yet another argument about the "correct" way to define depth of field, it might be better done by starting a new thread on that topic.

No, Bob's right: you need a well-specified resolution reference to give a CoC (circle of confusion), and that can be taken either by pixel dimensions or by approximating human visual acuity for a specific print size and viewing distance. The old depth of field scales from different manufacturers rarely used to agree, because there was not a universal agreement about the desired CoC with film.

In fact, this is part of the problem with infinity focus of some old lenses. It turns out that it was not uncommon for the infinity hard stop to be placed not at infinity, but at the point where infinity is covered by the DoF as they computed it. Unfortunately, for film most folks used a larger CoC, so using these lenses on high-resolution digitals reveals that they don't quite reach infinity. This is one of several reasons that adapters are often built just a little short....