When I replicated the timings given at the end of this chapter, their
relative rankings were drastically different from what is shown in the
essay, so I added the following note:
*[Be aware that the relative timings may differ for different versions of J
depending on code optimizations.] *

For anyone who is interested, the two sets of relative timings are:
   time each '9 comb 23';'9 comb2 23';'9 combi 23'  NB. From Ch. 34
+--------+--------+--------+
│0.251543│0.173648│0.199808│
+--------+--------+--------+
   (];~2%/\]) 0.251543 0.173648 0.199808
+----------------+--------------------------+
|1.44858 0.869074|0.251543 0.173648 0.199808|
+----------------+--------------------------+
   (];~2%/\]) >(10) time each '9 comb 23';'9 comb2 23';'9 combi 23'  NB. On
my machine
+--------------+------------------------------+
|1.26964 425.37|0.108453 0.0854202 0.000200814|
+--------------+------------------------------+
   JVERSION
Engine: j804/j64/windows
Release: commercial/2015-12-21 16:18:48
Library: 8.04.15
Platform: Win 64
Installer: J804 install
InstallPath: c:/users/devon_mccormick/j64-804
Contact: www.jsoftware.com


-- 

Devon McCormick, CFA

Quantitative Consultant
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to