Can Mozilla's Quantum project make Firefox king again?
The browser world has changed considerably since the release of the first version of the Google browser Google Chrome.
Google Chrome managed to snag a sizeable share of the browser market not only on the desktop but also on mobile.
Google pushed Chrome hard on its properties and via third-party agreements (Flash pushes Chrome up to this day), but that was just one part of why Chrome is used by the majority of Internet users on today's Internet -- at least on the desktop.
This rise had an impact on the then reigning browsers Firefox and Internet Explorer which both lost market share to Google's browser.
Mozilla found itself in a situation where it was clearly behind in many areas: in performance, responsiveness, and security for instance. Firefox still reigned in other areas, customization options, a superior add-on and personalization system, and general user control of the browser.
Mozilla Quantum
Quantum is the next step in Mozilla's plan to reconquer the browser market and provide its users with an improved way to browse the web.
Quantum, as David Bryant, head of platform engineering at Mozilla, puts it, is "Mozilla's next-generation web engine". Mozilla plans to land parts of Quantum in Firefox in 2017, and Bryant suggests that Firefox users will see "major improvements" by the end of the year.
So what is so special about Quantum?
Quantum is all about making extensive use of parallelism and fully exploiting modern hardware. Quantum has a number of components, including several adopted from the Servo project.
Implementation-wise, Quantum parts will replace their equivalent in Firefox's Gecko engine when they are ready. This allows Mozilla to ship those parts when they are ready which in turn means that Firefox users will benefit from the improvements they bring along with them immediately.
Quantum will replace parts that benefit from parallelization and offloading to the GPU.
The components that will likely make it into Firefox in 2017 are parallel layout, parallel styling, WebRender, and Constellation.
You are probably wondering how big of an impact those components make when they are introduced. The answer is it depends. First, on the machine that Firefox is running on. To use parallelization, there need to be multiple cores available. Second, on the website that the user accesses. On Wikipedia for instance, a site that is not optimal for Quantum optimizations, style and layout compute times drop by about 50% with four cores.
On Reddit however, a site that Quantum benefits from more due to the way the site is designed, compute time drops to a quarter or less of the Gecko compute time.
The worst case scenario -- nothing can be handled using parallelization -- still offers a 10% advantage over Firefox's current engine Gecko.
The following video is a presentation by Jack Moffitt who talks about Servo in particular and how it will benefit Firefox in the near future.
Quantum will make Firefox a lot faster and more responsive than it is today. The difference is significant, and will be noticeable when you compare Firefox's performance to that of Google Chrome, Microsoft Edge, or other browsers according to Mozilla.
You can find out more about Quantum on Mozilla Wiki.
Closing Words
Can the Quantum Project make Firefox king again in the browser world? I think it will be beneficial to the browser's market share, provided that Mozilla gets it done right and that Chrome or Edge won't introduce similar functionality at around the same time.
I doubt that it is enough to kick Chrome from the throne, but it is likely that Firefox will see an uptick caused by it.
Now You: What's your take on Quantum?
It blows my mind how pathetic people are. Firefox is the best browser, hands down. It still has more and better addons than any other browser. Firefox is better than it’s ever been. Australis is fine, it looks nice, I got used to it. WHO CARES. Live with it. So many whining children. These people are the reason Firefox is losing market share. You want to use spyware? Good, use Chrome.
years ahead?… lot more stable?…. c’mon.
https://www.reddit.com/r/waterfox/comments/4aqjy0/main_differences_between_waterfox_firefox/
read MrAlex94 posting. that’s it.
From the results, I’ve had from Firefox, over 25+ years. I can’t believe they will ever change!!!!!!!!!Why hasn’t someone from Mozilla, mentioned “Waterfox? . Been around for years as a 64 bit o/s, yet very few have heard or it!!!!!! All Firefox extensions and most “pluggins” work, It’s as solid as a rock , no crashes yet(been using it for years) It and Chrome, are my browsers of choice( we won’t mention “TOR”) That’s for serious shit.
After 25 years online. I’ll believe it when I see it. :)
No-one has mentioned “Waterfox”. Made by the same people as Firefox (which I have removed) due to constant crashes , etc. No problems with “Waterfox” , 64 bit , uses the same extensions , plugins, with none of the downsides!!!!!!!!!!! Have used it, ever since I found out about it!! Years ahead of “Firefox”, 64 bit, years ago!!!!!!!!!!!! Just not advertised.It and Chrome for me!(I use Thunderbird, as my default email program . After working out a few bugs, It’s brilliant. Better than Gmail(which still loses emails), no-one can fix that , including Google!!!!!!!!!!!!!!!!! I have 3 x i7 8 core, max memory, Bluray and DVD Burners, in all. 8 Terrabytes of storage. No problems with speed, dropouts, anything! I guess It just depends what you run.(Oh, I wouldn’t swap my Windows 7 64 bit , for Quids. Can make it do anything, not like “8”.”8.1″ which even Microsoft has given up on). Peter
Am I the only one realizing that the numbers quoted actually mean that Firefox will become even more CPU heavy than it already is right now?
All measurements made with 4 threads. Only reddit is around the same consumption (4x improvement in rendering time with 4x parallel, assuming full load — which is fair to assume here), Wikipedia uses double the CPU power, and a particularly unfortunate layout in Servo terms will consume almost 4x the CPU power to render the exact same page… >.< Those aren't figures to just ignore.
How is that going to scale on battery-operated devices like laptops, tablets, etc.?…
The video gives a reply for mobile. Because the new architecture makes full use of multiple cores, cores can be clocked down through power management where they previously couldn’t be. When a CPU is clocked down it runs slower, but the speed boost from the new architecture allows to reduce battery consumption by 40% while keeping the same speed that we currently have.
I don’t remember what exactly this was applying to, styling and layout I think ? i.e. what you seem to be talking about.
Experience also shows that Mozilla keeps a close eye on performance regressions of all sorts, as well as potential trade-offs. You can look up very detailed goals and observations for e10s in particular, where it’s neatly listed in one web page that I’m too lazy to find again. So if you see regressions, it’s highly likely that they will be fixed by the time this project hits Firefox Release.
maybe they should wait for quantum computers hehe.
to late. almost all my users have a habbit of having chrome in computer.
the only way is like chrome way having big brother. maybe samsung with mobile and tv solutions as default browser?
chrome was pushed by google search page, many softwares (lika avast).
yesterday i found forged chrome browsers. even virus creators use chrome :)
i like opera way of using blink engine. internal adblock, vpn are great out of box features.
still firefox as gecko engine is great browser for dual/single cpu with 1-2gb memory.
with firefox 53 killing vista and xp is not good idea. there is still plenty of old computers and refurbished pc.
it is question will firefox 52 will be enough for web mail, simple browsing in next 4 years which i expect that people will run this oldish os.
This is great news! I’m looking forward to the better performance.
“Can the Quantum Project make Firefox king again in the browser world?”
Certainly Firefox will be the undisputed superior product that other browsers will have to play catch up with. But product quality does not equate market share :)
honestly, do you think there’s no further development for chrome & co.? right now, firefox seems to be behind chrome and even edge on tests like jetstream and kraken.
personally, i don’t care about such tests.
Wait, he didn’t say that :O
Well either way competitors compete, so this is a big milestone in the history of web browsers brought to you by Mozilla. (Since nowadays people bash them senseless, I make a point to mention it :) )
Architectural development is a whole different beast. It’s like when Chrome first arrived: It simply was technically superior. (Despite being incomplete and lacking things that only a mature browser could have)
Last time I checked Firefox 32-bit on Windows was mostly above its competitors on benchmarks, but as you say it doesn’t matter that much when they’re so close.
On the other hand when you have a Firefox with WebRender that can render complex scenes at 200-300 FPS when current Firefox and Chrome are stuck at 10-15 FPS, it does matter quite a bit. (That also means more battery saved for scenes that all browsers render at 60 FPS.) Servo components are simply superior architecturally, and Google can only hope to reach that level by taking inspiration from the concept, just like e10s takes inspiration from Chrome.
Servo and Rust represent years of work. To evaluate the Edge that Firefox will have we’d need to have insider information on Chrome “on the side” development and their future plans. If it appears that the Chrome team has been working on an equivalent project for years already, then yes, Firefox and Chrome could still be roughly equal.
Note: On a smaller scale, Chrome and all browsers are already playing catch up with Firefox on asm.js related things. Firefox has been a few years ahead here, i.e. one browser having an edge for X years has several precedents. The biggest one is Chrome’s sandboxing, multi-process architecture. I expect Mozilla’s project to be even bigger and as Martin says, it will benefit all browsers in the long run.
The elite has always concerned a minority :) Champions as well, except that you have to fight to become a champion. “Champions” is more politically correct as well. I mean, in Queens’ “We are the champions”, replace “champions” by ‘elite’ and you’ll get a high-high eyelashes rockin’ attitude. Yeah!
Google is most likely already working on something to catch up but Mozilla has an Edge of several years, it seems.
Anyone here with an insider knowledge on Chrome team’s future development plans ?
Who knows, maybe Blink will start using Rust or even Servo components. (It would be fun but I don’t think that will happen.)
MOZILLA CORPORATION cannot fail with this ‘servo’ engine.
Otherwise they are gone, into the trashy past of IT failures they go !!
Come on Mozilla! Let’s make Firefox great again!
When this project will be achieved to run Firefox I will change my multiple machines for minimum 4 cores, it’s free, just around 500€ each… King Again.
David Bryant says…
“Quantum will replace parts that benefit from parallelization and offloading to the GPU.”
What does “parallelization” mean besides “offloading to the GPU”?
Is it taking advantage of multiple CPU cores? What?
Thank you for your help. I missed that in the article.
Article explains it already “To use parallelization, there need to be multiple cores available. ”
Parallelization in this instance is utilizing the CPU’s multi-core/threading.
Mostly how browsers perform, is from a design built way back before multi-cores were a thing (think about 16 years ago) for the consumer. Parallelization is a huge step forward into the modern era.
@ jern : fa-sci-na-ting! — Thanks for sharing!
Give me a 1,000 core processor and I’ll handle *all* available add-ons + *all* available plug-ins while having 9,999 tabs open and still won’t have the time to start the beginning of a contemplative yawn :)
WOW!
Back to another reality : my duo-core not only seems but *is* very shy now that it found out that the upper average number of cores rises to 16. The processor does a good job but I have to admit I’ve never confronted it to heavy multi-tasking. Also, we don’t always realize our speed until that runner in the park passes by like a dragster…
It’ll be time to wonder about tomorrow. My machine, new, was bought in August 2013 … 3years old is old for a computer….
@Tom Hawack
I found your Maximum Core question interesting so I did some research. For general computing 16 cores CPU’s are available. AMD has designed a 32 core for release in 2017.
However, if you are just interested in the BIGGEST, there is a 1000 core CPU (as predicted by Intel in 2010) and it’s amazing.
===================
https://www.ucdavis.edu/news/worlds-first-1000-processor-chip
A microchip containing 1,000 independent programmable processors has been designed by a team at the University of California, Davis, Department of Electrical and Computer Engineering. The energy-efficient “KiloCore†chip has a maximum computation rate of 1.78 trillion instructions per second and contains 621 million transistors. The KiloCore was presented at the 2016 Symposium on VLSI Technology and Circuits in Honolulu on June 16.
“To the best of our knowledge, it is the world’s first 1,000-processor chip and it is the highest clock-rate processor ever designed in a university,†said Bevan Baas, professor of electrical and computer engineering, who led the team that designed the chip architecture. While other multiple-processor chips have been created, none exceed about 300 processors, according to an analysis by Baas’ team. Most were created for research purposes and few are sold commercially. The KiloCore chip was fabricated by IBM using their 32 nm CMOS technology.
Do you think that would run FF fast enough?
So true that an application such as ‘Process Lasso” allows the user to associate (“CPU affinity”) a core to a given application. CPUs, multi-core, more and more powerful deserve to be taken to their full potential…
OK, the formulation is naive. But I think that’s the idea.
I linger to have a CPU with more than 2 cores only. I want more!
By the way, what’s the average and the ultimate number of cores nowadays?
when did they start talking about e10s? even when they started to “ship” alpha versions of firefox it was years ago…
the same we had with firefox x64. it was available long, a looooong time (again, years) in alpha channel before it finally made it into the release.
so let’s talk about “quantum” in let’s say 2020 again. if firefox is still around by then. :(
Firefox’s architecture has had its hands tied by the necessity to be compatible with third-parties that use Gecko and with add-ons. It’s freed from third-parties now, and add-ons will be protected from architectural changes once they are WebExtensions-ready.
A number of extensions won’t survive the change, but Firefox can evolve faster and not fear to lose any more extensions after that. Then Servo components are architected in such a way that it shouldn’t be a problem if third-parties use them.
I’m Tom Hawack and I do not endorse this statement ;)
Don’t be defeatist! (as I sometimes tell myself).
Firefox has been really bad for me in recent versions that I finally uninstalled it. It’s a mess for me. I’ve decided to move on to Vivaldi with Pale Moon as a backup browser. I hope Servo fixes the problems with sites that I’ve been having. If the promise of Servo is even half true, then this may be worth getting Fx back into my system.
@Dan
Yeah, while it’s possible that something changed with Firefox that made it suddenly less compatible with your particular PC ecosystem, it’s more likely something corrupted your install. When you feel like it you can:
– Make sure your Firefox profiles have been deleted when you uninstalled Firefox
– Make sure there are no remnants of the app in your system
– Reinstall Firefox
– Check that no unknown plugin or add-on have been auto-installed by crapware on your system
– See if that fixed the problem
– If not, go to about:support and restart Firefox in safe mode and see if that fixed the problem
– If not, the next step is indeed more involved (Bugzilla), so you can go back to Vivaldi/Pale Moon
Indeed!
@Parker Lewis – or just test with a portable FF
Firefox may become a mess, it never is out of the box. More anything is tweaked more it is likely to become a mess. I know, we all know the behavior of some users (some, I’m not saying that for you implicitly, Dan) when it comes to an OS, an application, adding this, removing that, manipulating settings inconsiderately and then being surprised to meet problems. Many users are aware of that and take care to tweak only when they know what they’re doing and always backup, be it simple settings. I’m no expert but I inform myself, here and elsewhere, read/listen experts, think twice, backup and then only add/remove/modify. This has to be a credo to avoid problems and enhance whatever. And, even when things seem to be ok, too much is not the right idea, adding one enhancement to another, one protection to another when redundancy or even over-lapped functions arise leads to sluggishness. The right settings and the right settings one with each other and all within an OS and all should be ok. Crashes, crashes … not one Firefox crash, not one Windows crash … when the computer is taken care of and not filled up like a bachelor’s closet that’s how it goes.
Give it a couple years’ time. They’ll forget about, deprecate, and eventually drop it by then, just like all their other projects.
Quantum has been in the works for several years already. You can find articles of it dating back to 2014.
Just like e10, Mozilla will take it’s time with it.
Hopefully it will make Firefox faster and more responsive WITH EXTENSIONS installed.
A bare Firefox installation is actually quite fast and responsive.
But as soon as you add in extensions like uBlock Origin, it feels sluggish and less responsive.
> A bare Firefox installation is actually quite fast and responsive.
Yeah, and straight Kool-Aid has no added sugar. But it’s not drinkable.
This is just FALSE. uBlock Origin makes browsing faster, as these tests show:
https://github.com/gorhill/uBlock#performance
Don’t spread wrong information!
Rafuse,
You could also do the same ;-)
Speed and performance are relative to the machine it’s running on. A bare installation can make pages quick but ANY addon could make sluggish.
While uBlock Origin works best with more modern PC’s, older ones might not benefit from it as much (or at all).
The graphs that you linked are also not of a typical webpage (note the “high-traffic” part). In many cases, load times vary by marginal numbers.
I do use uBlock Origins and my experience with it has been nothing but great. However, wrong information is a two-way street.
@Chris
Perhaps you feel that because of e10s ? It’s currently not enabled when you have add-ons installed (with some exceptions). That means Firefox without add-ons should currently feel faster than with most add-ons, since it received a speed boost that Firefox with add-ons has to wait a few more months for.
I have 70+ add-ons and 1 plug-in (Flash) installed and, as Firepok, rendering is fast and responsive. Only impact of many add-ons is on Firefox’s start delay. Anyway it’s not uBlockO which will slow down whatever, on the contrary : it brings much more than it takes.
Quantum. Gosh, when I read/hear certain comments it’s like if Mozilla would be behind when not innovating and wouldn’t catch up when exploring because it would be too late, because the game would be over… again and again this mentality which analyzes on the short-term when, unless an unrecoverable failure, time is always on the side of innovation… if you give time to time. 2018 is tomorrow and if first impact of Quantum will appear in 2017, great. Here, though, I’ll have to upgrade from mu duo-core sweet little machine to take advantage of Quantum’s enhancements. 2017 will be a great year :)
I have almost 20 extensions installed, still fast and responsive.
I never understand why people are complaining that Firefox is slow.
Hallelujah!
Servo has been talked about for years but AFAIK there’s been no indication whether it would take years or decades to make a difference. If Mozilla can reverse it’s bizarre failure to promote 64 bit builds, get e10s bedded down ( so far, a complete disappoinment for the 40% of users running add-ons [come on add-on devs, update your fecking addons!] ) and throw some Servo into the mix, those who have remained loyal for an underwhelming two thirds of a decade will finally be rewarded.
I just hope they aren’t going to butcher their extremely good layout and style code. As a web dev, Firefox is still the king for rendering both according to standards and as I expect. The cache is also much more sensible than Chrome’s cache. It’s been bad enough that Mozilla has butchered the dev tools situation by halving it. Now the newer native tools are incomplete and nobody is updating the Firebug code either. Madness! If they change the rendering , I might as well adopt Edge which is, AFAIK, the most standards accurate browser beyond Firefox. That is, if I can manage to get it running on Windows 10.
My 17 addons work perfectly fine with e10s. Then again, I’m using Aurora, so maybe your issue is that you’re using ‘stable’ Firefox which doesn’t support addons with e10s (yet). I’ve put ‘stable’ in inverted comma’s because Aurora is stable for me, no crashes.
What do you mean by halving it ? If you have links that talk about this, I’m curious.
‘The components that will likely make it into Firefox in END of 2017’ is correct sadly.
Therefore rendering will be at least 12 other month slow. So the fully ‘servo’ browser will be anytime in 2018.
Google’s performance tuning in Chrome till this time won’t hold on.
I think it’s too late for Mozilla.
What are you talking about.
1/ Firefox isn’t slow, unless you consider Chrome slow because mostly they are equal. It is gonna be *faster*, that’s not the same thing!
2/ e10s is a huge increase in speed and it is as we speak released to an increasing portion of the Firefox user base. It should reach 100% by what, Firefox 53 ?
3/ Chrome will not lose market share as long as Google keeps pushing it, but Mozilla can stay second. In fact it is second now, I think. The question is, will Chrome keep increasing until it reaches 90% market share ? Nobody wants that, not even Chrome fans.
4/ Market share isn’t what matters most in terms of survivability. Flat users matter most for this. Mozilla should have around 500 millions of those.
You’re 4th point is often missed. Apple has done very well as a distant “#2” to Microsoft in the desktop market. They have no need / interest / intention to become #1 if they are making tons of money and have a loyal user base. The same can be said for Linux, which is an even more distant #3. What do Linux developers and users care, as long as they can get what they want out of the system?
Firefox doesn’t need to return to its historically dominant market position to be successful. It just needs to stop the bleeding that has been going on for some years now. The goal should be to target the niche demographic of users who care about privacy, customization, and open source. If they can do that, the project will continue for many years to come.