Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add point_class_add #16

Merged
merged 3 commits into from Jan 9, 2015
Merged

Conversation

zhuomingliang
Copy link
Contributor

No description provided.

…se it's 2x slow, but Perl 6 multi method version one is the same speed
@japhb
Copy link
Collaborator

japhb commented Jan 1, 2015

(Apologies for the delay, I've been rather sick.)

I'd like a few changes before I merge, if you're willing:

  • Rename the files to luajit-point-class-add-*variant*, where the variant is a one or two word description of what is special about this version of the benchmark (e.g. luajit-point-class-add-named-build-bless for the variant that uses named parameters to BUILD() and blesses in the add() method; try to use equivalent names between nqp and perl6.
  • Merge the big comment blocks at the top of each file into a luajit-point-class-add-README, with all results from all variants represented in the comparison table (and remove all the extra verbiage from the individual files).
  • Use results from the same system in the comparison table; showing your timings on your system versus Mike's values on his system doesn't compare well, especially since he doesn't specify what his system actually is; you should specify what your test system is.
  • Add relevant entries to minibenchmarks.pl.
  • (Optional!) Add perl5 equivalents.

@zhuomingliang
Copy link
Contributor Author

Hello,
I'm not good at naming a good name for these variant, feel free to change to whatever name you like. :P
About timings, I've tested luajit is about 0.22s on my system, so I consider others are similar result. Actually, I don't care others, I only care about luajit/C++, and Java, which has JIT too.
I hope with these benchmarks, NQP/Perl 6 will be as fast as luajit someday. :)

@japhb
Copy link
Collaborator

japhb commented Jan 2, 2015

OK, understood about feeling uncomfortable about the variant naming; I can do that post-merge. Can you do the other things on the list and update the PR? That will save me some time I'd rather spend on some threading tests.

For the README, since you tested the luajit timings on your system, I'd say just separate the timings Mike Pall had from the timings you found, and include your luajit timing in with your NQP/Rakudo timings, so that people can see the two tables are roughly but not exactly comparable.

@zhuomingliang
Copy link
Contributor Author

Hello, I add a new commit, thanks.

japhb added a commit that referenced this pull request Jan 9, 2015
@japhb japhb merged commit 25425c0 into raku-community-modules:master Jan 9, 2015
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants