Pavlov's post is awesome, and I've seen many questions on blogs, in comments and on, while Shaver and Pavlov and Schrep and others have done a great job answering most of them. Here is a quick and dirty FAQ:

Memory usage in Firefox 2, Firefox 3 and IE7

Memory usage in Firefox 2, Firefox 3 and IE7

Question: What about IE8?

Pavlov's answer:

We attempted to run IE8 through the test but had problems with it showing broken images, giving error pages and freezing up while running the test. We had similar problems with Safari. It seemed to follow the IE7 line pretty closely up to about 350MB until it crashes very early on.

Question: What about Safari (on Windows, since the test runs on Windows)?

Shaver's answer:

We tried to graph it against Safari 3.0.4b on Windows: it looked very much like the IE7 curve until it got to about the 350M mark, and then it crashed, every time, even after we reduced the test load sufficiently to let IE7 complete it.

Question: What about Opera?

Schrep's answer:

Since folks asked I ran the latest Opera 9.5b in the exact same environment. It peaks around 240MB and doesn'tt free up any memory at the end (so ends at 240MB). Performance during the run is similar to Firefox but higher than Firefox at the end. It is significantly higher than Firefox 3 - which peaks around 220MB and ends at 85MB.

Question: What about using tabs instead of Windows, would that change anything?

Shaver's answer:

with tabs our results were better (about 10M less in use at the end). Unfortunately, IE7 crashed if you churned tabs instead of windows, and we'd already weakened the test pageset a lot to get IE7 that far, so we left it at that for now.

Question: Why not using jemalloc on Mac OS X?

Shaver's answer:

We're not using jemalloc on OS X in beta 4, and we're not likely to in Firefox 3 final. The win was much less there, especially on Leopard, where the allocation behaves very much like jemalloc itself.

Question: what does this mean in a mobile context? (Ok, the question was asked by Chris Blizzard :-) :

Blizzard's answer:

It’s pretty simple, really. What it shows to anyone who looks is that we’re able to hit the kinds of memory and performance requirements that mobile platforms demand. Along with that we’re able to bring our full platform, excellent web compatibility, a single source code base, a committed organization and a strong brand and identity and make that available to partners and users. Users who use our software on mobile devices can expect web sites that just work, access to add-ons all balanced against the hardware limits imposed by mobile devices. In essence, we can bring that no compromises approach to mobile, just as we’ve done it with the desktop. And Beta 4 is the proof of that.