Firefox 13 Review

Michael Muchmore

Performance
Before I get into benchmarks, a word about startup time. For the past few years, Firefox has lagged behind the competition in taking longer to get going, especially after a reboot. Firefox 7 went a long way towards remedying this, but Firefox 13 brings the browser within shooting distance of the competition.

On my 2.53GHz dual-core laptop with 3GB RAM, after a reboot Firefox 13's cold start time was 3.4 seconds. While that's hardly an eternity on a hardly screaming fast laptop, Firefox still trails Chrome's 1.5 seconds and IE9's 1.1 seconds, but it's better than earlier versions, which often took over 6 seconds. Safari and Opera held up the rear, with 6.6-second and 9.8 second cold startup times, respectively, so Firefox is no longer the straggler among browsers when it comes to startup time.

A restart of the browser without rebooting made the differences in startup time negligible—Firefox 1.4 seconds; IE9: 0.6 seconds; Chrome: 0.5 seconds; Safari: 1.4 seconds; Opera: 1.5 seconds. Here's a table summarizing the results:

Though Firefox had been much faster at JavaScript than IE8, IE9 changed that picture. With Firefox 9, the tables turned back again a bit, with Firefox's Type Inference speeding up JavaScript performance by analyzing JavaScript code and using this info to generate more efficient code with its JIT ("just in time") compiler.

The results of this optimization show up clearly on the JavaScript benchmarks below, with over 30 percent speed improvements in some cases. I tested with a Core 2 Duo 2.6GHz Windows 7 (32-bit) laptop with 3GB of DDR2 memory. I shut down any unessential processes for three averaged test runs.

On the popular SunSpider JavaScript Benchmark, Firefox picked up a bit, but the difference among browsers is negligible at this point. Most browsers have already optimized to the hilt for this test, so the results of all are pretty closely clustered.

On Google's JavaScript benchmark, V8, Firefox 4 showed a massive more-than six-fold improvement over its predecessor; since then, Firefox 9 showed the biggest improvement among its recent releases, with a 36 percent improvement over version 8. The browser still has some catching up to do with Chrome, though it beats the rest handily:

And on Mozilla's own Kraken JavaScript benchmark, Chrome has actually overtaken the test maker's own browser, Firefox, though not by much. Firefox 9 improved significantly—32 percent—over its predecessor here, and though it didn't add much performance, Firefox 13 is still far faster on this test than IE, Opera, and Safari. Mozilla contends that this benchmark reflects more realistic workloads than the other two JavaScript benchmarks, and it takes quite a bit longer to run.

With Firefox 4, Mozilla added hardware acceleration like that found in Internet Explorer 9. Unlike IE9's, Firefox works on any operating system; IE9 is limited to Windows 7 or Vista. I ran one Microsoft and one Mozilla demo test designed to show hardware acceleration, but keep in mind, the results of this test depend to a great deal on your graphics hardware. The Microsoft demo, Psychedelic Browsing, spins a color wheel and plays spacy sounds, reporting RPM as a result. Firefox does well on this test, beating Chrome and all other comers besides IE. Opera hasn't yet implemented hardware acceleration, and I only saw evidence of it in Safari when that browser was running on a Mac. Here were my results, using a 3.4GHz quad-core desktop with an ATI Radeon HD4290 graphics card:

One final test of hardware acceleration (which, it is worth noting, comes from Mozilla itself): the Hardware Acceleration Stress test, which spins a spiral of photos in the browser window and reports a score in frames per second. This test showed the browsers furthest along in implementing hardware acceleration (particularly for CSS) to good advantage (note the benchmark no longer reports frame rates over 60FPS, as that's the limit of standard LCDs):

Visit page four to read more our Firefox 13 review...