I've been using $RELATIVELY_FRESH_BROWSER_INSTALL again recently. I stopped when it was slower and less stable than $BROWSER_I_HAVE_NOW_BEEN_USING_A_LONG_TIME.
These days it often feels faster than $BROWSER_I_HAVE_NOW_BEEN_USING_A_LONG_TIME.
Yes, it would. It's not based completely on performance gains you may see in new updates. Profiles bloat up over time with data (history, cookies, cache, etc) and the bigger it is, the more the browser has to process. So, over time, your browser will slow down. A fresh install of a new browser would result in all this data not existing. If it doesn't have to parse hundreds of megs of files, it'll generally be faster.
All of those things are easily cleared, however. If I do a clear history/cache/cookies in Firefox, there's still a lot of miscellaneous data floating around in my profile, to the point where, last I recall, the Firefox team still recommends you outright dump your profile in case of difficulties.
There is nothing inherent in the design of A hypothetical browser that would make it slower over time, its just a matter of people make assumptions and those assumptions are often wrong, and then performance suffers.
Some things do accumulate. That is in fact the point of Firefox Reset[1].
The support page lists "Extensions and themes, website-specific preferences, search engines, download history, DOM storage, security settings, download actions, plugin settings, toolbar customizations, user styles and social features will be removed."
Installing extensions can dramatically slow down browsers. I wonder if the lack of extensions is responsible for the quickness users experience with a fresh install.
Absolutely – there's really common cycle I see repeated all over:
1. “(Firefox|Chrome) is slow”
2. “Did you try removing Ad-Block Plus, etc.?”
3. “Now it's fast again!”
I think we're well past the point where the browsers are going to need to start having some UI around monitoring and exposing slow extensions to make it easier for users to learn this.
Absolutely. When I first starting using FF and all the developer tools, it would grind to a halt. Now that I'm a lot more comfortable with the FF toolset, I have a minimal set of extensions and it feels much snappier.
I also use Aurora which is still super fast to me. I also develop with Chrome Canary and the stable Chrome version. Canary has times when its totally useless or when someone breaks the build and I can't use it for a week so until the next update comes out.
All in all, yes, less overhead means a snappier browser to me.
That's my biggest reason for switching from AdBlock+ to Privoxy + https://github.com/skroll/privoxy-adblock . Same blocking rules but uses a ton less memory, since each tab isn't running all of AB+ .
That's fine for image and flash ads, but provides no support for CSS blocking of elements. My understanding is that the CSS blocking is the slowest part, too.
I've been using Firefox for several months and have no regrets. On Ubuntu Chrome was doing all sorts of weird things, plus being slow. Firefox's tab groups is also a great feature.
On my Linux MINT machine, chrome keeps doing something[1] that ends up locking up my whole machine and I have to hard-reboot. It got to a point where I actually had to set up Ctrl+Alt+K to issue "pkill -9 chrome" so as soon as I see the mouse-pointer movement become non-smooth or music start to skip, I slam on those keys to kill chrome before I have to restart my whole machine. Then I just went to Firefox developer edition[2]. Is it better? I dunno, but I like its cool dark theme, my system hasn't locked up since and now I got all those cool extensions back again.
I had that problem with Linux Mint on some configurations too. I found that by going into chrome://flags/ and disabling everything related to GPU acceleration the crashes went away. YMMV, of course.
The discussion is about Google Chrome not working on Linux Mint. I said that Google Chrome works for me on Linux Mint. How is that not relevant?
I'm not saying there's no problem, I'm just offering my experience to demonstrate that perhaps the problem isn't just "Chrome runs poorly on Linux Mint", and to offer a counterexample to "Linux Mint is a buggy mess".
I can second the Ubuntu motivation. Chrome is buggy and unreliable on Ubuntu, and I was tired of the Aw Snap several times a week. Firefox appears to be much more stable, and Developer Edition looks better too.
I switched back to Firefox recently as well after using Chrome for years. Mostly it was due to problems on Linux, like Flash not working. I also love the work Mozilla has been doing -- Rust, Servo, and asm.js in particular -- which seem to be very promising vis-à-vis future versions of their browser.
Firefox won't have a process per tab. Even in the multiprocess versions, it has one process for the UI and one for all the tabs' content. That way it doesn't take tons of RAM for overhead if you have lots of tabs, but heavy tab content can't freeze the UI. You can change the number of tab-processes in about:config under dom.ipc.processCount.
Another advantage of using one process for all tabs is that Firefox can take advantage of security sandboxing of the content process without the overhead of tons of per-tab processes. (Quite a few Firefox users like using hundreds of tabs. :) Eventually Firefox will tweak the tabs/process balance to find a good compromise.
Firefox uses compartments per domain instead, which is a great alternative to process per tab:
"Some readers might wonder how compartments compare to per-tab processes as they are used by Google Chrome and Internet Explorer.
Compartments are similar in many ways, but also very different. Both processes and compartments shield JavaScript objects against each other.
The most important distinction is that processes offer a stronger separation enforced by the processor hardware, while compartments offer a pure software guarantee. However, on the upside compartments allow much more efficient cross compartment communication that processes code.
With compartments cross origin websites can still communicate with each other with a small overhead (governed by certain cross origin access policy), while with processes cross-process JavaScript object access is either impossible or extremely expensive.
In a modern browser you will likely see both forms of separation being applied. Two web sites that never have to talk to each other can live in separate processes, while cross origin websites that do want to communicate can use compartments to enhance security and performance."
Yeah, they have a MemShrink team now. https://wiki.mozilla.org/Performance/MemShrink They got serious about memory leaks in their own code, and even took steps to limit hogging by 3rd-party plugins. Now they host https://AreWeSlimYet.com/ to track regressions. Edit: and Chrome really does take a ton of RAM if you have lots of tabs open.
"and Chrome really does take a ton of RAM if you have lots of tabs open."
Yup, it does. My problem with FF years ago was with leaks. It would eventually bring my entire system to a state of paging hell until I killed it. That was years ago though, I've since become comfortable with chrome and haven't found aneed to go back.
Nope. Pinch to zoom "works", but it's the same zoom increments as Ctrl - and Ctrl =.
I've been using Firefox for a few years (switched back from Chrome), but it's IE for me on the Surface until someone else does proper touchscreen support. The only time I open Firefox on here is when I need a bookmark that's synced from my desktop or phone.
I was using Firefox in the past few months because Chrome felt slower after the new update.
I am back to using Chrome however. It's really hard to break the habit even though I made it a conscious effort to use Firefox and have enjoyed it thoroughly.
My only gripe with Firefox would be watching youtube I see a Adobe process that takes up a lot of memory even though I force HTML5, it seems to revert to Adobe.
All in all Firefox is the Chrome browser I once loved, but now that Chrome is working back to what it was doing, I forgot about Firefox.
I've been using $RELATIVELY_FRESH_BROWSER_INSTALL again recently. I stopped when it was slower and less stable than $BROWSER_I_HAVE_NOW_BEEN_USING_A_LONG_TIME.
These days it often feels faster than $BROWSER_I_HAVE_NOW_BEEN_USING_A_LONG_TIME.