[berkut]

Exactly the same with me: I tried switching to Firefox 63 last week on a 2015 MBP running MacOS 10.11, and compared to Chrome it seems to use a lot more CPU and thus makes my machine run noticeably hotter.

I guess this might be due to the new Rust layout engine I've heard about which is more parallelised?, but even just having a single tab playing youtube or a gif uses more CPU in Firefox, so maybe it's something with hardware acceleration?

[CmdrKrool]

It's funny how the promise of multicore CPUs was that appropriately engineered workloads could hit new heights of efficiency and overall performance but the cynical reality is that I'm glad for the most part that the partitioning between cores puts a hard limit on how much still commonplace single-threaded processes can take the piss.

[acqq]

> even just having a single tab playing youtube or a gif uses more CPU in Firefox, so maybe it's something with hardware acceleration?

My investigations (their bugzilla, the news about their company, I've been following for a long time) point to the opposite: it's the "old code" which nobody wants to improve: it doesn't help if the "new" code is fast to draw Some detail when the "old" decides to draw or update the stuff that doesn't even has to be updated or drawn. And it's not just drawing too, it's all the processing that happens during the life of the page with the "moving" images. Nobody, as "especially managers in Firefox." There was a number of developers in some Asian country, paid by Firefox, who were in charge for fixing in the old code, and they were simply fired. It seems it's not "strategic" for the managers to "fix" things, only to "experiment" with "new" things like ads or "new technologies" or whatever.

The "more CPU and GPU use" of Firefox is absolutely observable on any platform, not only on Macs. There are also the bugs submitted, but it doesn't seem that there's any priority in fixing them. From their perspective "it works, it's just that those who measure such use notice it." It's of course not so. The battery is empty earlier, the notebooks are hotter, the response lags. There are reported bugs demonstrating exactly what you observed: compared to any other browser even a single video on the "modern" site in Firefox needs much more processing power. It’s easy to reproduce in Windows too.

But obviously fixing the performance problems is not priority for the managers, when there's anything "new and shiny." Because they don’t see that as serious bugs: “the same page uses 10% of the max possible power in Chrome but 20% in Firefox.” “Who cares?” It’s too long-term goal.