That's interesting Thiago, in the chart JSP ranked slightly better than Tapestry but not as well as play. I'm not sure what the test case is like though but taglibs can easily get out of hand.
I think performance tests should compare complex web pages and form handling because trivial test cases are unlikely to stretch the frameworks to reveal substantial practical differences. John ----- Original Message ----- From: Thiago H de Paula Figueiredo To: Tapestry users Sent: Monday, January 21, 2013 8:28 PM Subject: Re: Tapestry performance On Mon, 21 Jan 2013 17:52:12 -0200, Lenny Primak <lpri...@hope.nyc.ny.us> wrote: > I'm pretty sure tapestry would blow JSF out of the water, just due to > the architecture differences. > As far as JSP, it is very hard to beat because it compiles directly to > java servlet code. I don't think so, as the generated Java servlet code may be slow (itself or by invoking other slow code). As far as I know, running JSP which uses taglibs is slow due to using a pool of instances and other overhead. Some insights by in the messages by Kevin Burton here: http://tomcat.10.n6.nabble.com/The-amazingly-slow-performance-of-JSP-profiler-results-td1971947.html -- Thiago H. de Paula Figueiredo --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org For additional commands, e-mail: users-h...@tapestry.apache.org
