Web Browser Grand Prix 4: Firefox 4 Goes Final

[mitch074]

To commenters:
- Fx4 is the most stable browser out there, even without tab isolation; session restore is useful in case of crashes, but also in case of abrupt computer shutdown - something that tab isolation doesn't solve.

- since 3.6.4 isolated Flash in a separate process, I've yet to break an official build (dailies or distro builds are another matter).

- most people don't go and spend hours setting up hardware acceleration, because as many said, browsers are so fast now that it's hardly visible. However, Fx4 can load an anti aliasing calibration tool, which solves a problem that IE9 users can't beat (and other browsers don't see): font fuzziness in hardware accelerated contexts.

- Do Not Track is not an ad blocker: it merely asks advertising networks to show adverts without trying to track users' habits. AdBlock+ is greatest on Firefox, eventhough the Chrome version got better. Fx4 does feature DNT.

- IE9 is available in Vista SP2 and Win7 only; Firefox is available, hardware acceleraed, on Windows 2k/XP/2k3 too. Hardware acceleration isn't available per se in Linux, howeve if one considers the speed of Fx 3.6 with XRENDER, it's not that much of a problem.

Fx4 isn't the fastest browser (but it sure is close). It's not the trendiest (but its UI and HTML5 support sure are good). It may not be the safest (but it sure allows safe browsing habits with NoScript). It's not always the lightest (eventhough it scales very well to demands and available power). But, dang it! It sure is the most useful.

 

[cjl]

[citation][nom]haroem[/nom]@luckyducky7"Adblock Plus, Adblock Plus, Adblock Plus.That pretty much sums it up. No amount of optimization will help you when it comes to loading web pages if you have a large amount of adframes and such to slow you down.So have your super-fast IE9: I bet the ads will appreciate loading quick too."Not if you use tracking protection.[/citation]
Does anyone else have a problem with tracking protection? I'd really like to give IE9 a serious try (I really would), and I like the interface and speed, but no matter how I've tried to configure it, tracking protection never seems to do anything. I've added a variety of tracking protection lists, I've made sure they were enabled, and at the end of the day, not a single damn thing was blocked. Until I can get that working, it's FF4 with adblock plus for me.

 

[Guest]

If this test shows anything, then it's probably that right now you can use just about any recent "big" browser you like.
As if anyone is REALLY going to notice the speed differences between them.
For the average webpage today it just doesn't matter.

Get the one you like best/that's most stable for you and keep it.

 

[Tomtompiper]

I can't tell if you are correct about IE9 as it is not available for any of my systems, where as I use Firefox on them all, XP, PclinuxOS, Mint and Android. So why have you promoted a niche product to Approved status?

 

[ohseus]

If MS couldn't make their browser run the fastest on the OS they also make that would be truly sad. Really isn't this article basically saying Ford transmissions work best with Ford engines?

I'll stick with Firefox. Ad Block and No script may have similar options in other browsers. What about "Flash Block" and "Down Them All"? Foxsync is another reason to use FireFox. I can sync between Windows and Linux, can IE?

SO... IE is out, I trust Google less than I trust Rosanne Barr with a box of donuts and have never really used Opera that much. (Though I have several Opera CDs)

 

[aznjoka]

Good review! IE9 is doing well wish I could install in my linux systems, Opera is doing good as well, there's a surprise

 

[Guest]

I really like how Firefox can open up many tabs at once without falling over and dying - When reading articles like this, i mash my middle mouse button on all the links on page 1, and open all the pages at once.

Firefox - They get tabbed browsing right. The ability to customise everyhting is great too. And seeing as i generally operate with 10-20 tabs open at a time, Firefox is definitely the way to go, at least in my case. The other browsers start to chug at that point.

 

[dfusco]

After suffering through the vast and mighty suckage that was IE 6,7 and 8 I don't even need to try the sure to be awful IE9.

 

[Brazilian Joe]

I have found that after a long time, with a heavy usage scenario, FF (at least the Mac version) starts to get a noticeable responsiveness lag, even when on a SSD.
It's way better, and I still like it, but I think I will keep using chrome preferentially for some time.

 

[epileptic]

Does anyone else have a problem with tracking protection? I'd really like to give IE9 a serious try (I really would), and I like the interface and speed, but no matter how I've tried to configure it, tracking protection never seems to do anything. I've added a variety of tracking protection lists, I've made sure they were enabled, and at the end of the day, not a single damn thing was blocked. Until I can get that working, it's FF4 with adblock plus for me.

--- If you don't care about how it works, just try using the first list here. Click the Add EasyList+EasyPrivacy TPL link. You might want to disable all other lists you already have.

--- Otherwise read below.

The lists Microsoft provide aren't used to block ads. They're used to prevent tracking only. You can write your own list and subscribe to it. The syntax is explained here. Here's a sample list adapted from this one:

[cpp]msFilterList
: Expires = 30
# blocked strings
- *-ad.cgi*
- *-ads/*
- *.2o7.net*
- *.ad-*
- *.ad.*
- *.adbrite.*
- *.adrevolver.*
- *.ads.*
- *.adserver.*
- *.adtomi.*
- *.advertising.*
- *.atdmt.*
- *.banner*
- *.blogads.*
- *.buysellads.*
- *.casalemedia.*
- *.click*
- *.directtrack.*
- *.doubleclick.*
- *.etology.*
- *.fastclick.*
- *.google.*/adfetch*
- *.googleadservices.*
- *.googlesyndication.*
- *.kontera.*
- *.linkbuddies.*
- *.mochibot.*
- *.overture.*
- *.pheedo.*
- *.qksrv.*
- *.quantserve.*
- *.roar.*
- *.slickkicks.*
- *.tradedoubler.*
- *.tradepub.*
- *.trafficzap.*
- *.tribalfusion.*
- */accresults.*
- */ad.*
- */ad/*
- */ad_*
- */adbot.*
- */adc_*
- */adclient.*
- */adcouncil/*
- */adframe.*
- */adgifs/*
- */adgraph/*
- */adimages/*
- */adinfo*
- */adlog.*
- */adlog/*
- */adrotator.*
- */ads.*
- */ads/*
- */ads_*
- */advert*
- */adview.*
- */affiliates/*.js
- */banner/*
- */banners/*
- */hotfreebies.html*
- */housead/*
- */liveads/*
- */pagead/*
- */phpads/*
- */pop.cgi*
- */pop.htm
- */pops/*
- */poptest.*
- */popup/*
- */printads/*
- */redir.asp*
- */skyscraper/*
- */softad/*
- */sponsor/*
- */sponsors/*
- */tw/adt*
- *120x240*
- *120x600*
- *120x90*
- *160x600*
- *234x60*
- *336x280*
- *468x60*
- *728x90*
- *_ad.js
- *_ad_*
- *_ads.js
- *_ads_*

- *_advert*
- *_adx_*
- *_banner_*
- *_borders/*
- *_superad*
- *a.p.f.qz.*
- *a.r.tv.*
- *ad-flow*
- *ad.trafficmp.*
- *ad_type*
- *adbot*
- *adclick*
- *adclix*
- *adclub*
- *adcycle*
- *adflight*
- *adframe*
- *adimage*
- *adknowledge*
- *adlink*
- *adlogix.*
- *admaximize*
- *admex*
- *admonitor*
- *adpulse*
- *adrunner*
- *adserv*
- *adsoftware*
- *adswap*
- *adtomi.*
- *aureate*
- *avenuea*
- *banner.*
- *banners.*
- *bluestreak.*
- *burstmedia*
- *burstnet*
- *clickxchange*
- *darkblue.*
- *darkbluesea.*
- *dbbsrv.*
- *exitpopup*
- *flycast*
- *focalink*
- *headerpopup*
- *hitexchange*
- *hitlist*
- *hitsites*
- *houseads_*
- *i.us.rmi.yahoo.*
- *imaginemedia*
- *intellitxt*
- *jsads*
- *linkads*
- *linkexchange*
- *linkpopup*
- *linkshare*
- *linksynergy*
- *media.fastclick*
- *paypopup*
- *popieen.*
- *popme.*
- *popunder*
- *popupad*
- *ps.interpolls.*
- *radiate*
- *secure.webconnect*
- *smartsize_*
- *spinbox.versiontracker.*
- *spylog*
- *subs.timeinc.*
- *toolbar.google.*
- *trafic.ro/*
- *us.a1.yimg.*
- *us.f.yahoofs.*
- *valueclick*
- *x.mycity.*
- *z.about.*
- *zdmcirc*[/cpp]

Put that in a file called mylist.tpl and then use something like that to install it:

[cpp]<button onclick="javascript:installTPL();">Install myList</button>
<script type="text/javascript">
function installTPL() {external.msAddTrackingProtectionList('mylist.tpl','myList');}
</script>[/cpp]

You can add this code somewhere on your website if you have one otherwise just create a simple HTML file and put it there. I don't know if there's an easier way to get your own list in there.

 

[emergancy exit]

i think they should do these browser tests on a reasonably priced laptop instead of a supercomputer lan gaming machine/workstation.

reason being is that on a really fast computer it puts all the browsers really close together on speed and the main function of a fast computer is not to brows the internet but run other software.

if toms used regular laptop or entry level consumer pc for these tests maybe people will take your review more seriously? i say that becasue the numbers for the most part on all of the tests are really close together i want to know if that is because the browsers really have similer speed or if the system is carrying the slower ones along?

 

[emergancy exit]

just saying they should use the types of computers most consumers buy and only use to browse the internet with instead of a gaming machine. there are millions of people who buy laptops and desktops each year and they basicly only use those computers to browse the internet with and thats prettymuch it use a machine like that for a test like this one

 

[mosox]

[citation][nom]lancelot123[/nom]Are Firefox fanboys and others really too stupid to realize that IE9 (as well as other versions) have adblocking "addons" you can get? Too bad ie7pro will never work with IE9, I loved that addon.[/citation]

Yeah a list with some ads and may are not. Didn't use IE in a few years, does it block scripts?

One can't compare open source with proprietary, check out this:
https://addons.mozilla.org/en-US/firefox/

Thousand of them, doing everything from downloading videos to proxies and whatnot.

 

[slo]

I'd like to see maxthon included in the next test aswell. I have all browsers in the test installed on my pc but i still end up using maxthon for the last 4-5 years. I find it to be faster in real world compared to Firefox 4 (havent fiddled with the lastest IE version tho.) and it also has the easiest to use interface and really just makes you do your thing faster.

 

[hixbot]

With Tracking filtering, and ad blocking host files there's no reason to have ads on IE.

 

[rantoc]

It boils down to taste and versatility and currently IE9 sadly lacks in the plug in department even when its the fastest, when comparable ad block + no script becomes available i will for sure give it a real chance.

Until those basics are available its a no brainer for FF4 imo

 

[tipoo]

Did you enable Chrome GPU accleration in about:flags for the hardware accelerationt test?

 

[adamovera]

[citation][nom]epileptic[/nom]@stm1185 IE9 does support ad blocking. It's called tracking protection. You can write a tracking protection list that will block the ads you usually see or get a more thorough one from the web. Either way, that feature is there. I personally just converted my Opera list and rolled with it.@adamovera There's a mistake in your last chart. IE9 should be weak in HTML5 conformance and you're listing it as a winner.[/citation]
Thank you, I corrected it. The CSS3 and HTML5 conformance results were swapped in the final table.

 

[adamovera]

@Electro1984 and tipoo:
No, that flag was not turned on. Chrome 10 does not have hardware acceleration by default yet. I am aware that you can enable it, though the feature is clearly listed as experimental. Until Google turns it on in Chrome 11, it's not a final feature.

 

[tipoo]

I'm rather certain OSX isn't on version 1.7

Any word on GPU acceleration in Safari?

 

[davewolfgang]

Did you enable Chrome GPU accleration in about:flags for the hardware accelerationt test?

As was stated, the whole point of these tests was to run them on the "default" settings, with nothing changed from a normal install.

If the browser makers have this "wow" feature, then they should enable it by default. Most don't because they don't always work, thus causing crashes, thus causing lower satisfaction ratings.

I understand that you run some of these test from an in-house web servers to get "equality", but in the real world everyone here does have to deal with the "live" internet, and having the browsers trudge across the internet with all the other traffic would be a good test. And since you can all have them use the same IP from the testing site, that shouldn't be a bad thing and "considered" equal. My two-tenths of a dollar.

 

[adamovera]

[citation][nom]TA152H[/nom]Firefox is still a poor man's Opera. IE is still a poor man's Firefox. They all copy from Opera, anyway. Most innovations start there, and the lesser browsers pick them up later.But, let's be real. When you're measuring milliseconds, it's not that important. Who notices? The look and feel, and personal preference will always matter more than that. After trying to like Firefox for a week, I finally gave up. It's not bad, but Opera is just easier, and fits better. Some will prefer FF4 too. It's really personal preference. Where Opera falls on its face is memory, as indicated. Even after you close Opera, it seems to take longer to give up all the memory. Where this would help is reloading prior pages, probably, since they probably hold onto this data to speed that up. But, it's probably more bad than good.I haven't used IE9, but IE8 was so bad, I don't know how anyone could enjoy using that awkward, horrendous browser. It's just painful to use, in the true Microsoft tradition.The summary about how it's unavoidable that IE9 is the fastest is pure stupidity. It's beaten in a lot of tests, and how one decides to interpret that data, based on their own criteria of usage and importance could easily come up with a different conclusion. For example, I don't think it's too far fetched to say that something differing by less than 1/2 a second is irrelevant since it won't impact the human experience. The whole concept that all rankings are equal is wrong on two accounts; it ignores magnitude, and it assumes each test is equally important. Both are obviously fallacious. If TH wants to crown IE fastest on this crude basis, that's fine, because there's a basis, and everyone will have a different one. But, to say this conclusion is inescapable is ignorant and uninformed, and ignores even the slightest bit of statistical understanding.[/citation]
This series was started at a time when speed was a major issue, and three of the five Web browsers claimed to be the fastest - all three pointing to different benchmarks, of course. So, we decided to run every benchmark we could in order to find out which of the three could really claim to have the fastest Web browser. Going by the numbers, and using every test we could gather, IE9 is (for the time being) the fastest Web browser on Win7. BTW, the final table listing winner/strong/weak serves to balance out the magnitude and also only counts each category of testing once (no matter how many tests we have in that category). If you have a better solution, we're all ears, but so far the suggestions for scoring systems have mostly come from people who clearly have a dog in this fight, and they obviously favor one browser or another.

 

[adamovera]

[citation][nom]Tomtompiper[/nom]I can't tell if you are correct about IE9 as it is not available for any of my systems, where as I use Firefox on them all, XP, PclinuxOS, Mint and Android. So why have you promoted a niche product to Approved status?[/citation]
I too am in a similar situation. The vast majority of the time I use Linux but switch between just about every OS on a weekly basis. So that means Chrome, FF, or Opera. But Windows makes up 90%+ of the PC market, and Windows 7 is their newest product, so I can't really agree with you on the "niche product" statement.

 

[victomofreality]

if you award a point scale for each lvl google chrome actually comes out ahead of IE by 2 or 3 points....

 

[adamovera]

[citation][nom]emergancy exit[/nom]i think they should do these browser tests on a reasonably priced laptop instead of a supercomputer lan gaming machine/workstation. reason being is that on a really fast computer it puts all the browsers really close together on speed and the main function of a fast computer is not to brows the internet but run other software. if toms used regular laptop or entry level consumer pc for these tests maybe people will take your review more seriously? i say that becasue the numbers for the most part on all of the tests are really close together i want to know if that is because the browsers really have similer speed or if the system is carrying the slower ones along?[/citation]
That's a popular request and one that we'll be exploring at some point (probably not when we're trying to coincide a WBGP installment with a new browser release, though). Quick question: How low do you want to go? Laptop or netbook? Price range?