Posted by anisotropic on June 14, 2007 at 7:14pm
Ran across this 'speedtest' page today, it has interesting results especially cross-browser:
http://mootools.net/slickspeed/
If you run it on FF ( at least on my iMac and a nearby Vista machine ) moo and prototype are much faster, but extjs wins hands down on Safari ( OS X ) and IE7 on Vista. jQuery is middling IMHO the benefits in developer satisfaction are worth the speed cost vs prototype.
Discuss.
Comments
Surprising
I was surprised how many times they returned different result sets.
newbie thoughts
I have only tried jquery and my experience wasn't brilliant - it didn't work correctly in XHTML+SVG mixed namespace pages. I also came across mootools whilst evaluating MODx for a specific project.
Both jquery and mootools offer loads of eye candy. I think on the basis of only a cursory walk through that I prefer the mootools API over jquery.
My 2p
Rick
jquery ? eye candy ? where ?
jquery ? eye candy ? where ?
heh: .hide() | .show() == eye candy?
I don't really see jQuery as being eye candy heavy compared to moo or something like Yahoo. The lesson I took away from this especially when doing the cross-browser comparison is that not all JS frameworks work the same way on all browsers. I almost see this page as the start of some sort of compliance test framework that each library can use to validate not only their speed but their functionality on different browsers. example: extjs seems to have some very fast performance on some test, but they completely fail others. It seems to me like moo/prototype is optimized for Firefox?
@rbeton: do you have a simple example of where jQuery failed in a mixed namespace page? I've used it for simple things in mixed XUL/HTML pages, but I can imagine your case being something Ressig never considered a common use case.
As well, I've heard that the JS implementation in FF 3 is going to change over to Adobe's recently contributed JS vm, which may change things for JS libraries.
interesting
I was really surprised to see the difference in results for prototype between IE and Firefox. I'm not talking about the speed of the queries but what the queries actually found.
Update, jQuery 1.2 has been released
There seem to be some great improvements, according to the same 'slickspeed' benchmark:
http://dev.jquery.com/~john/slick/