Re: Knight's tour

[Simon Allen]

I am impressed. Would it be possible to incorporate your code into the original service it now runs?

Simon

On Sun, 10 Feb 2019 at 15:45, Doug Stewart <address@hidden> wrote:

On Sun, Feb 10, 2019 at 7:40 AM Doug Stewart <address@hidden> wrote:

On Sun, Feb 10, 2019 at 6:56 AM Simon Allen <address@hidden> wrote:

Very well done.  Yes I will keep Octave in the loop.  Can they be incorporated into the code?

Simon 

We normally bottom post on this list.

Try this.

https://www.dropbox.com/s/oqftcdjq4d8liqg/spiralNumbers.m?dl=0

 

Now that I have spent some time figuring out spiral-numbers, I can understand the code at

 https://oeis.org/A316667/a316667.m.txt    much better.

 With my version of spiralNumbers the move number 1 is at the centre of the square matrix, but

they seem to call that spot x0=0 y0=0. So a simple offset should convert between the 2 systems.

this line

 xy2num = @(x,y) SpiralNumbering(x,y);

is making a function xy2num(x,y)  that if you give it an x,y location

in a spiral array,it gives you back a move number. 

this part remembers what we have done.

 % add current square to sequence
    seqX(end+1) = x0;
    seqY(end+1) = y0; 

seqNum(end+1) = xy2num(x0,y0);

this part

  % apply all possible moves

    x = x0 + dx;
    y = y0 + dy;

gets all 8 new positions.

So I think I am close :-)

--

Simon