Recently, there has been a few of discussions on some blogs about the memory leaks and performance of IE vs. Firefox (Scobleizer, LifeHacker). This is a complex subject matter because performance depends on the task being executed, and I don't think you can easily define what is the most common tasks for a web browser. Sure, you can say things like how fast it renders a page, or how efficient is its javascript engine, but that is not the reality of the average usage of a browser. To be honest, I don't even think you can define average usage since there are so many people using the browser in their own personal way. Some people open multiple windows, some open multiple tabs, some open multiple instances of the browser. Then there are extensions installed, sites visited, frequency between reboots, etc.
Yesterday, I performed a non-scientific test on IE 6 vs. FF 1.5.
At 4 PM yesterday I closed all browsers' windows on my PC, then opened IE and FF, and navigated both of them to Bloglines for My Feeds. Bloglines now uses AJAX and updates the feed list automatically every few minutes.
When I started the task, the memory footprint for each browser was this:
Firefox: 29.2 MB
IE: 33.5 MB
When I've got to the office this morning at about 9AM, the browser windows were open for 17 hours straight, with nobody interacting with it but the Bloglines AJAX. The result:
Firefox: 37 MB (increase of 27%)
IE: 195 MB (increase of 4,821%!!!)
Yikes! IE sucks... Not quite, IE was pretty bad, but that could be because the Bloglines team used FF to develop their AJAX framework, and never did a good job of testing it on IE and getting rid of the most obvious memory leaks.
So the quest for the answer of who's faster continues...