Taco Cowboy's Journal: Journal: Unscientific testing of browsers

Yesterday I did a totally unscientific test on several Window-based browsers.

I came across a page ( http://pastehtml.com/view/bmp0uzp8a.html ) set up by some Ukrainian hackers asking ppl to ddos some Ukrainian government websites in protest of the dns deletion of a Ukrainian file sharing site, or something like that

I used an old Pentium Duo-Core machine, with 4GB of RAM, running WinXP SP3 to test the following browsers:

Google Chrome
Apple Safari
Mozilla Firefox
Opera
Internet Explorer
K-Meleon

Before I ran the test I updated all the above browsers to the latest official version - for IE, I used version 8, which is the last version that XP can run.

For each browser I only ran one tab, with the above page on it, and nothing else.

Before I clicked on the button I used Process Hacker ( http://processhacker.sourceforge.net/ ) to set all the browsers to Realtime Priority â" and for Google Chrome I set all 3 instances of Google Chrome to "Realtime"

Here are the totally unofficial results so far:

Internet Explorer did not permit me to access the page. It told me that that page is dangerous or something.

Opera ran in slow motion. It updated the count once every 20 to 30 seconds, or so

Safari ran slower than Opera, and in about 8 hours Safari froze up.

K-Meleon ran smoothly at first, almost as fast as Mozilla Firefox, but slowed down significantly after 20 hours.

The fastest of the bunch is Mozilla Firefox, about twice the speed of Google Chrome, which came second.

In fact, Mozilla Firefox runs the page so effortlessly that I opened up 3 tabs all running the same thing, they still ran without any hiccup.

Finally, the memory usage, which I checked with Process Hacker -

For Google Chrome, the 3 instances occupied 135 MB of RAM, 80 MB of RAM and 64 MB of RAM respectively.

Mozilla Firefox has only one instance in memory and it used 156 MB of RAM.

Interesting but useless

[mcgrew]

It doesn't matter to me how fast a poorly written page renders; if the html is crap the content probably is as well. Computers' speed, even an old 500mz Dell I'm fixing up for a friend, are enough that on a normal web page there's really no difference. What matters to me is how well the browser adheres to standards, and what its interface is like.

I'v been using FireFox since it was simply Mozilla and had a fire-breathing lizard for an icon.

They're using IE7 where I work, and it's utter crap because it won't