Beefy Boxes and Bandwidth Generously Provided by pair Networks
Do you know where your variables are?
 
PerlMonks  

Re: 5:10: Why is this slower than 5.8.8?

by eserte (Deacon)
on Dec 24, 2007 at 09:52 UTC ( #658860=note: print w/replies, xml ) Need Help??


in reply to 5:10: Why is this slower than 5.8.8?

Here are my results (amd64-freebsd 6.2, perl5.8.8 from ports, perl5.10.0 self-compiled):
$ perl5.8.8 ~/trash/test.pl Benchmark: running set_inherited_class, set_inherited_object for at le +ast 3 CPU seconds... set_inherited_class: 5 wallclock secs ( 3.14 usr + 0.00 sys = 3.14 +CPU) @ 403854.65/s (n=1268356) set_inherited_object: 4 wallclock secs ( 3.16 usr + 0.01 sys = 3.17 + CPU) @ 441967.29/s (n=1401865) $ /usr/local/src/perl-5.10.0-sysmalloc/perl -I/usr/local/src/perl-5.10 +.0-sysmalloc/lib ~/trash/test.pl Benchmark: running set_inherited_class, set_inherited_object for at le +ast 3 CPU seconds... set_inherited_class: 4 wallclock secs ( 3.15 usr + 0.01 sys = 3.16 +CPU) @ 309119.05/s (n=975657) set_inherited_object: 5 wallclock secs ( 3.08 usr + 0.00 sys = 3.08 + CPU) @ 402004.47/s (n=1237420) $/usr/local/src/perl-5.10.0-perlmalloc/perl -I/usr/local/src/perl-5.10 +.0-perlmalloc/lib ~/trash/test.pl Benchmark: running set_inherited_class, set_inherited_object for at le +ast 3 CPU seconds... set_inherited_class: 6 wallclock secs ( 3.12 usr + 0.00 sys = 3.12 +CPU) @ 378921.38/s (n=1181169) set_inherited_object: 6 wallclock secs ( 3.15 usr + 0.00 sys = 3.15 + CPU) @ 469993.13/s (n=1479744)
perl with sysmalloc seems to be the worst, while the both perls with perlmalloc have similar results.

Replies are listed 'Best First'.
Re^2: 5:10: Why is this slower than 5.8.8?
by Anonymous Monk on Dec 24, 2007 at 16:33 UTC
    You forgot 5.8.8 without perlmalloc (not from ports) which would be the fastest (at least on i386)
      I'm curious to know under what benchmark you ran these tests and what the results were. Also, were there any other flags in the perl build that be factors, i.e. what's the output of perl -V?
      -- Ian Tegebo

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://658860]
help
Chatterbox?
[ambrus]: and they try to use the obsolete features that PHP has to support only for compatibility with old scripts. C++ and PHP both have the problem that people can't forget the past, because when they google "PHP" plus the problme they want to solve, they find b
[ambrus]: ad code examples.
[ambrus]: I'm not trying to recommend PHP, but I think it has way too bad a name because of its past.
[ambrus]: This is different from MS Word, which was already a good editor in the pre-unicode days (in word for windows versions 2 and 6, which ran on windows 3 but also on windows 95), only it wasn't trying to solve the task of writing maths papers back then.
[Discipulus]: ah ok, sounds reasonable; with no fear: Perl all life long
[ambrus]: Mind you, LaTeX is currently still useful for writing math paper or snippet content without styling in such a way that the
[ambrus]: formatting conventions of a journal or website can be quickly applied to it, and MS Office and LibreOffice has not quite solved this (although it's better for this than it used to be),
[ambrus]: which is sort of a drawback compared to the ages of typewritten manuscripts representing content only to which the typesetter applies formatting, but that process required much more manual labor.
[ambrus]: If you want to typeset a manuscript, you can still do much less work then in the manual typesetting ages and get good formatting.
[ambrus]: All with only cheap modern computers and software.

How do I use this? | Other CB clients
Other Users?
Others having an uproarious good time at the Monastery: (11)
As of 2017-09-26 11:10 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?
    During the recent solar eclipse, I:









    Results (293 votes). Check out past polls.

    Notices?