AlgoMantra
No matter how many state variables you conjure up, you're going to have to
> look up the nearest neighbour's state variable to calculate the next value.
> In your system, I'm not sure if they are automatons at all. The cells seem
> to behave completely independent of their environment, almost autonomously.
> Comprenez?
>
> What I explained only attempted to address your "update order" question, I
was assuming that you'd fill in your own function for updating which would
look at the environment. In fact the whole point is to keep the environment
constant while all cells base their calculations on it and only *then*
update it. If you re-read my last post I think you'll find I did address the
subject of cells looking at their neighbours.
I wrote it with Wolfram's 1d 2state version in mind but it should work for
Conway's life or anything along those lines as well.
I was assuming all cells are governed by the same rule or set of rules and
so suggesting making those rules a part of the system of updating and not of
a member function of the cells itself. If it's all the same anyway there is
no need to create large and complicated objects taking up memory and
generally complicating matters. That's just a question of the architecture
of the program though, that need not affect how the end result behaves.
Perhaps I was misunderstanding the whole question?
Kas.
Received on Tue Sep 16 2008 - 23:06:41 BST