Jon Aquino's Mental Garden

Engineering beautiful software jon aquino labs | personal blog

Thursday, December 28, 2006

Page-Load Times for Firefox 2, Internet Explorer 7, and Opera 9.2

IE 7 seemed to be faster than Firefox, so I decided to measure page-load times for the two browsers on four popular websites.

Histogram

My test setup is shown below. None of the three browsers had any extensions installed. Page-load times were measured using IBM Page Detailer on my IBM T43 laptop with a wireless connection to a cable modem, in Victoria BC, Canada.

Screenshot

IBM Page Detailer measures the total page-load time, including all CSS, Javascript, and images. (It can also display a "Gantt chart" of the load times of each of these assets, making it a great for performance analysis).

6 Comments:

  • Correction: Opera 9.1.

    Also, data is below:

    Cute Overload, Firefox 2, 10.6
    Cute Overload, Firefox 2, 10.6
    Cute Overload, Firefox 2, 10.2
    Cute Overload, IE 7, 8.4
    Cute Overload, IE 7, 7.7
    Cute Overload, IE 7, 10.3
    Cute Overload, Opera 9.1, 4.6
    Cute Overload, Opera 9.1, 8.9
    Cute Overload, Opera 9.1, 5.3
    Yahoo, Firefox 2, 4.8
    Yahoo, Firefox 2, 4.4
    Yahoo, Firefox 2, 4.5
    Yahoo, IE 7, 2
    Yahoo, IE 7, 4.2
    Yahoo, IE 7, 3
    Yahoo, Opera 9.1, 4.2
    Yahoo, Opera 9.1, 5.3
    Yahoo, Opera 9.1, 3.5
    Amazon, Firefox 2, 7
    Amazon, Firefox 2, 10.8
    Amazon, Firefox 2, 4.4
    Amazon, IE 7, 8.5
    Amazon, IE 7, 5.6
    Amazon, IE 7, 4.4
    Amazon, Opera 9.1, 5.3
    Amazon, Opera 9.1, 24.7
    Amazon, Opera 9.1, 8.7
    Digg, Firefox 2, 4.8
    Digg, Firefox 2, 8.2
    Digg, Firefox 2, 11.9
    Digg, IE 7, 5
    Digg, IE 7, 7.3
    Digg, IE 7, 7.7
    Digg, Opera 9.1, 5.9
    Digg, Opera 9.1, 8
    Digg, Opera 9.1, 5.9

    By Blogger Jonathan, at 12/28/2006 10:44 a.m.  

  • Ooh. :) My surname is Aquino, too, but I'm from the Philippines. Your blog sounds interesting.

    Happy New Year! and cool stats... ;)

    By Anonymous Anonymous, at 12/30/2006 4:03 p.m.  

  • Elaine Aquino - welcome! Glad you found the stats intriguing.

    By Blogger Jonathan, at 12/30/2006 4:07 p.m.  

  • Except for FF on on Cute Overload and Yahoo (That's the homepage, I assume?), the results seem to vary quite a lot.
    Are the measurement based on a full load after hitting the refresh button?
    Was there any other network activity on the computer at the time, something else using the CPU, or anything else to influence it?

    By Anonymous Anonymous, at 1/01/2007 11:38 a.m.  

  • Hi Yaron - Indeed, there is a second or two of variance in the results. I simply hit Refresh and waited for the page and all its elements to load, as tracked by IBM Page Detailer. This was on my home internet connection.

    By Blogger Jonathan, at 1/01/2007 1:02 p.m.  

  • I was so going to make IE7 my default browser, but there's one thing about it that's a deal-killer for me: if you zoom in to enlarge the text, you suddenly get a horizontal scrollbar that you need to use to see all the text. I believe the IE folks will be fixing this (by adding a "Fit to Width" option), but until then, IE7 zooming is unusable for me.

    And the problem with Opera is that some websites are crippled when viewed with that browser, e.g., you don't get the slider in live.com image search.

    By Blogger Jonathan, at 1/01/2007 9:51 p.m.  

Post a Comment

<< Home