Will this version survive months of usage with multiple open tabs without performance deteriorating massively like the regular firefox? I left FF after my 10th profile reset / reinstall to get back to "normal" performance.
I'm kind of hoping that Eloctrolysis never becomes main-line. Right now I have a browser that isolates misbehaving web pages (Chrome) and a browser that allows me to have a million open tabs simultaneously (Firefox).
If I could only choose one I'd choose the browser that isolates misbehaving web pages but both is even better.
Electrolysis will initially only have two processes: one for chrome (i.e. UI) and one for web content. More content processes may be used in the future, but we will move very slowly and carefully in that direction. I consider the several-hundred-tabs-open use case to be an important one, and will do my best to ensure it continues to work in the future.
I personally think the number of browser processes shouldn't exceed the number of tabs because you won't get any responsiveness improvements beyond that, and it'll keep the memory usage reasonably low.
Chrome doesn't isolate misbehaving webpages. The pages are grouped together under a few parent processes. The more pages you have open, the larger the group and a page crash will take out many more pages at the same time.
Originally the idea was for every page to be isolated but that didn't scale so well.
In my experience this kind of response was typical a few years ago, but Firefox's performance has improved so much over the past few years that complaints are much less frequent, and when they do occur there is now typically quite a bit of pushback.
Electrolysis will help with some performance problems, but not all.
I haven't followed (and probably don't know nearly enough anyway) details of e10s, but their claim is contrary to your view:
> Performance would improve because the browser UI would not be affected by poor performance of content code (be it layout or JavaScript). Also, content processes could be isolated from each other, which would have similar security and performance benefits.
This is almost certainly caused by an extension. If you use Firefox with few or no extensions (e.g. 1Password doesn't cause problems) you'll find this doesn't happen.
Although badly-written extensions do cause a lot of problems, please don't assume they cause every problem. It's not always true and comments like these give the impression that Firefox developers simply pass the buck every time somebody complains.
Having said that, if someone is experiencing bad performance and they do have extensions enabled, temporarily disabling them is a good diagnostic step. If performance improves, it's clear that it's an extension at fault, and then bisection can be used to work out which one. Otherwise, it's clearly a Firefox problem and a bug report (bugzilla.mozilla.org) would be very helpful!
Isolating a problem is always a good idea – but scope is also a factor: the original claim that Firefox becomes slow simply from normal usage would require everyone at Mozilla not to have noticed during a multi-year performance push. I've never worked there but that seems unlikely.
Pretty much. I never close Firefox on my laptop, and haven't seen any performance degradation for over a year now.
Chrome on the other side, starts degrading my entire system's performance forcing it to hit swap memory the second the string of tabs on my top bar threatens to stretch across the screen. No need to wait for 'months of use without closing'...
