November 21st, 2007

Memory test - Firefox 2.0.0.9 vs Firefox 3.0 b 1

Posted by Adrian Kingsley-Hughes @ 12:40 am

Categories: Software

Tags: Mozilla Firefox, Memory, Beta, Web Browser, Web Browsers, Internet, Adrian Kingsley-Hughes

Does Firefox 3.0 beta 1 do a better job of handling memory than earlier versions?  In a test put Firefox 3.0 beta up against Firefox 2.0.0.9 in a series of tests.

Before I go any further, a few disclaimers and notes.  First off, I’ve carried out this test on a single system running Windows Vista Home Premium on which Firefox had not been previously installed.  The system has 2GB of RAM.  Both Firefox 2.0.0.9 and Firefox 3.0 b 1 were installed fresh using a standard install.  For each test I visited the same web pages and did my best to make the browsing the same on both versions.

OK, with that out of the way, on with the tests.  I simulated three different browsing scenarios:

• Loading a five pages into the browser
• Loading a single page and leaving the browser for 10 minutes
• Loading 12 pages into the browser and wait 5 minutes

Here are the results:

Firefox 2.0.0.9

• Loading a five pages into the browser - 35,640KB (img)
• Loading a single page and leaving the browser for 10 minutes - 47,852KB (img)
• Loading 12 pages into the browser and wait 5 minutes - 103,180KB (img)

Firefox 3.0 b 1

• Loading a five pages into the browser - 38,644KB (img)
• Loading a single page and leaving the browser for 10 minutes - 63,764KB (img)
• Loading 12 pages into the browser and wait 5 minutes - 62,312KB (img)

Check out the complete screenshot gallery.

Just to give us a baseline, I repeated the 12 page test using Internet Explorer 7 and found that the browser used 89,756KB (img), more than Firefox 3.0 beta 1 but substantially less than Firefox 2.0.0.9.

This is interesting.  Initially it seemed that Firefox 3.0 beta 1 was consuming more memory than 2.0.0.9, but during the twelve page test I started seeing what I’ve been seeing before - spiraling memory consumption when the browser is under significant load.  I’m certain that if the browser had been left open longer, memory usage would have continued to rise.  I didn’t see much signs of Firefox 3.0 beta 1 doing this.  Certainly based on this test and from using Firefox 3.0 beta 1 today, I do think that things have significantly improved.

Anyone else taken Firefox 3.0 beta 1 for a spin?  Any thoughts, feelings or observations?