Skip to main content

Web Browser Grand Prix: Chrome 20, Opera 12, Firefox 13

Test Suite And Methodology

Big Changes

We replaced GUIMark 2 Flash, GUIMark Java, and Encog Silverlight with five tests from RIABench Flash, Java, and Silverlight. We also added RIABench JavaScript to our JS testing, a long overdue replacement for JSBenchmark and V8. While RIABench consists of ten separate tests, two of the tests aren't available for all four technologies, one is currently spitting up a 404 error, Safari has a problem with one, and IE9 has a problem with another.

We had to temporarily replace Dromaeo DOM with Acid3 due to the former having issues with the WebKit-based browsers under Windows.

Another CSS test was added to balance out the Maze Solver CSS3 benchmark.

The Mozilla Hardware Acceleration Stress Test was replaced with WebVizBench, which has no highest score limit.

We've added a general responsiveness observation to the 40-Tab Memory Usage test.

And finally, we found a security test that is still relevant and not passed by any of the contenders.

Our tests are no longer placed into the core, observation, dated, and quarantine groups. With the massive refresh to the benchmark suite and the introduction of composite scores to cover every category of testing, this is simply no longer necessary.

Web Browser Grand Prix Test Suite v11

The table below lists all 34 benchmarks (consisting of 66 individual tests) currently in our suite, along with the version number and link (where applicable), and the number of iterations performed.

Benchmark NameIterations Performed
Performance Benchmarks (24 Benchmarks, 56 Tests)
Cold Startup Time: Single Tab3
Hot Startup Time: Single Tab3
Cold Startup Time: Eight Tabs3
Hot Startup Time: Eight Tabs3
Uncached Page Load Times (8 Test Pages)5
Cached Page Load Times (8 Test Pages)5
RIABench JavaScript (5 Tests)3
Mozilla Kraken v1.12
Google SunSpider v0.9.1 Mod2
FutureMark Peacekeeper 2.02
Acid35
Maze Solver5
CSS Stress Test and Performance Profiling - Tom's Hardware2
GUIMark 2 HTML5 (3 Tests)3
Asteroids HTML5 Canvas 2D And JavaScript2
HTML5 Canvas Performance Test2
Facebook JSGameBench v4.12
Psychedelic Browsing2
WebVizBench2
Mozilla WebGL FishIE2
WebGL Solar System2
RIABench Flash (5 Tests)3
RIABench Java (5 Tests)3
RIABench Silverlight (5 Tests)3
Efficiency Benchmarks (4 Benchmarks/Tests)
Memory Usage: Single Tab3
Memory Usage: 40 Tabs3
Memory Management: -39 Tabs3
Memory Management: -39 Tabs (extra 2 minutes)3
Reliability Benchmarks (1 Test)
Proper Page Loads3
Responsiveness Benchmarks (1 Test)
General Responsiveness Under Load3
Security Benchmarks (1 Test)
BrowserScope Security1
Conformance Benchmarks (3 Benchmarks/Tests)
Ecma test2621
Peacekeeper 2.0 HTML5 Capabilities1
HTML5Test.com1

Methodology

We restart the computer and allow it to idle before benchmarking. Most of our final scores are an average of several iterations. More iterations are run for tests that have short durations, lower scales, and/or higher variance. Any obvious outliers (usually network hiccups) are removed and retested.

Individual detailed methodologies and information regarding composite scoring is described on the corresponding benchmark pages.

  • mayankleoboy1
    1.what the benchmarks dont show is that in Firefox , if a tab has a heavy page with a lot of CPU intensive workload, the complete browser UI starts stuttering. That means the browser UI is on the same thread as the page loading.

    2. in the 40 tab test, try working in a tab during the loading of the 40 tabs. you will find lots of difference between browsers. FF hangs, Opera and Chrome remain fluid.

    3. how about a test where a browser is using 1GB+ RAM and you are trying to open/close tabs. Then see the UI responsiveness. most browsers can easily handle 800MB RAM. but which browser easily handles 1.2GB+ RAM ?
    Reply
  • mayankleoboy1
    IE9 does so good on HTML5 HWA accelerated benchmarks because its able to offload more of the processing to the GPU.
    i tested this and found that during a HTML5 benchmark, IE9 had the least CPU usage, and most GPU usage amongst all the browsers.
    Reply
  • mayankleoboy1
    How many firefox users dont use ADblockPlus ? very very less.Also ABP developer is a regular contributor to the Firefox source code.
    maybe you should do a few memory benchmarks with ABP installed just to realistically judge what 99.99% of FF users go through.
    Reply
  • ben850
    WTF Chrome is already on 20?? It ninja updated to 19 just a few days ago..
    Reply
  • lethalsam
    i won't ever use a browser a browser WITHOUT AD BLOCK Plus. (ABP)

    ABP works wonderful on Firefox, i RARELY see any ad. While I have used ABP on Chrome BUT its doesn't block half the ads.
    I know its Not Google's fault, its just that ABP developers are putting more effort with Firefox.

    So for me, Firefox > Chrome.
    Reply
  • adamovera
    @mayankleoboy1: 1+2) Interesting, I'll be looking for that next time 3) That would require a different workload for each browser.
    IE9 does so good on HTML5 HWA accelerated benchmarks because its able to offload more of the processing to the GPU. i tested this and found that during a HTML5 benchmark, IE9 had the least CPU usage, and most GPU usage amongst all the browsers.Really interesting, what utility do you use for measuring GPU usage?
    How many firefox users dont use ADblockPlus ? very very less.Also ABP developer is a regular contributor to the Firefox source code.maybe you should do a few memory benchmarks with ABP installed just to realistically judge what 99.99% of FF users go through.I'd estimate ABP usage on FF at around 5% or less based on ABP and FF usage statistics. Besides, that would give FF an unfair advantage.
    Reply
  • mayankleoboy1
    Really interesting, what utility do you use for measuring GPU usage?

    MSI afterburner for GPU. windows task manager for CPU.

    @mayankleoboy1: 1+2) Interesting, I'll be looking for that next time

    i sent a mail regarding this to Chris. but maybe i sent it too late for this article...
    Reply
  • mayankleoboy1
    if you open multiple tabs together in chrome, it can use each CPU core for each tab. so if you have a quad core, and you open 4+ tabs together, the CPU usage will be 100% (using all 4 cores) during the tab loading time.
    but if you run 4 instance of dromaeo in 4 tabs, the CPU usage is still 25% (using only 1 core).
    so chrome is not completely multiprocessing.

    in IE10 beta, if you run 4 instances of dromaeo benchmark in 4 tabs, it uses all the for cores. so we can expect better multiprocessing from IE10 and win8 :)
    Reply
  • adamovera
    @mayankleoboy1: I got that email yesterday or the day before, this article was completed a few days before that. Sorry, my inbox usually gets a few pages deep after a doing long benchmark-heavy article.
    Is Dromaeo (the DOM portion) working in Chrome for you? I could not get it to finish in Chrome or Safari on any of my Windows machines.
    Reply
  • mayankleoboy1
    i ran the javascript benchmark that ran fine. Didnt run the DOM benches.

    BTW, i run chrome dev version. so that could make a difference.
    Reply