Chrome Beats Edge and Firefox in 'Browser Benchmark Battle: July 2018' -- Sometimes (venturebeat.com)

An anonymous reader quotes VentureBeat: It's been more than 20 months since our last browser benchmark battle, and we really wanted to avoid letting two years elapse before getting a fresh set of a results. Google Chrome, Mozilla Firefox, and Microsoft Edge have all improved significantly over the past year and a half, and as I've argued before, the browser wars are back. You can click on the individual test to see the results:

SunSpider: Edge wins!
Octane: Chrome wins!
Kraken: Firefox wins!
JetStream: Edge wins!
MotionMark: Edge wins!
Speedometer: Chrome wins!
BaseMark: Chrome wins!
WebXPRT: Firefox wins!
HTML5Test: Chrome wins!

Chrome looks to be ahead of the pack according to these tests. That said, browser performance was solid across all three contestants, and it shouldn't be your only consideration when picking your preferred app for consuming internet content.
Chrome wins in four tests, beating Edge's three wins, and Firefox's two wins.

Comments

obviously

[AlexanKulbashian]

If Firefox and Edge inhaled as much memory as Chrome, i'm sure they might also perform a little better

Re:obviously

[Z00L00K]

Today I'm usually less worried about performance and more about feedback, privacy, security and stability of the browser.

What I really miss in the Microsoft browsers is the feedback about what the heck it's waiting for - or if it's still waiting for a server response.

Re:obviously

[thegarbz]

If Firefox and Edge inhaled as much memory as Chrome, i'm sure they might also perform a little better

Then they should do it. RAM is cheap. It's in my system to be used as fast as possible. Cache the Jeebus out of websites if you have to. My server has 32GB of RAM currently with 31GB actively used for by the filesystem for caching.

Low memory footprint is something for RaspberryPis, not desktops. Give me speed!

Browser benchmarks

[Anonymous Coward]

Please tell me again why time-based browser benchmarks matter at all, when the differences are measured in milliseconds.

Re:Browser benchmarks

[gweihir]

They don't. But since most people (including the ones publishing such benchmarks) really do not understand anything about browsers or what matters and what does not, we will get these demented comparisons for the foreseeable future.

Re:Browser benchmarks

[Dutch Gun]

One issue that matters to me is responsiveness and perceptible speed of the browser UI. In my experience, Firefox beat out Chrome in this regard, as the Chrome UI had a few noticeable stutters under heavy load, while the Firefox browser stayed responsive. That's probably not going to show up in a benchmark of any sort, as it's a very subjective experience.

I had briefly considered switching to Chrome after Mozilla pulled it's "Mr Robot" plugin stunt, and so I tried it out for a while. It was a pretty slight margin, but Firefox just *felt* faster to me, likely because of UI responsiveness. But beyond that, I missed a few of Firefox's minor quality-of-life features. Edge seems very much a take-it-or-leave-it experience. Firefox is becoming more like that, but still not as much as with Chrome.

And whatever problems Mozilla may have, and whatever idiotic decisions they still make, I still trust their motives more than Google or MS.

Re:Browser benchmarks

[mcswell]

The reason is...well...it's because...err...well, you know...um...I dunno.

All seriousness aside, my choice of browsers (Firefox) is because it's easier to use, because it has a _real_ menu system, and because I can make configuration changes that I need.

Chrome has none of this. I can't use the backspace key because G, in its wisdom, knows better; I might some day be filling out a form, and accidentally hit the backspace key, resulting in backing out of the form instead of erasing a single character. (That's never happened to me using FF.) Yes, does the same thing, but that's a *much* less convenient set of keystrokes.

And I can't open a new tab in Chrome using FT, I have to mouse some icon. (Yes, thank you, I am a keyboard user.) I suppose there's some other keystroke in Chrome, -something-or-other, but I don't know what it is, and anyway it conflicts with other uses of keys.

My bookmarks: I imported my Firefox bookmarks, but they're buried somewhere way down in Chrome's lousy hamburger excuse; again, maybe there's some way I can quickly get there, but I don't see it. Bookmarks Bar, you say? Let's see, the reason for removing the Title Bar was to give me more vertical space for the web page I'm reading. So I want to give that space back for a lousy toolbar? Why can't I have the bookmarks *directly* accessible from a menu? Any reasonable application (like Firefox) allows you to do that.

At work, where my choices are Internet Explorer or Chrome (which is the only reason I have any experience with Chrome), I go to Sharepoint (yuch, but I'm stuck with it) and try to open an Excel doc. It comes up inside Chrome in some kind of marginally useful spreadsheet-in-browser applet. I can never do anything I want in that applet, so I invariably tell Chrome to open the spreadsheet in a real copy of Excel. Extra keystrokes to get there, despite the fact that I always know I'll need to do it.

None of the above would be a problem if Chrome offered any options. Redefine keystrokes? Nope. Rapid access to my bookmarks? Nope. Automatically open Excel files in Excel? Nope.

Who cares if Chrome is marginally faster at opening web pages? Not me!

Re:Browser benchmarks

[tsa]

About FF's quality-of-life features: you know what I hate? That if you accidentally or purposefully close the last tab, the whole browser just disappears. All browsers except Firefox do this. That alone is reason enough for me to keep using FF.

Re:Browser benchmarks

[MrL0G1C]

You probably set an option aeons ago, I think a fresh Firefox install also closes the browser. But still, Firefox (Waterfox here) gives you the option.

I prefer Waterfox because it allows me to keep the classic look whilst having multiple processes. Which leads to the question - if Waterfox can do this then why can't Firefox?

Re:Browser benchmarks

[tepples]

What else is the browser supposed to do visually when the user has closed the last tab? Keep the window open with no tabs in it? Keep the application open with no windows open? Other than macOS, does any major PC GUI operating system even have a concept of an application remaining open after its last window is closed?

Re:Browser benchmarks

[thegarbz]

One issue that matters to me is responsiveness and perceptible speed of the browser UI

The browser UI? Why are you using the browser UI? Do you use the Windows UI or the applications within it? :-)

I know what you mean, but these days the interaction I have with things are very much within the rendering window of the Browser. The Browser is a glorified alt+tab alternative. Increasing Javascript performance means increasing UI speed of things people actively use including Office365, Gmail, Facebook, etc.

So?

[Anonymous Coward]

I am not entirely concerned with the speed of the browser, at this point the speed difference is in the unnoticeable order of milliseconds.

My primary concern is advertisers and bitcoin miners tracking me or utilizing my machine. In this regards firefox wins hands down because Google and Microsoft both have vested interests in making a dirty browser which serves them but appears to serve us the user.

Until these companies change their souls to be better citizens within our society I will never allow their products upon my machine, it is too much unknown risk for no reward. They have only themselves and their actions to blame for burning through their goodwill karma with the public, google most especially given how beloved it once was.

Re: So?

[im_thatoneguy]

Microsoft has a massive financial disincentive to monetize the small edge revenue possible vs the huge cost of people ditching Windows.

Hmmm

[hcs_$reboot]

Safari missing? Nice of you, it is based on good intentions I presume.

Re:Hmmm

[DontBeAMoran]

And to us, Edge is the rounding error that only Windows 10 weirdos would ever want to use.

Re:Hmmm

[DontBeAMoran]

Maybe he's a Linux nurd himself?

Re:Hmmm

[dk20]

according to this, there isnt a whole lot of OSX / IOS safari users.

https://www.netmarketshare.com...?

on the desktop windows is like 88% with OSX being arond 9% (still much higerh then the "linux weirdos" as was posted earlier.

on mobile devices, android is 70% vs 26% ios.

Clearly no one "wants" to use Edge.. unfortuntly like all instances of windows, there really isnt a way to uninstall that peice of crap.

Re:Hmmm

[Shirley Marquez]

It is impossible to do an apples to Apples comparison between Edge and Safari because they don't run on the same OS. These benchmarks chose to concentrate on the OS with the larger market share.

Re:Hmmm

[thegarbz]

I tired Safari once, but I couldn't figure out how to install it on Windows 10.

Re:Hmmm

[hcs_$reboot]

VirtualBox

Re:Hmmm

[thegarbz]

I'm sure that will do wonders for its performance.

Who cares?

[Anonymous Coward]

The thing that need battling, is the bloated "web"!
An ad blocker was the biggest speed improvement that actually meant something in the last years.
Blocking stupid visual effects (like 3D animations and background videos) should get the web back to mostly smooth even in the slowest browser.
And the most ideal solution, not abusing the web to run applocations in the mother of all inner-platform effect software design anti-patterns, would mean a super-smooth web experience even in Mozilla 0.6 from 2001!

Re:Who cares?

[tepples]

And the most ideal solution, not abusing the web to run applocations in the mother of all inner-platform effect software design anti-patterns

What would you recommend instead? Writing the applications in Java and requiring everyone to download a Java VM? Writing the applications in Python and requiring everyone to download a Python interpreter? Writing the applications in Swift or Objective-C and requiring everyone to buy a Mac?

Not sure why

[rsilvergun]

but benchmarks or not Chrome "feels" faster than Firefox these days. As for IE, it's JavaScript compiler is dog slow so pages take forever to load, making the entire experience pretty rough since pages still transition a lot (AngularJS and one page sites haven't really taken over like everybody thought they would...)

Re:Not sure why

[Anonymous Coward]

Not generally true anymore. Chrome feels more sluggish then Firefox, Safari, or even Edge for a looooot of users. And not because of "crufty profiles" or "bad drivers" or "too little RAM" or other classic excuses, either. It's frankly starting to show its age, even if it still has a lot of life left in it.

In fact Google is even artificially crippling other browsers by using not-ready-yet technologies in their own web services which only Chromium supports well. See for instance their use of Web Components in the newest YouTube redesign, which makes it much slower and more RAM-intensive on other browsers.

Long gone are the days where Chrome was genuinely king on its own merits, just like the days when Google cared to make cross-browser web apps instead of barely paying lip service to any browser but their own. About the only non-artificial benefits it has left to its own name now are site isolation and their JS debugger.

Re:Not sure why

[KiloByte]

Sounds like AdBlock/uMatrix fail. No one cares about that hundred megs of Javascript crapware if you don't load it.

Re:Not sure why

[SNRatio]

Could you make a Firefox plugin that replaces stock images on websites with the top result of a google image search for "stick figure" plus the title of the stock image?

"stick figure" Trump. "stick figure" Kanye.

The web would be so much better.

Who cares? Seriously...

[gweihir]

Browser speed is not a factor today. The only reason for these comparisons is that most people do not have what it takes to do more than compare a (meaningless) number.

Re:Who cares? Seriously...

[thegarbz]

Browser speed is not a factor today

Ahhh someone who still uses MS Office. Get with the program man the browser Javascript speed is directly related to how complicated and fast the modern version of the "application" works. If Office365 can have less UI lag then go for it.

If this were a discussion of Linux kernel outperforming the Windows kernel you'd be all over that. Time to give up on the idea of a web browser showing some text on a screen and time to face the reality of modern computing that is the glorified thin client running a browser based "OS"

Re:Doesn't Matter

[gweihir]

You must be one of these "cyberterrorists" I keep hearing so much about! Nobody else has anything to hide, after all. I think they should lock you up immediately, or better, "accidentally" kill you on arrest.

IE 6?

[Billly Gates]

What about IE6 performance? It's the PHB ultimate browser as Chrome makes his internet's sites on the LAN look funny

Interface is more important

[Stormwatch]

A difference of microseconds doesn't matter if the program has a shitty user interface. Firefox is the only one of those three that has a proper interface -- after you restore the title and menu bars, that is. No, cramming the menu bar into a goddamn hamburger button is never acceptable for a desktop application.

Re:Interface is more important

[williamyf]

Amen.

At least in mac, the menu bar is there by default all the time (os imposed), so there is nothing to restore in that front.

HAving said that, I like my tabs in the title bar (even with the 16:10 display of the mac, vertical space is at a premium these days).

On the other hand, I hate how firefox dicks around with the buttons on the interface. Lucky for me, I am in the ESR channel, so this happens only once a year, not every six weeks (chrome), or every six months (edge).

Re:Interface is more important

[theCoder]

With vertical space at a premium, you should look into the Tree Style Tab extension for Firefox. It puts your tabs down the side, and also organizes them in a tree structure, so that new tabs open under the tab they were opened from. You can collapse groups of tabs as needed. Personally, I don't know why all the major browsers keep tabs on the top. With widescreeen monitors (at least on desktops/laptops), tabs on the side makes much more sense.

Of course, I had to stay on FF56 to avoid losing my TST extension, though I think the latest version of FF re-enables some features to allow it to work properly again.

Re:Interface is more important

[tepples]

With widescreeen monitors (at least on desktops/laptops), tabs on the side makes much more sense.

Not if you're splitting your 1920x1080p display down the middle to show two 960px-wide windows.

Re:Interface is more important

[Stormwatch]

It's kind of pathetic that 1920x1080p is still standard when cell phones have better resolutions.

Re:Interface is more important

[thegarbz]

What interface? All you need is tabs and a scroll bar. The rest of the interface is dependent on the "apps" that you run in your browser these days, and Office365 looks the same on Firefox and Chrome.

Can we just agree that it doesn't matter?

[DalM]

There are a few websites -particular specialty government websites- that don't work right with this browser or that one, but for the most part, everything works fine regardless of browser.

So who cares?

Firefox isn't Chrome or Edge

[Anonymous Coward]

So Firefox WINS!

Re:One word

[hcs_$reboot]

web developers target two browsers: Chrome and IE

IE? Thanks for the fun!

Matters?

[markdavis]

They go to all that effort to benchmark and don't even list the versions of the freaking browsers. Plus, only MS-Windows... no Linux, no MacOS. (And they didn't use a zero scale on several of the graphs.)

In any case, I am not sure it matters. Looks to me like all three were fast. Other factors probably matter more now...

I am much more concerned with using a browser that is truly open-source, multiplatform, tries to respect privacy as much as possible, and community driven. I guess you know which browser I am using (because only one of those three matches).

Re:Marketing value of browsers

[darkain]

And then there is Opera: most supported rendering engine (Blink), with the advantage of having ad-block and VPN built in without plugins.

Re:An extension benchmark would be better.

[KiloByte]

Depends on the benchmark. It would completely fail if security,

Bullshit. These days, conventional exploits hardly matter -- blackhats (NSA and Russian mafia) don't reveal their toys, so these get fixed only when noticed by someone else, and that happens with similar speed for all major browsers. What counts is prevention, and for that you need powerful extensions. These exist only for XUL as webext doesn't expose sufficient APIs to allow blocking crap. Ergo, Waterfox and Firefox <=56 run circles security-wise around Google-Spyware, Microsoft-Spyware, and Firefox Quantum.

I don't run an ad blocker

[rsilvergun]

in either browser. If ads are annoying me I stop going to the site.

Re:I don't run an ad blocker

[KiloByte]

These days, most sites don't have a single source of annoying ads -- they have several. There's no option but to block those, and the war against spammers (ads are a form of spam) is going so badly that opt-out blocking is a losing proposition, you need opt-in.

Speed isn't the biggest problem

[sremick]

So basically some Chrome fanboy is paranoid about articles such as the one at The Verge which outline the far bigger problem with Chrome, and so it trying to deflect attention: "Hey, look! Chrome is 0.001ms faster than Firefox (sometimes)!"

https://www.theverge.com/2018/...

It seems people have started to just lazily accept the browser monoculture of a Chrome-dominated web with no acknowledgement that this is the exact same nightmare we went through back in the day with Internet Explorer. We're seeing underhanded and sneaky bundling deals and drive-by trojan installs (which then make Chrome the default), as well as a push for proprietary markup that only works with Chrome and subsequently websites that REQUIRE Chrome for use.

Why are we giving Google so much of a free pass for all the things Microsoft got raked over the coals (and taken to court) for? It seems peoples' memories are quite short, but I can assure you that a lot of us are STILL dealing with the fall-out of websites requiring Internet Explorer to this very day.

Re:Speed isn't the biggest problem

[tsa]

Not only are People's memories short, but there is a whole new generation of people who never saw the monkey run.

What I care about is

[aliquis]

1) Which one slows my computer down the most with most tabs.
2) How easily can I save and shut down all those tabs and get back to a useful state.

Category 1 goes like Chrome, Opera Beta, Firefox, Edge or something such. Hard to decide between the first two.
Category 2 I don't even remember which one let you save all tabs at once even through multiple windows but basically with extensions they all do it well.

What I'm using now is Opera Beta though with V7 sessions. I know for sure V7 sessions has felt slow after using it many times and Chrome has felt slow with all the hundreds of thousands of bookmarks so .. Not really great anyway.

But that's what matter for me. Not how fast they render one webpage or run some benchmark application in one.

Re:Doesn't Matter

[gweihir]

Irony challenged AC is irony challenged.

Re:An extension benchmark would be better.

[serviscope_minor]

These exist only for XUL as webext doesn't expose sufficient APIs to allow blocking crap. Ergo, Waterfox and Firefox

Given I'm running uBlock origin and NoScript on Firefox 60, I suspect your assessment may not be accurate.

Wetware/carbonware bottleneck

[williamyf]

The wetware or carbonware that is put between the Keyboard and Chair is the main bottleneck in performance.

If you change many things in the browser, like the interface, or break a lot of plug-ins, the wetware/carbonware gets adapatation problems, and productivity suffers.

Chrome changes every six weeks or so.
Edge changes every six months or so.
Firefox ESR changes every year.

So, logic dictates, go with firefox ESR.

Firefox never in third place

[PineHall]

An interesting detail I noticed in the nine tests that Firefox was in second place except for the two times that it was in first place. It was never in third place. I think that means Firefox is solid across all tests. Looking at first place finishes only, you see that Firefox ends up in third place, but if you add up the rankings Firefox ends up in first place. I think all the browsers are pretty good so maybe these tests are not that big of a deal. I think the point is that Firefox and Edge have improved so much that Chrome not the must-use browser that some people used to claim. I use Firefox so I am happy to see that it is performing well.

nurd

[williamyf]

Maybe nurd is a portmenau of nerd and turd?

Having said that, at one time or another I have used Windows mac and Linux on my desktop, and better not tell you about the weird stuff I administered in my servers (Sinix or VMS anyone?).

So, I do not care what OS anyone uses, as long as they do not pontificate about it.

OSs and software in general are tools, use the best tool for the job at hand, and get done with it.

You never see people arguing if a screwdriver is better than a chisel... It should be the same about OSs

Re:nurd

[tsa]

But some screwdrivers are better than others. Magnetic, hardened steel tips and so on. It's the same with OS's.

When You Block Javascript

[Bing Tsher E]

Which browser performs the best when you block javascript? Which browsers make it the easiest to set configuration rules that blacklist third-party scripting? Which browsers then gracefully display the 'corrected' web content after the script infestation has been removed?

That is the kind of benchmarking some of us would find more relevant than gee-whizz speed measurements of the excrement that "web developer" discharge.

Re:Oh good, more useless benchmark contests.

[Excelcia]

Not sure why the parent was modded down. This isn't inaccurate. When was the last time that browser rendering speed was the rate limiting step in how fast something displayed. Or, if it was the rate limiting step, when was the last time that the difference in rendering speed between two browsers was even relevant in a time scale beyond an eyeblink?

But which one uses the least memory

[halfdan the black]

I couldâ(TM)nt care less if a page takes a microsecond longer to load. What I care about is not chewing up all the memory on my machine because of sloppy coding and memory leaks.

Browser wars

[Clived]

What about Opera ? I seem to be using it a lot these days, love the built in VPN and its quite fast

Comments ?

Re:Browser wars

[Anonymous Coward]

What about Opera ? I seem to be using it a lot these days...

Last I heard, Opera stopped using their own distinct engine in... was it 2013? Thereabout. They switched to the same engine as Chrome, so there's no point in making Opera a separate contender in browser tests anymore.

Re:Browser wars

[thegarbz]

Comments: I'd rather an American company hoovering all my data up than a Chinese one.

Who wants a browser

[AHuxley]

made by an ad company to ensure the consumer gets their ads?

Cut the cheese

[Mats Svensson]

This is just the beginning!
After all the needless things, like bookmarking, navigation, GUI, etc has been removed, FF will really fly!
And as an added bonus, It will report itself as "Google Chrome".
Its what the kids want!
Right?
Right!

Safari is 2nd market share overall, 1st on mobile

[taharvey]

You have data problem. Region, education, occupation, access, software quality, and income all factor into usage.

So what you find is in the USA. Chrome is 47% market, followed by Safari at 31%. Everything else is in the noise.
http://gs.statcounter.com/brow...

If you restrict that to mobile devices. Safari is 50% market, followed by Chrome at 41%.
http://gs.statcounter.com/brow...

The problem with your data is that selling lots of $50 android phones, and cheap windows boxes doesn't mean people use them. Also, given the professional, corporate, and tech industries strong leaning toward Macs over Windows (even the Microsoft office I was in recently was all iMacs) - they will use the web 10X more than joe average.

Safari way faster... and left out

[taharvey]

For several years now Safari has outpaced Chrome nearly across the board in javascript and DOM operations.

I've been tracking it as a web app developer because it has serious implications for UX on mobile devices. The DOM operations can easily be 3-4X faster in safari, when combined with the iPhones processing advantages stacks up a 10X difference in performance between the average iPhone and the average android phone. Its a big problem for javascript app developers.
examples:
https://bugs.chromium.org/p/ch...
https://discuss.emberjs.com/t/...

The fact the Safari is left out of these comparisons puts a reality distortion field on the market, and keeps Google from getting their act together.

Re:Safari way faster... and left out

[Roger W Moore]

For several years now Safari has outpaced Chrome...

It needs to because it has to load all the ads. I never used Safari because of the absence of a decent adblocker.

Re:Safari way faster... and left out

[Shirley Marquez]

You seem to be discussing mobile devices. iOS is a special case; all competing browsers are guaranteed to be inferior to Safari because Apple requires that they use the HTML and Javascript services provided by iOS, which are basically an out of date version of Safari. The choice on iOS is easy because Apple has rigged the game.

The benchmarks cited here are on systems where all the browsers they tested run natively - that means Windows. Safari for Windows was discontinued ages ago so it's out of the picture. It would be interesting to see how Chrome and Firefox compare with Safari on a Mac, a platform where all three get to run their own code, but then Edge would be left out because it's Windows-only. A comparison of Chrome and Firefox on Android and Linux would also be nice to see. (Neither Edge nor Safari is available there so it would be a two horse race.)

OS Dependent

[Roger W Moore]

They should also average all the browser performance over Mac, Windows and Linux or state which OS the results are valid for. It is unlikely that Edge will win any speed contest outside Windows because it will have to run inside a VM.

Re:OS Dependent

[DontBeAMoran]

Only test browsers available on the OS:
Windows: Edge, Chrome, Firefox
macOS: Safari, Chrome, Firefox
Linux: Firefox, Chrome

Browser Benchmark Battle: July 2018

[najajomo]

Is there any difference between Apple, Linux and Windows in running these tests. Does the fact the most of Edge is baked into the kernel add anything to it's performance?

Safari costs $499

[tepples]

Safari costs $499 (source: Best Buy). Most people (except professional web developers) aren't willing to buy a second computer or second phone just to run a web browser. These comparisons include Edge despite it being exclusive to Windows 10 because Windows 10 has a much larger installed base than macOS and X11/Linux.

Re:Safari costs $499

[DontBeAMoran]

But that doesn't make Edge free either.

Re:Safari costs $499

[tepples]

I concede that Edge is not free software, unlike Chromium and Firefox. I concede that acquiring an Edge license is not without charge, unlike Google Chrome and Firefox. But the fraction of PCs worldwide that are already licensed to run Edge is far greater than the fraction of PCs worldwide that are already licensed to run recent Safari. The Edge fraction is above the threshold, the Safari fraction not so much.

"Never bothered implementing" features for years

[tepples]

Chrome isn't based on WebKit any more

Chrome for Chrome OS, X11/Linux, Windows desktop, and macOS isn't based on WebKit. Chrome for iOS is still based on WebKit, as is every single other browser that runs on iOS.

If your website doesn't work on WebKit, it's probably because you're using new standards that WebKit either never bothered implementing

Correct. Web developers are complaining that WebKit "never bothered implementing" things that Firefox and Chrome have long supported. It was last to get WebGL, for example. It still doesn't play WebM, as far as I'm aware. Even something as old as <input type="file"> was completely unsupported in Safari for iOS for five years.

Re:One word

[tepples]

Almost everybody on mobile uses Chrome, which is not stuck with Webkit.

As I understand SuperKendall's posting history, he's referring to the "sheer number of mobile devices as well that" are "stuck with WebKit". StatCounter claims that Safari and other WebKit wrappers made up half of all mobile page views in the United States over the past month: 50.84% Safari, 41.63% Chrome, 4.88% Samsung Internet, and less than 1% each for Puffin, Firefox, UC Browser, Opera, etc.