On Fri, Jan 5, 2018 at 5:42 PM, Hendrik Boom <hend...@topoi.pooq.com> wrote:
> Long ago in the 60's, on an old PDP-11 running an early Unix, I > implemented a Live using quad-trees for storage after I noticed by > drawing on a lengthy printout that at each level of the quadtree, only > about half of the quads has objects in them. The empty ones I could > leave out of storage and not do much computation with them (except at > the edges in case they wee to come back to life). > > I could get simply *huge* Life patterns on this rather small machine (by > today's standards) without running out of RAM. I did have to rescale > the screen display several times. A pixel ended up meaning something > like "there is something in this 8x8 block". > > Well, what had to be used as a pixel on an ascii-text-only terminal. > Must have been incredible at the time! Can you tell how close to the 1980's HashLife algorithm [1] your implementation was? It seems quite related at least. The first time I saw these exponential speed ups, I was blown away. People seeing your implementation likely felt similarly. I would love to see a clean Scheme/Racket implementation of HashLife. [1] https://en.wikipedia.org/wiki/Hashlife, impressively implemented in Golly [2] [2] http://golly.sourceforge.net/ -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.