Thanks a lot for taking the time out to explain this stuff in detail! Will go through your solution shortly.
On Sunday, April 3, 2016 at 12:02:48 PM UTC+5:30, Francis Avila wrote: > > I had some fun with playing around with faster solutions. > https://gist.github.com/favila/0573e3f644dea252bdaaed5be9d1519f > > The biggest speedup comes from avoiding set creation in expanded-range > (i.e., the function that produces the collection of affected coordinates) > and ensuring that the ops run on the accumulating set of on-lights using > transients. clojure.set/* functions require both items be sets and does > not use transients internally, so it was much slower. > > Another big speedup comes from encoding the light coordinates more > efficiently. You can encode a light as a number (in my case, a long, with > high bits the x coordinate and low bits the y coordinate) instead of a > vector. This creates fewer objects which are easier to hash. > > Finally, I tried an approach which doesn't use sets, but instead naively > creates a 1000x1000 array of booleans and mutates it in place with every > op. This is the fastest approach: 4 seconds on a 2010-era i3! I'm sure a > proper matrix library (e.g. core.matrix) could do even better. > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
