The real subtle detail is a diagonal pattern
about 5 times the the resolution and running the opposite direction
and at a different angle to the alaising output. The alaising
output contributes nothing. If this is subtle detail, what does it
convey about the real image?
Peter, from this, we should be able to deduce the precise
resolution (in the Nyquist sense). I.e., we know what the real
spatial frequency of the lines in that section. We know what the
image turned out to be (the lower frequency bands, which is nothing
but aliasing error). From that, we should be able to derive the
folding frequency.
I bet it is just the spatial frequency of the sensors
.
BTW, the slightly diagonal lines should do a bit better on the D60,
since the spacial frequency of the green sensors on the Bayer is
highest at 45 degrees to the vertical.
To paraphase a different popular saying: "Nyquist -- it is not just
a good idea, it is the law!"
By the way, a lesser known fact is that although Nyquist proposed
his "theorem" in the 1920's, it was not proved until Shannon's
Master's thesis in the late 1940's. And, here, in the next
century, we are still discussing if we can "see" beyond what the
Shannon-Nyquist sampling theory dictates
.
