On Wed, 2003-11-12 at 15:31, Stas Bekman wrote: > At the moment it looks like if you build your perl with -Uuseshrplib > -Uusethreads and don't enable debugging, and use a good selection of > optimization flags (like Mandrake 9.2's perl does), you will get the > best performance.
Whee! Benchmarking is fun! Those are interesting numbers, but can this really be telling us that compiling with threads gives better performance? I guess we need to isolate that one option to get the answer. It was just pointed out to me by Sam Tregar that perlbench is heavy on math, and therefore not representative of what most of us do in our apps. It looks like you can replace the stock tests with your own, and I'd like to try it with a test that selects something from a database and runs it through a template. Sam also pointed out that the Intel compiler has a free evaluation download... When I get a little time,.I'll give these a shot. - Perrin -- Reporting bugs: http://perl.apache.org/bugs/ Mail list info: http://perl.apache.org/maillist/modperl.html