Slashdot Mirror
Firefox 4 Regains Speed Mojo With No. 2 Placing
CWmike writes "With the release of Firefox 4 Beta 7 this week, Mozilla has returned to near the top spot in browser performance rankings. According to SunSpider JavaScript benchmark suite tests run by Computerworld, the new browser is about three times faster than the current production version of Firefox in rendering JavaScript, and lags behind only Opera among the top five browser makers. Mozilla launched Firefox 4 Beta 7, a preview that includes all the features slated to make it into the final, polished version next year, on Wednesday. Beta 7 was the first to include Mozilla's new JavaScript JIT (Just In Time) compiler, dubbed 'JagerMonkey,' which shot the browser's performance into the No. 2 slot behind the alpha of Opera 11."
On my aging PentiumD/2.8GHz:
No, there's not, because the outcome would inevitably be "every browser is fast (and you won't notice a difference) given enough RAM, CPU power etc". In reality, it's not things like Javascript microbenchmark performance that matter: browsers CAN be faster or slower, but whether you can get a few ms less on Sunspider is going to be lost in the noise. The real differences will stem from other factors.
That's because you're looking at it wrong. JS performance isn't going to improve the current browsing experience, that is perfectly true. The part you are missing is that the JS performance matters for FUTURE browsing.
Why, you ask? One word: Flash. HTML5 exists purely to replicate most of Flash in native HTML. Flash is heavily scriptable and runs complex apps (see any heavily produced Flash game) with passable performance (most of the performance problems with Flash are due to the way the graphics stack was designed [it sucks]). If Canvas, Video and Audio tags are going to replace Flash, the JS needs to be comparable to Flash's ActionScript performance — preferably better to insentivize migration.
Because we're nerds, and we always look for the overly complicated solution to every problem.