On Jul 3, 4:30 pm, Shawn Hoover <shawn.hoo...@gmail.com> wrote: > Interesting. Maybe I'm crazy, but when running java -server, replacing > vectors/assoc with arrays/aset slowed my project down from 4000 > iterations/second to 100! Similarly, removing multimethods for each > instruction and loop/recur to churn through them and replacing with one big > fn generated up front resulted in a massive slowdown. I haven't profiled > yet, but I plan to do so because I'm so surprised by the results.
Well, I did one more thing: implement the VM core in Java. Only the storage and the instructions. Quite ugly, one class, everything is public static final, and this thing is fast! I get ~65000 iterations/ s. It would be interesting to find out why machine_data[pc] = machine_data[r1] + machine_data[r2]; is so much faster than: (aset-double machine-data (+ (Array/getDouble machine-data r1) (Array/ getDouble machine-data r2))) but I guess, at this level, it would involve comparing the generated JVM bytecode for each case but that is way over my capabilities at this moment. -- IgorR --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
