I mean also the fact that the browser most people used at that time was bad may have been a factor. Chrome was a noticeable step up from IE.

For me, Phoenix/Firebird/Firefox was a noticeable step up from IE, and Chrome was a noticeable step up from Firefox. Chrome did especially well relative to Firefox when I had a lot of tabs, and for JS-heavy sites.

I bounced back and forth several times:

* I tried it for a month in January 2010, and went back to Firefox because Chrome was slower and less stable: https://www.jefftk.com/p/google-chrome

* I tried it for ~6m in 2011, and switched back to Firefox because I missed side tabs: https://www.jefftk.com/p/side-tabs

* In early 2012 I switched back to Chrome, despite the lack of side tabs (which I still miss dearly) because it was enough more performant.

(Disclosure: I work for Google)

Try switching once more, and turn on Webrender this time - it's still beta, but I had no issues and it made my FF Chrome-fast.

Several reasons for not switching back, some idiosyncratic:

* I trust Chrome more than Firefox on security

* I'm used to Chrome's debugging tools, and would need to learn Firefox's

* I use the Chrome password manager, and it would be annoying to switch

If Firefox were enough better I would still switch, but at this point my understanding is it's more like "Firefox is no longer slower than Chrome" not "Firefox is a step up from Chrome in the way Chrome once was from Firefox".

That's understandable though I imagine it's unlikely we'll see something like that ever again, right? To me it seems kinda like how race car engineers used to be able to discover some major advancement in their engines that would give them wins for a few years but nowadays it's much more homogenous and worked upon for much less significant gain. I don't know much about the deeper webdev stuff though so who knows, but it seems like that could only ever really possible with a shift in a browser trying some novel shift in architecture.