The Speed of Eval in Some Lisp Implementations.

moved to kazimirmajorinc.com

9 comments:

  1. that's a lot of Lisp programs you have there!

    times here on an iMac are:

    622
    307
    676
    757

    ReplyDelete
  2. Very nice, but... WTF is each numeric column?
    A table with no references is worthless.

    ReplyDelete
  3. The numbers are evaluation times of four expressions, given above in Newlisp, Scheme, Common Lisp and Clojure on my PC. Other dialects might have slightly different code.

    ReplyDelete
  4. ran into this post. was looking for the date the article was written...

    ReplyDelete
  5. Xah, the first version is written 28 Dec 2008, but new items are added continuously; for example, Clojure 1.2 was added yesterday.

    ReplyDelete
  6. i hate it when i couldn't find the date a article was written on the web. Did you intentionally didn't want to include that? or i missed it somewhere?

    ReplyDelete
  7. Yes, I intentionally didn't included that. But I forgot why I did it, it was long time ago. Now I only assume that I have good reason. At least it is visible in address of the page:

    http://kazimirmajorinc.blogspot.com/2008/12/speed-of-newlisp-eval-test-v100.html

    ReplyDelete
  8. Kazimir, could you add Stalin to your set of tests? Also, please consider testing Chicken and the latest update of Racket.

    ReplyDelete
  9. Mark C, I just did it for Racket and Chicken; not for Stalin because it doesn't support EVAL. Also, there are two different results for Chicken - for interpreted code, using csi, and compiled, using csc.

    ReplyDelete