Firefox 3.6 vs Chrome vs Safari – Javascript Performance

Firefox 3.6 brings with it an updated javascript core, which tries to bridge the performance gap between Firefox and Chrome/Safari . The  performance is virtually unchanged from that of the beta release, which is contrary to what I had expected. I had assumed that the beta release would be an unoptimized release, ideal for debugging (debug symbols left in etc)  but this was clearly not the case.

As usual, Firefox performance on the v8 benchmark is pathetic where Chrome is more than 10 times faster.It is 24% faster than version 3.5.4 in V8 but it is clearly not enough. In the sunspider test, chrome is 2 times as fast as firefox. In this test, 3.6 is 17% faster than 3.5.4. Safari too comfortably beats Firefox in both these benchmarks. In fact, Safari is running neck-to-neck with chrome on the sunspider.

Javascript performance is what matters most in browsers these days as more and more web applications become javascript heavy like Zimbra, GMail, Google Docs and Google Wave. Firefox is sluggish on some of these applications like wave but chrome handles them with aplomb. Firefox performance has a long way to go before it catches up to Chrome and I’m afraid that bridge is growing. Firefox was my favourite browse and I hope it can recapture that loyalty soon.

You should follow me on twitter here

Not surprised

What is the unit for the time it takes to execute sunspider? What is sunspider? What does V8 measure?

On my Mac, when I compare the latest Chrome Dev to the latest Webkit nightlie the Webkit nightlie is almost always 100 ms faster than the latest Chrome Dev using java Sunspider. Can you tell me why are you finding Chrome slightly faster than Webkit? Thanks!

and Opera???? ¬¬

@Rafael: Linked to the sunspider page. It is the most widely used javascript benchmark and was created by the webkit team

@durbrow: I’m not finding chrome faster than webkit. I’m finding chrome faster than safari. Safari does not always have the latest webkit.

@faczever: Opera’s latest release is not yet optimized for OS X as far as i know and is in beta. Once they do a proper release I will include it.

Given that you are already comparing the beta version of Chrome with the release versions of Firefox and Safari, Would it not make sense for completeness’ sake to include the nightly builds of Chromium, and Webkit-on-Safari-4.0.4? Or do these make Webkit seem better, so it doesn’t quite get your point across? :)

@scott: Ideally I would like to include only release versions as development versions maybe unoptimized. Chrome dev channel is as close I can get to a release version which is officially blessed by Google. If apple blesses a certain build of webkit I will be happy to include it. From past benchmarks Nightlies (chromium and webkit) have varying performance. Which is why, they are not included if possible.

Leave a comment

(required)

(required)