Hacker News Reader

Comment by luhn | Hacker News Reader

I attempted to switch to Firefox earlier this year, as part of a personal exodus from Google. I ended up switching back to Chrome because I couldn't tolerate the poor performance.

I'm giving it another shot now that Electrolysis is there. I've only been using it for half an hour now, but my initial experience has been promising.

View on HN

Welcome back! Fair warning, legacy add-ons can perform worse under multi-process Firefox, since any synchronous APIs now require cross-process locking. To the greatest extent possible, try to limit yourself to add-ons badged with "Compatible with Firefox 57+" on addons.mozilla.org, which are guaranteed work well under Electrolysis.

Also, get ready for Firefox 57. Just one example, in one benchmark, but the last data point on the right is accurate: https://screenshots.firefox.com/1DYGktRBpiIJBc7T/arewefastye...

What particular change is responsible for the performance improvement seen in that benchmark? Not anything related to Quantum/Servo surely; I expect those benchmarks to reflect Javascript performance exclusively, and I haven't heard anything in a long time from Mozilla regarding Spidermonkey improvements.

The change in that benchmark was from https://bugzilla.mozilla.org/show_bug.cgi?id=1369140, as part of the Quantum Flow (https://wiki.mozilla.org/Quantum/Flow) effort.

That particular benchmark stresses DOM and layout performance, that test just programatically clicks all the delete buttons in the page under test.

There's been a steady march of performance improvements in Spidermonkey, though they haven't gotten very much press. They've been rearchitecting most of the inline caches so there's more sharing between the baseline and optimizing compiler and less in the way of performance cliffs. See https://bugzilla.mozilla.org/show_bug.cgi?id=1259927 and its dependencies (https://bugzilla.mozilla.org/showdependencytree.cgi?id=12599... for the dependency view). Things like https://bugzilla.mozilla.org/show_bug.cgi?id=1328140 led to significant improvements to the sort of megamorphic code you commonly see in frameworks (e.g. https://bugzilla.mozilla.org/show_bug.cgi?id=1328140#c14 shows 25% improvements on Ember benchmarks).

This work has been ongoing; a bunch of it landed in 53, more in 54, and more has been landing since. What makes it a bit harder to advertise is that it's aiming at reducing performance cliffs and improving the chance that real-life code lands on fast paths, not on improving the performance of those already-fast paths. WebAssembly is where the "make the fast bits faster" action is, mostly.

I looked for an ad-blocker badged "Compatible with Firefox 57+" and noticed ABP and uBlock Origin didn't have this badge.

I was hoping FF would get something like Safari's Content-Blocking Rules. For such a common use case its an area ripe for optimization.

There is a webextension version of ublock origin. I'm not entirely sure ff 54 supports all the needed APIs.

uBlock0.webext.xpi @ https://github.com/gorhill/uBlock/releases/tag/1.12.4

Edit: I've disabled the addon signing requirement and all seems to be working. Hopefully it gets on the AMO soon.

ublock will work fine with e10s. It's not a webextension yet, but it is e10s compatible already

Firefox 57 will no longer support XUL extensions at all. It's due to be released in November 2017.

ublock is actively maintained, and afaik they have been working on a webextensions version, so I assume they'll be ready by then if not soon after.

In the meantime ublock works fine and is e10s compatible.

Is there a way to search for only 57+ addons? I took a quick gander and didn't see anything.

Yep: https://addons.mozilla.org/en-US/firefox/tag/firefox57

With 30+ extensions installed, It would be nice to list incompatible extensions in the about:addons page. From about:support page, I see Multiprocess Windows 0/1 (Disabled by add-ons)

You might want to install the Add-on Compatibility Reporter available from https://addons.mozilla.org/en-US/firefox/addon/add-on-compat...

thanks for recommending this.

It lists dozens of incompatible installed add-ons such as cookie monster, Live Http Header, UserAgent Switcher, Vimporator etc.

I wish there was an easy way to see which of my currently-installed addons are not compatible with Firefox 57+, instead of having to look at which I have installed and searching for each of them individually on addons.mozilla.org.

I did search and it looks like HTTPS Everywhere, LastPass, and RescueTime aren't compatible. Privacy Badger is compatible, which may satisfy people looking for an ad blocker.

Try out Firefox 55 (Beta or Developer Edition). It puts a big "Legacy" badge next to each add-on that is not a WebExtension in about:addons.

Hah, every one of my addons is legacy. FireGestures, LastPass, Session Manager, Tab Mix Plus, and uBlock Origin.

LastPass and uBlock Origin have Chrome versions so I have some degree of confidence they will be ported over.

FireGestures is absolutely essential to me, and no confidence there. Once Mozilla stops supporting XUL extensions I guess I'll use a Firefox fork or switch to Vivaldi. After... geez... 13 years of using Firefox. They finally forced me away.

Hopefully Vivaldi supports customizable mouse chord gestures by then.

There are already several gesture add-ons that are compatible with Firefox 57 and beyond: https://addons.mozilla.org/firefox/search/?q=mouse+gestures&...

Foxy Gestures in particular claims to be deeply inspired by FireGestures: https://addons.mozilla.org/firefox/addon/foxy-gestures/

I'm dubious, because after nearly a decade, Chrome has exactly zero mouse gesture addons that I consider remotely usable. None of them work consistently. Not a single one is comparable to FireGestures.

But thanks for the link to Foxy Gestures-- I'll give it a shot!

Edit: Tried out Foxy Gestures. Not usable. First, no support for mouse button chording. Second, every single gesture pops up the right-click menu. Third, gestures don't work maybe 75% of the time-- difficult to tell if that's due to the right click menu popping up or if it's just plain broken.

Seems like it's this bug at fault. Mouse gesture Webextensions are broken on MacOS and Linux.

https://bugzilla.mozilla.org/show_bug.cgi?id=1360278

We're still nearly half a year away from Firefox 57's release, so many add-ons, like Foxy Gestures, are still going to be a bit rough. There's no harm in sticking with FireGestures for now, knowing that replacements are in the works.

(Though it's probably worth letting the Foxy Gestures author know what bits seem to be missing as a FireGestures user.)

Use Firefox ESRv52 with XUL support until June 2018 https://www.mozilla.org/en-US/firefox/organizations/ Pale Moon started a fork of the current XUL-based Firefox. Hope it will be popular. I hate XUL but a lot of people love it. I hope it will live with Pale Moon. Why would you use Vivaldi? A browser is a very personal tool, why use proprietary software for it?

> switch to Firefox earlier this year ... I couldn't tolerate the poor performance

Everyone still seeing poor performance in Firefox 54, I personally would recommend to do a Refresh: https://support.mozilla.org/en-US/kb/refresh-firefox-reset-a... because usually it is an old add-on or wierd setting that causes it. At least it was in my case, and after Refresh firefox was running very fast

I don't understand what you people are seeing who make these performance complaints. I switched to Firefox last year because Chrome was giving me a lot issues (often hanging and crashing when I had many tabs/large pages open). So far Firefox has been working like a dream for me.

I've often read FF is faster, HN is one of the few places I see Chrome being called faster. Maybe it depends heavily on the number of tabs and which sites are loaded? For me FF, whenever I tried switching, was always way slower and prone to freezes. But then I have 32GB Ram, 5 tabs always open (pinned), between 3 to 20 extra tabs at the same time and quite the number of extensions.

I really wish I could agree.

On a core i7 with 16GB and a completely fresh Windows 10 install that has never had Firefox on it (so we can't blame an old profile on the machine), Firefox with a single plugin (Keefox) is noticeably slower than Chrome. Just logging into my Google account I saw surprising amounts of basic UI stuttering. Something like Inbox (which is admittedly a pig) loads slowly and scrolls sluggishly.

Chrome, by comparison, is responsive and snappy.

I wish it were different. I really want to like Firefox, but on this machine, with some very basic tests it's clear that, for me, it simply does not compare...

Some legacy addons are poorly written. Have you tried a fresh Firefox profile with only uBlock Origin on it?

I've switched to Firefox full time (unless using F12 tools) - Electrolysis helped a lot. It has gotten very much faster and snappier, while Chrome feels laggy, particularly when closing or switching tabs. I've tried disabling all extensions, but it didn't help.

I've been using e10s for quite a while now. Yes, its great that they have reduced memory consumption by a huge margin. Unfortunately, I still feel page load times, scrolling, overall snappi-ness of chrome (I use ungoogled chromium) or overall browsing experience are way better than Firefox.

What about Chrome's dev tools do you find preferable? I tend to find myself hampered by lack of functionality like easy replay of network requests, and don't spend a lot of time in Chrome's toolset for such reasons - I'd be interested to hear a different perspective.

At least for my case, I found that Firefox debugger would sometime hang for a bit. It was slowing me down. I'll definitely give it a shot on my next JavaScript heavy project.

I had the same trouble with complex applications in Firefox 45, but haven't encountered it since then - 51 and since have been much more performant in general, and that goes for devtools as well. (I suspect it may have something to do with https://wiki.mozilla.org/Electrolysis/Release_Criteria/Jank#... - but that's just a guess.)

In my case doing a "refresh" did help: https://support.mozilla.org/en-US/kb/refresh-firefox-reset-a...

To confirm that Electrolysis is being used for you, open a new tab and go to about:support, and double check what "Multiprocess Windows" reports.

If it still doesn't work for you now you may want to give it another try once Project Quantum is further along the way (a year?):

https://medium.com/mozilla-tech/a-quantum-leap-for-the-web-a...

Servo looks super fast, but they don't seem to want to just rip-out Gecko and put Servo in there, just replace bits and pieces over time. I think that's a mistake, despite the hurdles of doing that, and especially if the end result in terms of performance is something like 60-70% of Servo as opposed to 100%.

https://www.phoronix.com/scan.php?page=news_item&px=MTgzNDA

https://www.phoronix.com/scan.php?page=news_item&px=Google-S...

> they don't seem to want to just rip-out Gecko and put Servo in there

It's hard enough to change the wheels of a car while driving on the highway, you really don't want to do all four at once.

Car analogies (and generalizations in general) are not so useful.

I've seen many cases (and been frustrated) working in a corporation where high levels of caution end up creating much more total work to make changes incremental. And when you do finish, the code is often not is nice as if it had been all done at once (at which point it probably gets left as is as more important things than refactoring, in the eye of business people, come along).

It would have been hard for us to get the real-world battle testing needed to get Servo's style system into shape without rolling it out in Gecko.

https://github.com/servo/servo/pull/16971 is an example of Bobby Holley, a Gecko developer (who's awesome, by the way) landing almost a full rewrite of the multicore scheduler, making tons of sites faster in Servo.

Welcome back! There are a lot more things in the pipeline. V57 in November will be crazy fast, even better than the current one.

It's an odd situation. Firefox has many fast parts but the whole is cranky. Chrome is sleek in almost every department.

As much as I don't "like" it.. I use Chrome 99% of the time

Try Vivaldi - it's basically Chrome without the Google-everything, supporting almost all extensions too, and pretty fast.

Vivaldi is closed source though.

Chrome is too, they only open the chromium core.

No , Chromium is almost 100% chrome , just without Google Service specific features .

How can I run chromium as a daily driver on windows? What about automatic updates?

Choco packages chromium nightly builds for Windows.You could automate 'choco upgrade chromium -Y' (Daily snapshot builds).

Otherwise: http://chromium.woolyss.com/#windows

Which packages http://www.henrypp.org/product/chrlauncher in the portable edition

Features like the updater / appmanager, which is borderline malware and one of the reasons I uninstalled Chrome (and boy did it take some effort). Regardless, Chrome is not fully opensource, so there is no way to "calculate" the % you mention. Google might be running a bitcoin miner in Chrome and you would have no way of knowing.

The degree of trust required to run Chrome is equivalent to the one you need to run Vivaldi; the only difference is who you trust (Google vs a bunch of ex-Opera developers). If you don't trust anything that is not opensource, you shouldn't run Google Chrome - you should run Chromium or Firefox builds published by trusted parties.

Try Brave then? FOSS, based on Chromium, created by Brendan Eich.

Wow, that is pretty gross, makes me wish chromium was GPL.

But that would prevent Google releasing the closed source Chrome.. And if khtml had been GPL apple couldn't have released the closed source Safari browser based on it - which would have meant that webkit might never have become mainstream - and chromium probably would have never even existed.

https://vivaldi.com/source/

This isn't the entire source code. It's just the changes that they've made directly to Chromium. They can still bundle all kinds of closed-source stuff alongside that, just like Google does with Chrome.

Reply on news.ycombinator.com