<<How about an example; say m = 3 with [2,7] and [5,10]?>>
If the members of P(m) are 2 and 3, and if x = 2, y=7, x'=5 and y'=10,
then
N(x,y,m)=N(x',y',m), c(x,y,m)=c(x',y',m) and o(x,y,m)=o(x',y',m). This
does
not tell us a great deal. In intervals in which o(x,y,m) is not equal to
o(x',y',m), then one will not necessarily find -- indeed one might be very
hard pushed to find -- that N(x,y,m)=N(x',y',m) and c(x,y,m)=c(x',y',m),
which is why I constructed arrays in which there is no prescribed
determinant of the distribution of occupied ('black') cells. With these
arrays, one can simply state that N(x,y,m)=N(x',y',m) and
c(x,y,m)=c(x',y',m) for a given [x,y] and [x',y'].
|
