If you are using a focusing screen to focus, then your ability to discern the plane of sharpest focus is limited by the DOF that the screen is able to show. A split-prism screen can be very accurate; a standard matte screen less so. Still, there are little tricks, like wiggling the focus back and forth and settling down in the middle, that can help you \"guess\" at where center of the focus should be, which means in practice you might be able to get fairly precise focus.
But if you are talking about the lens itself, focus shift by definition does not occur when the lens is focused and shot wide open--it is indeed the difference in the plane of sharpest focus that occurs when the image is exposed at an aperture different than the focusing aperture. As such, it is perhaps better to say that there is focus shift, and focus uncertainty. The former is a property of the lens, and the latter is a property of the method of focusing, be it AF or some variety of MF.
For instance, you can avoid focusing screens entirely and simply use Live View.
Mar 11, 2011 at 03:13 PM
Previous versions of wickerprints's message #9395074 « Risk of focus shift issues in new 1.4/35? »
