Firefox 55 reduces startup time by 90 percent and memory usage by 75 percent

[Redaktion]

A Mozilla developer has tested multiple versions of the browser and found that the current beta version improves startup times by over 90% and reduces memory usage by 75% when compared to the latest stable version. While these results are specific to the extreme test profile and operating system used, they show that we can expect a positive performance improvement from the next release of Firefox.

https://www.notebookcheck.net/Firefox-55-reduces-startup-time-by-90-percent-and-memory-usage-by-75-percent.237000.0.html

[Gleb]

Nice try Mozilla, but no ;D

[Jahn]

Ah, good ol' fanaticism: the bane of knowledge and reason, the holy grail of marketing.

[Craig Ward]

No doubt that I'm a fan of Firefox (which I made clear in the article). However, I mention that those performance gains are specific to the test scenario used, and I also showed that I'm skeptical about the level of improvement when used with Windows memory management and when the web pages are actually rendered. The figures presented were true and accurate for the test scenario.

Certain websites that I visit seem to bloat Firefox quite quickly, and I can end up with 2 GB of RAM used with only 25 or so tabs open. While I don't expect my memory usage to suddenly go down to 500 MB after the update, it would be very surprising if there wasn't some improvement from FF54 to FF55 based on the developers test scenario.

[tienku]

I'm not satisfied with Firefox 55. I constantly get the message: a web page may be slowing down your browser. Other times, when accessing any websites Firefox lags showing the clock-like wheel and the ventilator works at its maximum as if Firefox were consuming too much ram and I can't use FF until it fixes itself.

The only improvement for me is that startup is much faster now, but overall performance has devolved.

[Craig Ward]

Tienku - That's not good :( I daily use I've found that memory usage is lower for me in Firefox 55, but not to the extent found in the test reported on here. Unfortunately I think it shows the limitation in the testing method used by the developer where he wasn't connected to the internet so no websites actually loaded. The way in which the browser is handling webpages has a big impact on performance.

I haven't been getting any errors that I wasn't previously getting (some websites still cause memory leak problems).