On Wed, Apr 16, 2008 at 12:10:54AM +0200, Leopold Toetsch wrote: > Am Freitag, 11. April 2008 21:02 schrieb Nuno 'smash' Carvalho: > > Greetings all, > > > > I just posted a little Parrot benchmark in my use.perl's journal > > Just a reminder: > > Please don't use unoptimzed builds for benchmarking.
I agree with Geoffrey that optimized builds should be the default. For example, chromatic posted a blog recently that said: (http://www.oreillynet.com/onlamp/blog/2008/04/multiple_dispatch_now_please.html) Download the new Parrot release next Tuesday, 15 April 2008, then type: $ perl Configure.pl $ make $ make perl6 ... and you too can play with this in your own code. Anyone doing that is likely to get a poor impression of performance. I'd suggest a simpler approach than Geoffrey's: The default 'make' target could default to a reasonably safe portable optimized target, but be overridable by an env var. That way curious end-users and benchmarkers get good performance by default. Possibly not the ultimate, but far better than now. Developers working on parrot (wanting unoptimized/debug quick builds) would just need to set an env var in their .profile, for example, and carry on as now. No changes in their procedures, and no changes in the docs (other than to mention the env var). Then chromatic's recipie works well for all. Tim.
