Originally posted by biz-engineer Pixel shift really is something the collect bits of sharpness when the lens itself out-resolve the sensor. On the K1, I can generally see a difference between pixel shift and non pixel shift with the 100 macro stopped down. With zooms, I see no difference and due to the constrains of pixel shift shooting, images are, more often than not, sharper without pixel shift than with pixel shift. That said, on the K1, pixel shift is a work around to get EFCS in OVF mode, the tick is to do a pixel shift capture handheld and disable pixel shift when converting the raw into jpegs; this way, you get tack sharp images free of mirror and shutter induced vibrations, without having to use live view. If you use a K3II, or K70, using pixel shift and discarding it in post processing does the same: you get electronic shutter with OVF and you can shot hand held because the 3 extra pixels shifted frames are discarded so there is no motion blur in the resulting image.
How do you know a lens is out-resolving the sensor? Are there specific tests for that? I am really curious about that!
I thought about using Pixel-Shift with my old gems (K 35 3.5 and K 28 3.5, maybe stopped down) and would really like to know if they can outresolve the sensor (At open aperture or maybe they can't at all, no matter how closed the aperture is?)