Web Browser Grand Prix 9: Chrome 17, Firefox 10, And Ubuntu

[forestie]

[citation][nom]mayankleoboy1[/nom]IE9 64 bit performs very bad in comparison to the 32 bit builds.For firefox/waterfox, on Windows, using 64 bit builds has the following1. Native performance increase due to 64 bit.2. Performance degradation due to the fact that the MSVC does not have the same memory optimizations for 64 bit as for 32 bit. so overall the experience of 64 bit FF/WF is the same as 32 bit builds.For 64 bit Ubuntu, you get the 64 bit FF by default..For a really great optimised FF, use PALEMOON. @AdamOvera : 32/64 bit should be clearly mentioned in the article.[/citation]
@mayankleoboy:
-haven't compared IE9 32 vs 64 because I use neither of them (as said I'm 99% a FF->WF user). You're stating IE9 64bits is much slower than 32bits counterpart, I'm ready to believe you but please provide facts, figures, numbers, hard evidence. This is surprising for me but why not after all?
-Regarding the performance decrease due to MSVC not having same memory optimizations for 64bits as for 32bits, again I'm ok to believe you but please provide facts. This too I find surprising because when you're optimizing for 64bits, you get not only wider datapath performance increase, but also you have less processors to optimize towards (subset of the existing 32bits processors) -> more targeted optimizations, new instructions, higher common denominator in instruction set / cache size / performance features etc.
-agree with your statement about Ubuntu 64bits, you get 64bits everywhere on the Tux platform. My email by the way was targetting Windows where it is not obvious whether 32bits or 64bits version of the browser is used.
-regarding Palemoon vs Waterfox, after reading articles and descriptions it would seem to be the opposite, Waterfox more/better optimized than Palemoon. Side advantage for Waterfox: closer to original FF source/features (at least for benchmarks, trying to compare apples to apples). And faster updates too for WF vs PM.

To conclude, another reason I would like to see more info/comparison between browsers including both 32bits and 64bits versions is that 64bits version has the potential to be more secure (64bits address space -> efficient ASLR feature whereas weak with 32bits address space, No-eXecute bit per default etc).

Please provide your facts I would welcome them.

 

[DigitalWarlord]

I would like to see the next "Web Browser Grand Prix" be test on all three OS platforms. It would be nice to see the comparison on a single test.

 

[Guest]

A good configuration of firefox may change benchmarks drastically. Most people use browser as they are delivered. Is this the case with this test also?
It gves a speed gain of 1000% if the browser is good configured and it maintains this speed everytime you restart.
Although I read Silverlight I personally digust this type of propretairy software. I rather like Real-media or ogg that is running on every O.S. without distinction.
Libreoffice 3.4 is far better than Microsoft Office (any version) too.

 

[tomfreak]

looks like firefox is only good on a machine like E-350 wheres GPU is a lot faster than the CPU. Hardware acceration on zacate ftw. On other platform like Sandy bridge I will stick with Chrome

 

[annymmo]

[citation][nom]mitch074[/nom]Firefox performance took a dive starting with version 4, where all hardware acceleration was disabled: before then, in version 3.6, XRENDER was used when available (it was 4/5th as fast as IE9 on the same PC) while it is now really slow - it's all software.Moreover, the only driver enabled for hardware acceleration on Linux is the Nvidia driver: according to Mozilla (and verified by yours truly on AMD and Intel hardware), most display drivers in Linux suck when it comes to 2D rendering - ouch. Note that Mozilla and Google could add shims to circumvent those bugs, but they don't -not worth the effort, especially when driver makers could fix their bugs rather easily, leaving the browsers broken yet again.[/citation]
The right way for this is to write unit tests.
The mesa devs can test constructions Google and Mozilla en fix the bugs before they become mainline.

 

[adamovera]

[citation][nom]mayankleoboy1[/nom]just wondering if use of a DX11 capable GPU will change scores in some HTML5 and other benchmarks as the browsers use DX11 assisted rendering.Also, AMD driver support in linux is poor compared to Nvidia. For future Linux articles, can you use a Dx11 based Nvidia GPU?[/citation]
I'm still working on getting comparable DX11 graphics cards from both vendors for the software articles.

 

[adamovera]

[citation][nom]PreferLinux[/nom]Who wants to guess that the poor Linux Flash and WebGL results were because Flash and WebGL don't use hardware acceleration with that graphics card and driver? I would be thinking so.[/citation]
[citation][nom]txsouthpaw[/nom]Why would one select an AMD/ATI GPU for a competition featuring web browser based hardware acceleration on Ubuntu? Also, I am able to view any of the webgl demos on the Chrome experiments page with ease (50+fps) using Chrome 17 on Linux Mint 12. How is it that webgl is listed as not being supported by Chrome in Linux?[/citation]
I went with the AMD/ATI card over the Nvidia card because 1) it's what we usually use in the WBGP, and 2) it was kicking the Nvidia cards ass in locally-installed commercial 3D gaming. In retrospect, I should have gone with the Nvidia solution for this article.

 

[adamovera]

[citation][nom]mayankleoboy1[/nom]how many of those top 40 sites use HTML5?i think that the HTML5 scores should be weighed by a factor of the percent of top40 sites that use HTML5.This way actual importance of HTML5 can be judged in real world.[/citation]
Probably none of them. Interesting suggestion, I'll take it under advisement for weighing the categories in future editions.

 

[adamovera]

[citation][nom]forestie[/nom]The OSes that are used are 64 bits but the browsers are mostly (all?) 32bits on Windows, and probably 64bits on Linux.Internet Explorer has 64bits builds on Win7, and Firefox has "almost" a 64bits browser on Windows too: Waterfox, which is a semi-official Firefox for 64bits Windows. Waterfox in particular claims huge improvements over base 32bits install, I would like to see how that translates into real-world.Not sure about availability of 64bits editions of other browsers on Windows.Here are my wishes:-clearly mention if the 32bits or 64bits version of the browser is used-where applicable and relevant, test with both 32bits and 64bits variants. I would like to see IE and FF split into 32 and 64 variants on Win for example.I personally migrated from FF to WF on my machines 3 weeks ago and find it noticeably faster in everyday use. WF is now my main browser.[/citation]
[citation][nom]mayankleoboy1[/nom]IE9 64 bit performs very bad in comparison to the 32 bit builds.For firefox/waterfox, on Windows, using 64 bit builds has the following1. Native performance increase due to 64 bit.2. Performance degradation due to the fact that the MSVC does not have the same memory optimizations for 64 bit as for 32 bit. so overall the experience of 64 bit FF/WF is the same as 32 bit builds.For 64 bit Ubuntu, you get the 64 bit FF by default..For a really great optimised FF, use PALEMOON. @AdamOvera : 32/64 bit should be clearly mentioned in the article.[/citation]
All the Windows browsers are 32-bit and all of the Linux browsers are 64-bit. It basically goes by what people actually use. Due to package management, 64-bit Linux users will always automatically receive the 64-bit builds, while (other than IE9) there are no official 64-bit builds for 64-bit Windows users.

 

[adamovera]

[citation][nom]Marcus52[/nom]This sentence makes no sense to me. how can it "win" where it "simply cannot compete"?[/citation]
While it wins in performance, in reality it cannot compete in that sector due to lack of titles.

 

[adamovera]

[citation][nom]blazorthon[/nom]We see these browsers, but the optimized versions seem to get no attention. For Firefox, it would be nice to also have a look into Waterfox and Palemoon. It would also be nice to see the TOR version of Firefox, but that's undoubtedly a lot slower.There's probably even more Firefox derivatives. On Ubuntu there's Minefield and many more.I use Comodo Dragon instead of Chrome and woulds like to see those two compared too. I'm not sure about it being faster, but it claims to be even more secure and it gives me somewhat more control over the browser. Importantly, it still works with Chrome apps I don't know of a modified Opera or safari, but there's probably many browsers similar to IE. The only one I remember is MiniBrowser, but there are more.Otherwise, it's a great article, except for the several typos. There were just too many! I stopped counting after three. Understandably, this is a large article so it has more chances for a typo to slip by proof-reading, but some of them were very inconveniently placed and I had to infer what they were supposed to be.[/citation]
Maybe in the future we can do a remix edition of the WBGP, but this one was for Linux. Sorry about the typos, I didn't get a chance to re-proofread this article one last time before it went live. Thanks!

 

[adamovera]

[citation][nom]adrianlegg[/nom]Only thing perhaps missing is how they handle a lot of big picture sites, Firefox does eat a tooons of memory (for like 30-60 big picture) while others seem to use only for those on screen.Beside that great review.[/citation]
Nice suggestion, I too have noticed Firefox loads images slower than Chrome, I see it all the time in my NewEgg/TigerDirect emails (which are essentially big pictures). Maybe I can figure out a way to adapt the page load/startup timers for image loads. Thanks!

 

[adamovera]

[citation][nom]spotify95[/nom]That was close! If it wasn`t for the strong finishes, Chrome 17 would have swept the board! I hardly saw Firefox at all in the winning column for windows 7. And, need I say this: MAZE SOLVER?! Firefox was awful at that! Im not using something that could be awful at some features.....Well, I might be with the minority, but it is Chrome 17 for me, I will never leave Chrome. I may download Firefox as a backup (so that if Chrome fails, I do not need to use IE) but thats all it gets for me.Firefox, youre awful!!!!![/citation]
But that is ONE test, not a category composed of several tests, one solitary test. Out of 50+ tests, it tanks one. That is a rather solid record.

 

[adamovera]

[citation][nom]DigitalWarlord[/nom]I would like to see the next "Web Browser Grand Prix" be test on all three OS platforms. It would be nice to see the comparison on a single test.[/citation]
Ooof, that is a timing nightmare, that's 5 browsers on Windows, 4 on OS X, and 3 on Linux, any of which can update at any time. If Chrome, Firefox, or Opera updates mid-testing, that's 3x the re-testing. I had to re-test Chrome AND Firefox each for this one, and still missed FF 10.0.2. Throw in the lag time for the Linux versions to make it into the repos... Probably not going to happen until I can figure out a way to automate large portions of the suite.

 

[srap]

I'd love to see a WBGP with the latest developer version of each browser, using the x64 version if available, just to see where are they evolving. Of course, it should be only a single test and not seasonal like the normal.
Or maybe a Mail Client Grand Comparison, I am certain you would be able to compare them in a very detailed and objective way.

 

[Why_Me]

Been using Chrome since it first came out and been loving it since day one. I have IE for a back up....never really use it but it's on here somewhere. Simple browsing here and my needs aren't anything complicated as far as a browser goes. Chrome seems to keep it simple, clean, and quick.

 

[antilycus]

Ubuntu uses actually Debian stable and changes at a MUCH faster rate than debian does, leaving to a mess of linked files and packages (from what I have seen). While Ubuntu is probably the most advertise, Debian is BY FAR, the most stable, plus it uses the tried and true GUI GNOME.

If Tom's is going to include linux in their benchmarks (which I am vastly appreciative of) can they include debian squeeze (or stable) with either 2.6.32-38 or 3.x linux kernel?

PS: I would love to see a test... Windows 7 against Windows 7 on a KVM w/ full hardware support (though I don't think there is direct access to the GPU on the host side)

 

[thedexmonster]

Definitely agree we need a 32bit/64bit comparison here. Flash, Java, and Silverlight all have 64bit versions out now as well as Waterfox and IE9.

 

[srap]

@thedexmonster: I asked for dev. version tests for a reason: to get the official x64 fox (not a cheap third-party remix), and the x64 Opera 12.

 

[peevee]

A lot of users browse with laptops nowadays. Please test how long it takes to deplete a battery with a gven browser/OS combination. Take the most used laptop, MBP, and test it with MacOS, Win7 and Ubuntu. See how long the battery holds watching youtube, and how many page views on these top 40 sites you can make before the battery is depleted. You know, both graphics layer and network layer (when browsing through WiFi) matter, and they are very different with these browsers.

 

[peevee]

And please, do a test with AdBlock installed. The performance is vastly different due to different extension architectures. And who uses virgin browsers? Not Tom's readers.

 

[mrsphex]

[citation][nom]spotify95[/nom]@016ive: I totally agree!! Chrome is the king of web browsers, in both Windows and ubuntu! Someone must have worked things out wrong, since Firefix won virtually nothing in Win7, yet it takes victory?! Funny stuff... I`m starting to just use my own tests rather than Toms Hardware graphs. Chrome is king.[/citation]
I love chrome so much Sad thing is it got infected with that damn Smart Search Virus crap and I can't get it off and it's a damn lot harder to browse with that thing on -.- So lame

 

[blazorthon]

Firefox uses so much less RAM than the other browsers that it's problems when it runs out aren't too important.

 

[Guest]

The graphics oriented test results aren't very meaningful for Linux. If you had used an Nvidia card with Nvidia's binary driver graphics performance on Linux would probably have been more or less equivalent to Windows.

 

[notsleep]

where's 64-bit opera? version 12 build 1293 is quiet stable for native 64-bit support