Revision: 159
http://pure-lang.svn.sourceforge.net/pure-lang/?rev=159&view=rev
Author: agraef
Date: 2008-06-01 11:29:47 -0700 (Sun, 01 Jun 2008)
Log Message:
-----------
Comment change.
Modified Paths:
--------------
pure/trunk/examples/recursive.pure
Modified: pure/trunk/examples/recursive.pure
===================================================================
--- pure/trunk/examples/recursive.pure 2008-06-01 18:24:53 UTC (rev 158)
+++ pure/trunk/examples/recursive.pure 2008-06-01 18:29:47 UTC (rev 159)
@@ -3,13 +3,12 @@
(http://shootout.alioth.debian.org/). It can be used to measure aspects of
the performance of the Pure interpreter and compiler. 2008-22-04 AG */
-/* Pure 0.1 (release build using gcc 4.1 on Linux) does the benchmark with an
- input value of n=11 in 1m28s on my Athlon 2500+. (This is the user cpu time
- as reported by 'time', which includes startup and compilation times.)
- That's a factor of 34 against the C benchmark on the same machine, which
- puts Pure in about the same league as MZScheme. That's not bad considering
- that Pure is a fully dynamic language. Most of the credit goes to LLVM, of
- course. ;-) */
+/* Pure 0.3 (release build using gcc 4.1 on Linux) does the benchmark with an
+ input value of n=11 in 1m23s on my Athlon 2500+. (This is the user cpu time
+ as reported by 'time', which includes startup and compilation times.) That
+ corresponds to a factor of 31 against the C benchmark on the same machine,
+ which isn't bad considering that Pure is a fully dynamic language. Most of
+ the credit goes to LLVM, of course. ;-) */
ack x::int y::int
= y+1 if x == 0;
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
