[ericra]

Just a little experiment I did to test the inefficiency of websites. When looking into this, I found that most people measured website efficiency mainly based on the amount of network requests or just general subjective slowness. I wanted to see what the objective results were when isolating a website and observing CPU usage directly.

[fuzzfactor]

Excellent research.

I think this helps to explain how a particularly enshittified page loads OK, but if you don't close that tab before you move on, it can be trouble later on. Even if you add a number of lightweight tabs and everything is fine when clicking between them, clicking back on the "overloaded" tab tips the balance so much more than it should, that everything freezes.

[ericra]

Thank you.

It surprised me quite a bit that so many resources were being used even though Firefox is not in focus. It makes sense from the browser perspective, though, since you don't want to completely stop a web app when tabbing away. Almost all CPU usage is the result of the active website's design.

It actually made it appreciative of how Firefox uses almost no CPU "passively", and I'm curious if the same thing would be true for Chrome. If you discount memory usage, the browser is surprisingly lightweight.