My query was related to this bit:
Any live cell with two or three live neighbours survives.
Any dead cell with three live neighbours becomes a live cell.
All other live cells die in the next generation. Similarly, all other dead cells stay dead.
It may not matter on current processors, but on the 68000 or z80, that sort of inefficiency would make a program slow.
Conway's Life is troublesome to program because you need the neighbours from the previous generation, so you can't just itterate over the screen altering stuff as you pass, because if you did you'd change some neighbours that you needed for calculations on cells you hadn't got to yet. My solution was two arrays mapped to the screen, which alternated between being the previous generation and the current one, then write the screen when the current generation was finished. That being so, I think it was more effective to write each cell once, rather than write every cell to dead then write in the live ones.
So, my version would be:
Any cell with 0,1,4,5,6,7 or 8 neighbours is dead.
Any cell with 3 neighbours is alive.
Any other cell (which can only have 2 neighbours) that was alive is alive.
A cell with 2 neighbours that was dead is dead.
There are more lines, but less testing, and it should be faster.
"My way is the way of not fighting":- the book of Enter the Dragon.
