Performance: Firefox vs. Safari vs. Google Chrome vs. Opera
Written by Vygantas Lipskas on December 4, 2008
It’s time to run one of the most popular JavaScript performance tools: SunSpider JavaScript Benchmark.
We will be testing Firefox 3.0.4, Firefox 3.1 Beta 1, Safari 3.2, WebKit r38794, Google Chrome 0.4.154.29, Opera 9.62 and Opera 10 Alpha 1 builds.
Let’s begin, shall we?
Final Results
Lower is better.
Tests were running on fresh Windows XP SP3 and web browsers installations. Results are quite strange though.
Lowest ms - WebKit r38794 (best?)
Highest ms - Google Chrome 0.4.154.29 (worst?)
Be the first to know. Subscribe to our RSS Feed 
Just to note: SunSpider is just one of many benchmarking tools.
Strange though. I also tried the two Operas and the latest Chrome on XP SP3 with Sunspider…
Opera 10a1 was about 50% faster than it’s predecessor, but Chrome was still about 3 times faster than Opera 10…
Indeed, I tried to re-run them but results were about the same anyway.
Sunspider runs its tests about five times, and makes statistic. It SHOULD give the same result every time…
I tried 9.6 and 10 alpha on a winxp Eee pc. 10 was just 6 percent better overall. It was actually worse on most of the tests, and only came out ahead because it lowered the regex time from 1300 ms to 300. Also ran celtickane and both had around 950 ms total. This is not what I expected.
Some relevant reading, by John Resig, for those interested
http://ejohn.org/blog/accuracy-of-javascript-time/
http://ejohn.org/blog/javascript-benchmark-quality/
I see, Internet Explorer would not fit the chart… :)
BTW, I’m pretty surprized to see Chrome at last place.
Very odd results.. Opera was always *the best* in SunSpider tests.
It hasn’t been since the release of Safari 3. And Firefox 3…
Opera 9.62: 6633.2ms
Opera 10 Alpha 1: 5588.6ms
There’s a notable difference with your results.
The results depend on your hardware, OS and other software you have installed and running.
How relevant is javascript performance in my daily browsing?
In other words what percent of browsing time is typically spent executing javascript?
Bill,
Depends on what sites you are using. As I am not a fan of web mails (pop3 is my thing), social networks, etc, etc. js is very minimal here.
[...] Performance: Firefox vs. Safari vs. Google Chrome vs. Opera. [...]
Opera has serious security issues, its easy to get spyware, adware and virus using Opera, thats why I deleted it. I am sticking to firefox, its customizable adblocker, flashblocker and other other security is pretty good and its also qute fast, may not be the fastest, but fast enough and its secure.
@Jim: Are you crazy?
according to Secunia Firefox has 209 vulnerabilities out of which 3% are unpatched. source:http://secunia.com/advisories/product/4227/?task=statistics_2008
Opera on the other hand has 46 vulnerabilities out of which all are unpatched.
Please stop spreading FUD
Also, Opera is really fast at rendering pages. Firefox and Chrome are as well. Just because Safari(webkit) beats the others at this doesn’t necessarily mean that it is the fastest out there. Anyway, wasn’t Google Chrome built on webkit? Strange!
Most of us do not go to javascript intensive sites all the time. On an average page only 4% accounts for javascript. Do a page loading test from numion.com or something. Anyway , this test is run in an artificial environment. Maybe that explains why Chrome fared this badly.
don’t know what you were seeing, but opera has only one unpatche bug!
http://secunia.com/advisories/product/10615/
and then only opera 6 and opera 5 with again one security advisor is open:
http://secunia.com/advisories/product/82/
http://secunia.com/advisories/product/81/
(nobody will uses this software, there are more holes, because of found new lecks in later software, unpatched in old!)
–> only one unpatched secutity bug. what did you see?
@FasterOPera, you’re not understanding what’s being discussed. Not sure if you’re trolling or confused, but maybe I can help.
First, yes Chrome is built on top of WebKit, but that’s just the browser/HTML/rendering engine. WebKit separates the JavaScript implementation from the browser, so Google chose to replace the WebKit’s JavaScriptCore with their own JS engine: V8.
Second, as for JavaScript only accounting for 4% of “an average page,” again we’re not talking about the part of the browser that parses HTML. We’re talking about the runtime engine that interprets and executes the JavaScript code. That lives in its own box in the browser (JavaScriptCore for WebKit, V8 for Chrome). That’s what is being tested. How quickly it executes the tasks that it’s being asked to do is what we’re interested in. A lot of sites do use a lot of JavaScript code, especially for the smooth Web 2.0/Ajax interactions that you see with all the major web-based mail clients and many other sites.
here is the independent speed test
i think they are right
http://www.evilscience.org/internet-explorer-vs-firefox-vs-opera-vs-chrome-vs-safari/