Firefox 8 20% Faster Than Firefox 5 441
An anonymous reader writes "Thanks to continued improvements to start-up and first paint performance, tweaks to memory footprint and garbage collection, and the addition of a new 2D graphics backend called Azure, Firefox 8 is some 20% faster than Firefox 5 across all major metrics — and actually about equal with Chrome 14 on JavaScript and 2D rendering performance. Azure (which is new with Firefox 7) replaces Cairo, and instead of dealing with Direct2D and Quartz, it allows Firefox to deal directly with the Direct3D and OpenGL subsystems — resulting in a 20% speed boost under Windows, and probably even more under OS X."
First post (Score:5, Funny)
Your post could've been here if you had a faster web browser.
What about Firefox 6? (Score:3, Insightful)
I know it's been said before, but this new release system is fucking retarded.
I'm this close to dumping Firefox on every machine I touch.
Re:What about Firefox 6? (Score:4, Interesting)
I agree that the race-to-the-highest-number game is silly; but it is the silly, albeit visible, symptom of the FF team having a fire lit under its collective ass by Chrome. Arguably, while a lot of what gets full numbers really should just be point releases, FF's quality today is relatively better than it has been in some time. Are you really going to let the version numbers get in the way?
Re:What about Firefox 6? (Score:4, Insightful)
Because "if our plugins didn't fail to run because Firefox decided to throw version number semantics overboard, they would be running 20% faster" is not a good position to be in.
Re:What about Firefox 6? (Score:5, Funny)
Re: (Score:3)
Silly you.
Re:What about Firefox 6? (Score:5, Funny)
Firefox 8: "The Ocho"
Re: (Score:3)
Firefox 8: "The Ocho"
Then I dread the loudmouthed point five release.
Re: (Score:3)
Why are we comparing version 8 to version 5? That's insane. The releases are so fast now they can't even measure adjacent releases to each other!
This appears to be a damage control sound bite; a way to reassure customers that their project mismanagement has an upside.
Dumping Chrome and Firefox because of numbers? (Score:3)
I know it's been said before, but this new release system is fucking retarded.
I'm this close to dumping Firefox on every machine I touch.
I think it's silly to dump both Chrome and Firefox because of their release systems (which are identical - both release a new major version number every 6 weeks).
I guess you can use other browsers than Chrome and Firefox. But those other browsers release new versions with new features very rarely. Is the *version number* enough of a reason to not use Chrome and Firefox? I don't think so - even though I thought it was silly when Chrome started with it, and when Firefox decided to do it as well.
Re: (Score:3)
Yep, it's just a number.
eg. I'm sure it went from 4.05 to 4.08 you'd be all "Yippee! A new version!!". The only difference is in your head.
Re:What about Firefox 6? (Score:5, Funny)
Firefox 8 has a version number that is 100% greater than Firefox 4. Now there's a big improvement for you.
Re:What about Firefox 6? (Score:5, Insightful)
Tell that to the extentions that constantly break on new major version.
Re:What about Firefox 6? (Score:4, Informative)
Addons hosted on addons.mozilla.org are now automatically checked for compatibility with new versions (by checking API calls used by the addon) and are bumped to show as compatible. If not, an email gets sent to the addon developer alerting them to the fact their addon is broken, and what exactly is broken about it.
This will start happening for the release of 6.
Re:What about Firefox 6? (Score:4, Interesting)
Tell that to the extentions that constantly break on new major version.
Also, tell that to Mozilla's extension approvers, who won't approve any extension marked as having a maxVersion greater than 8, and that was 7 up until a few days ago.
Granted, you could distribute your Extension through your own site instead of the main Firefox Add-ons site...
Re:What about Firefox 6? (Score:5, Insightful)
Is it possible to check compatibility of installed extensions before your upgrade? Other than going to the homepage of each extension, that is.
If not, that would be a nice option:
"A new version of firefox is available. Checking list of extensions for compatibilty issues ... Done! From 10 installed extensions, the following are incompatible with the new version. Would you still like to upgrade?"
Re: (Score:3)
Re:What about Firefox 6? (Score:5, Insightful)
It is now. It wasn't once before. At one point, version numbers had meaning, that is until marketing and people such as yourself decided it really didn't matter.
I've worked doing both support and development and for me, a version number is not 'just a number'. It tells me something about what I can expect.
The number should be as follows: major.minor.patch.build
A major version bump means there WILL be compatibility issues. API changes require a major bump. Massive UI/Feature changes warrent a major bump a lot of the time.
A minor version number means it probably won't break anything unless you have some dependency on a specific bug that has been fixed, which should be unlikely if you followed documented APIs and such. May introduce minor new features, should mostly be features that you missed on the release or where incomplete due to deadlines. Occasionally the 'heh, I forgot that obvious small feature that makes a world of difference' goes here, but for the most part this should be reserved for fixes that may accidently break compatibility.
A patch version number change means you shouldn't notice any difference other than a security/bug fix.
The build number is a unique identifier for the developer to reference internally as an exact point in time snapshot of the package. Thats the number that allows the dev to reproduce the exact same build.
With the exception of a major number change, all other changes shouldn't result in massive breakages, for instance a major version change is the only thing that should break Firefox plugins en masse, and it should be avoided if there is any possible way to do so.
I want to know what to expect from an upgrade, traditional version numbers help me at a glance determine my risk.
Firefox (and Chrome) is taking an approach that I could only describe as amateur and completely lacking of any experience of having to support someone elses software product. It indicates you (Mozilla) don't give a flying fuck about what happens to the people who use the browser or what their experience is like, just that they get their way.
And for the record, going from 4.05 to 4.08 generally would result in my response being something like:
Great, another Firefox release, wonder how much this one breaks?
The only people who randomly see a new version of software and get excited for no reason aren't relevant to this conversation. Normal people (the majority of us) have better things to do than worry about the latest release of Firefox. Anyone who does get excited needs to seek counseling.
Re: (Score:3)
If you think the versioning system is stupid then why are you putting Firefox on every machine you touch? You want everyone to see how bad it is out of vengeful spite or something?
It seem to me that by putting it on every machine you touch that would get more people using it. I don't get it.
By 'dumping Firefox on every machine I touch' he means dumping it off, i.e. removing it from every machine he touches.
And he is right about the pain of this release schedule. I want a stable web browser to view webpages. I don't want to update it more than about once a year.
Re: (Score:2)
"Options->Advanced->Update->Ask me what to do..."
History has shown that for most people, manual update doesn't work.
Plus... if you really are putting Firefox on every machine you touch then you should be grateful to them for saving you a lot of work every time an update appears.
Re: (Score:2)
...yet those same "enterprises" happily support Flash (which does constant update demands), no?
Re: (Score:2)
Firefox isn't for enterprise.
No?
So what is?
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
Meh, nonsense anyway. Everyone knows that real open source projects never have version numbers greater than 0.999
Comment removed (Score:4, Insightful)
Re: (Score:3)
Dumping Firefox because their project management is clearly lost and out of focus makes sense. 20% faster is no reason to ignore the dysfunction.
Firefox 6 is already old-and-busted (Score:5, Funny)
Firefox 6 is so out of date, my parents will probably use it when it comes out.
For those confused (Score:5, Informative)
Re: (Score:2)
I was worried. I'm just downloading FF6 Beta and suddenly I'm 2 version behind. This isn't rapid release, this is TARDIS programming.
Re: (Score:3)
So wait. Does that mean I get a sonic screwdriver with a new download of Firefox?
Re:For those confused (Score:4, Funny)
So wait. Does that mean I get a sonic screwdriver with a new download of Firefox?
No, it means you get a shower of sparks and your ass dumped in an awkward situation because you hit the wrong button. ;)
Re: (Score:3)
I don't care what the "reasoning" is - this is just ridiculous versioning. At this rate in 2020, we'll be using Firefox 153. It will be confusing for the users.
The whole purpose of the point system is to separate major changes from trivial changes. 4.01 is a bug fix. 4.1 is an upgrade. 5.0 means you gave your program a whole new look (or possibly a complete rewrite).
I'm still using Firefox 3.x and hearing they are now on version 8 makes me think the MARKETERS have taken over Mozilla, instead of the per
Re: (Score:3)
You'll be using simply "Firefox". Version numbers are for internal project tracking only.
Then why display them at all? Why do we know there is a Firefox 8?
If it was done for the MARKETERS, wouldn't you expect Firefox versions to advertise their version number?
So you're saying that it isn't used for marketing purposes ... like the model year on a car isn't either I presume. If it wasn't used for marketing purposes why change it? Why not stick with what everyone else is used to and what ever other commercial software package uses already?
If it was JUST an internal reference point we wouldn't be having this discussion, but what happened is marketing learned of version numbers a few years back and
Re: (Score:3)
Re:For those confused (Score:4, Funny)
So, does that mean it has a 5 minute summoning animation??
Re: (Score:2)
Mozilla hasn't suddenly started to number their versions geometrically, although that would be hilarious.
Try it and chrome will one up them by releasing Chrome .... google...
Should I quit this IT stuff and do stand-up comedy?
Re: (Score:3)
have they fixed (Score:2)
the random mysterious lockups? there are times about 4 a day that firefox locks up to any responses for a few seconds. Most of the time on Slashdot, but I have seen it on MSDN and other sites.
Some days it's so pervasive I switch back to Chrome.
Re: (Score:2)
Re: (Score:2, Troll)
Half the Internet has reported that bug.
It's often caused by having a page in one tab doing something slow (loading from a slow server, lots of JS, something like that) and locking up all the other tabs because they aren't properly independent in Firefox.
Someone will be along shortly to bleat about how this will be fixed in Firefox 15 or something, because apparently it's now OK for anything that is mentioned in a footnote on a roadmap to be announced as if it's an existing feature in production builds, eve
Re: (Score:2)
By the way, is this true for Chrome, too? It's said to be parallel but by my experience (on a dual-core system) some other busy tab can quite efficiently jam the current one I'm browsing.
Re: (Score:2)
As far as I can tell, neither Chrome nor IE has suffered from this to anything like the same extent for literally years.
Re: (Score:3)
By the way, is this true for Chrome, too? It's said to be parallel but by my experience (on a dual-core system) some other busy tab can quite efficiently jam the current one I'm browsing.
I tend to find it's DNS that causes the jam-everything problem. If your DNS is slow to respond, everything grinds to a halt.
Re: (Score:2)
Re: (Score:3)
Feel free to explain to me how that's possible on a system with 12 GB of memory, while Firefox is using 700MB, and I have no other applications running.
And before I forget, I have an SSD, so I often don't even notice it when applications really do start swapping. Nonetheless, on one of the faster desktop PCs money can buy, Firefox freezes regularly.
It's doing it right now, pausing for a about 500ms before responding to keystrokes every 10 seconds or so.
It also does it on my work laptop, which has 8GB of mem
Re: (Score:2)
Wow, talk about version inflation (Score:5, Insightful)
You fools are only benchmarking Firefox 8!! Well I benchmarked Firefox 14 and it's plus 10 faster than Firefox 4.
I appreciate the benefits of rapid versioning and release cycles, but really, this is ridiculous.
Re: (Score:2)
You fools are only benchmarking Firefox 8!! Well I benchmarked Firefox 14
Boy, are you behind. I just downloaded Firefox 17 and it blows Firefox 15 out of the water!
Re: (Score:2)
Forget that, I am using Firagafox 51 for a more-potent browsing experience.
Re: (Score:2)
I got it ... but, I wasted many hours of my life to be able to instantly recognize it.
Re: (Score:2)
Re: (Score:2)
I'm giving up my mods on this thread to post this, but I simply couldn't resist jumping in with Buzz Lightyear's
Personally, I think we should just start going to build numbers for every project. Having Firefox 76,326,358 would certainly take care of that annoying Chrome rivalry. ;-) Then perhaps geometric sequences, Taylor series, quantum dynamics... software versioning would never be the same again!
Re: (Score:2)
Nah, Google will just start using Googlebids as their versioning scheme. (With one Googlebid being equal to version pi-billion for any other browser). That'll make short work of the browser VERSION wars (to heck with the browser wars).
Firefox 61 (Score:5, Interesting)
By my calculations, if Firefox had started this version numbering scheme with its start in 2004, we would now be running Firefox 61.
If they Mozilla had adopted it in 1998, this would be Firefox 113.
Bonkazoids.
Re: (Score:2)
I think when they started out they just used a different browser name every version?
Re: (Score:2)
No, it was Firebird for quite some time, untill they got forced to change it in 2004.
Re: (Score:2)
It needs to go to plus 11
At least it's not roman numerals... (Score:2)
FFIX would be really confusing.
random bugs in rendering abound... (Score:2)
simply because they split the rendering... surly it would have been better to test it and leave a "compatibility " view there ?
and what about support for PowerPC ?
regards
John Jones
iOS browser does this from day 1 (Score:2, Informative)
Re: (Score:2)
Firefox 6 & 7? (Score:2)
What happened to them, I'm still on 5 :)
Re:Firefox 6 & 7? (Score:5, Informative)
From what I can tell, Mozilla seems to have four versions of Firefox being developed and/or maintained at any given time:
Current - Whatever is currently released. Only bugfixes usually get ported to this release. Currently FF5.
Beta - Feature-frozen and reasonably stable, but not quite ready for prime time. Will be the next release. Currently FF6.
Aurora - Feature-frozen, but not stable. Early QA happens here, though it gets more fleshed out in Beta. Currently FF7.
Nightly - This is where the new feature development happens. Currently FF8.
When it's time for release, everything gets promoted: when FF6 is released, FF7 will become Beta, FF8 will become Aurora, and new development will start on FF9.
I kind of like the idea of putting new code through two entire cycles of public testing. All the same, I do wish that Mozilla would add a Long-Term Support cycle every few versions, akin to Ubuntu's LTS cycle, that people could count on to be supported for more than just a couple of months.
It is true that sane IT departments upgrade their browsers regularly, but not all IT departments are driven by sanity. This is a sad fact that Mozilla needs to account for, and there's a tested model out there that isn't too dissimilar to Mozilla's own. They should seriously look into adapting the differences.
Re: (Score:2)
Essentially a copy of Debian development:
Stable
Testing
Unstable
Experimental
Someone should make an ISO standard for naming, or a design pattern for standardized naming, or similar.
Re: (Score:2)
Re: (Score:3)
FF just pushed me ("Strongly Recommended") to upgrade to Firefox 6. My favorite part is after the upgrade installs, then it runs a check to see which of my plugins are compatible. Hmm, you think maybe it would be a useful feature to run the check BEFORE doing the upgrade so then I can make an informed decision about whether to upgrade?
Re: (Score:2)
You can keep your blinking text and ad-serving plugins. I'm sticking with NCSA Mosaic.
And so this means that.... (Score:2)
Ridiculous (Score:3, Interesting)
Do you have any idea how it complicates Web contracts? We used to be able to say "your website will be compatible with current version-2 of the browsers" but now that would be ridiculous. We'd never be able to deliver since we would be stuck in a infinite testing loop.
We'll have to start writing "your website will be compatible with Firefox 5" and by the time we deliver Firefox 12 will be available. I guess we'll have to add a clause about how Microsoft, Google and Firefox are all teenagers who compare their peni- I mean version numbers to feel good about themselves.
Apple aren't being childish with the whole issue and using sane version numbers. And Opera has been out for quite a long time, though they do seem to be jumping into the version bandwagon as of late.
Re: (Score:2)
Could you state a date before which the browsers are compatible? Something like this:
-"Ahoy, sweet customer! We will make sure the product is compatible with $LIST_OF_BROWSERS released before $MONTH $YEAR
Great... yet another version of Firefox to support (Score:5, Informative)
We just got our web site rendering correctly under Firefox 5, and now there not one but THREE new versions in beta that we also need to test with.
Just a quick note from the web developers and web site QA testers around the world to the Firefox development team... you're really starting to piss us off.
Re: (Score:2)
Mozilla gave you the finger (Score:3)
Not all addons are on AMO nor in pure JS (Score:5, Insightful)
If your addon is on addons.mozilla.org
Does addons.mozilla.org offer private hosting of bespoke addons used by a single company? Does addons.mozilla.org offer hosting of addons whose use requires payment? Or is addons.mozilla.org intended solely for addons intended for public use at no charge?
they've begun automatically testing addons for compatibility
I seem to remember reading that any add-on incorporating a native code component will automatically fail the test.
Re: (Score:2)
Must write some really convoluted CSS/HTML. I've never really had something render differently in Firefox 3.6 vs. 4 or 5. If it renders in both IE7 and IE8 AND any version of Firefox, it will render in just about anything modern.
Do not make much sense right now (Score:2)
Re: (Score:3, Funny)
It'll be alpha on Friday, beta on Sunday, released next wednesday and discontinued next friday.
doesn't matter; I'm using chrome (Score:2)
google owns me
When will they add per-tab processes? (Score:2)
Chrome and Internet Explorer (as of version 8) support per-tab processes. That is the one major feature I would like Firefox to implement so that if a page's JavaScript or Flash goes bonkers it doesn't take down all my tabs.
Also, Google is changing the way their accounts work in a few weeks that will prevent users from being able to access multiple Google accounts in the same browser. I am not entirely sure on the particulars, but wouldn't per-process tabs work around that to allow us to have multiple Gmail
pfft FireFox 8 (Score:4, Funny)
Rendering (Score:3)
Most of the comments have been about the version numbering...
I'm curious about the change to rendering. It seems to me they're saying, "these OS layout engines (Quartz et al) are too slow - we'll just route around them". Understandable, but it's kind of a shame that they'll presumably be re-solving a lot of the problems that Quartz et al deal with (e.g. are they going to do their own font rendering?), and I wonder why their concerns can't be addressed by altering Quartz.
I'm not criticising the decision, I'm just curious as to the reasoning that goes on when such decisions are made. (I'm always interested in the practical examples of why those lessons they drum into you at university about the myriad benefits of code reuse, standing on the shoulders of giants etc don't really pan out in the real world.)
Is the job they're doing fundamentally different? (such that rendering via Quartz was the wrong idea in the first place)
Is there some key component that fundamentally could not be in Quartz? (maybe, embedding videos or somesuch)
Is it that Quartz isn't open source (or Apple cooperative enough) and so Mozilla can't realistically get them to fix it in a sensible timescale?
Is it that they'd have to do this with all the vendors, which isn't feasible?
Is it that abstracting on top of different vendors' APIs turns out to be too much of a headache? (maybe a pure-Quartz implementation would be as fast as the OpenGL version but it's all the Mozilla layers above Quartz that are sub-optimal?)
I wonder if their rendering engine will be released as an independent library that Gnome/KDE etc could incorporate if they wanted to.
Re:Rendering (Score:4, Informative)
Precisely the opposite. It's our previous abstraction layer that's too slow, and we're replacing it with a thinner one, starting with the easier things like Canvas. See Introducing the Azure project [mozilla.com] and Azure vs Cairo [basschouten.com].
Re: (Score:3)
Actually, Quartz might end up getting used as a backend for the Gecko-internal Azure API on Mac at least for a bit (just like Direct2D will be on Windows). See http://blog.mozilla.com/joe/2011/04/26/introducing-the-azure-project/ [mozilla.com] for some discussion of the tradeoffs here. There's some discussion about why it might be preferable to use a GL backend instead of Quartz on Mac (long story short, a GL backend is needed anyway for Linux and Android, and at that point it might make more sense to use it on Mac to
64-bit flash on Windows (Score:3)
Firefox a cause of Technological Moral Panic (Score:5, Funny)
When I saw "Firefox 8" in the title, I fell into a panic. What happened to 6 and 7? People weren't meant to upgrade their browsers to new major version numbers weekly! No one could possibly survive that pace, their mouse buttons will burn out at the furious pace necessary to install that often! Think of the effect that has on the women and children!
Time for an ask Slashdot (Score:3)
I used to work for a web development company and it was always a pain to keep or get web-sites working with various versions of browsers.
With Chrome I would have told customers , "Hey, if it happens to works with Chrome that's just great, but we can't continually test against new versions of Chrome".
Now I work for a medium size company and we have limited the number of browsers our internal web interface will work with. Currently it is with Firefox. But now it appears that we are going to have to move away from Firefox. I hate to go back to IE but it appears that is where we are heading.
Sorry Firefox, but we can't just keep regression testing at your whim.
So maybe it is time for someone to ask, what is the recommended browser for corporate use?
Re: (Score:2)
Re: (Score:3)
no. Why? because it's called seamonkey.
Re: (Score:3)
Seriously. I wish they would change the name. I'm embarrassed to tell people what browser I use because the name is so stupid.
Re: (Score:3)
My poor little SeaMonkey is only up to 2.1. Somebody obviously needs to get their sh*t together!
Well, SeaMonkey 2.2 has been out for three days now...
Re: (Score:2)
Their competitors like... Chrome? Who introduced this version numbering system in the first place?
Re: (Score:2)
The difference is that Chrome was doing this from day one so people could just accept it as a quirk; FF had an established system and changed it for no great reason, imho.
Re: (Score:2)
I tried switching to Chrome again but, like the last time, the extensions weren't half as good. Even the ones developed by the same teams as FireFox! From what I understand it is the difference between the two browser extension APIs that is causing the problem. So, I'll stick with FireFox for now.
Re: (Score:2)
I tried switching to Chrome again but, like the last time, the extensions weren't half as good. Even the ones developed by the same teams as FireFox! From what I understand it is the difference between the two browser extension APIs that is causing the problem. So, I'll stick with FireFox for now.
My web browser is adblock plus, flashblock, noscript, xmarks. Also firebug. They define my "internet experience". I currently use firefox as the backend, as long as the backend stays out of the way of adblock and noscript, and cooperates with xmarks, I'm pretty happy. If chrome makes a better backend for my addons, then I'll switch to chrome. Not till then. All I need is a working, configurable ad blocker, flash blocker, JS blocker, and bookmark syncer.
Re: (Score:2)
Re: (Score:2)
"animations for the websites I design"
Spotted your problem right there.
Re: (Score:3)
The 32/64-bit situation for Mozilla right now is the following:
Linux: both 32-bit and 64-bit builds are available and supported starting with Firefox 4. Default download is probably 32-bit.
Mac: On 10.6+, the browser runs 64-bit by default and the plug-in process runs 32-bit. On 10.5, both run 32-bit due to OS-level bugs in 64-bit support.
Windows: Only 32-bit builds are supported. 64-bit builds exist, but are somewhat buggy. Shipping them involves fixing those bugs and addressing the issue about not bei
Re: (Score:3)
You can download it and see. In my testing, Firefox today kills Firefox from back then in performance. JS is about 30 times faster. Start-up is about 5 times faster. Rendering is much faster. UI responsiveness is way ahead. It's a slam dunk. Go get Firefox 1 for yourself and give it a try.
Re: (Score:3)
The author of this article only reports performance numbers for Microsoft's Windows OS and Mac OS X but fails to report the actual performance under Linux. Pretty pointless article with such limited numbers.
Pointless because it's only relevant to 98% of Firefox users? It would it be nice if every website reviewing Firefox has 6 machines (or VMs) so they could report on win32, win64, Mac32, Mac64, Lin32, and Lin64? Actually, make that about a dozen different OS versions. Win XP, Win Vista, Win 7, Mac OS X 10