Chrome 27, Firefox 22, IE10, And Opera Next, Benchmarked

[snadge]

flawed testing!! - ive used Chrome on many PC's and laptops and can definitley say it is 4-5 times quicker than firefox at startup... even when loading from HDD it is almost instant!! no way is it 5.8 seconds on a cold start lol....that should have raised eyebrows at TH and they should have re-checked their testing setup...for me it starts cold in about 1 second and does so on many machines I use... based on that I refuse to accept these test results

 

[apple cyanide]

maybe GOOGLE paid off UBUNTU to run chrome ? GOOGLE only allow their staff to run OSX or GUBUNTU after all

 

[belardo]

Isn't Firefox 25 really FireFox 5.0? okay, this is what it looks like: If IE10 and Chrome had a baby, then you have ff5.0/25/wtf..... AND IT IS UGLY.

Imagine... if only you could keep ff from autupdating.

 

[belardo]

Is it me, or does Opera and other browsers look like crap under Win8? All flat...?

Hope Opera next retains the great functions of Opera 10~12... it still kicks Chrome's butt in usability. ff... still half way useless. Operas worst problem is that print-outs are mostly useless.

 

[RedJaron]

Firefox works perfectly for me. It is a big of a memory pig using about 1/2 GB right now but not so bad considering I have 65 tabs open.

It hasn't crashed in a long time and also the new releases fixed the serious memory leak.

But the download manager is a bit annoying but oh well.

IMO, Chrome sucks + I don't trust google products.

I rarely had it crash, I just hated the memory leak, or something similar to it. Open a few tabs, no problem. Open a lot and suddenly it's using 1.5GB of memory. Close off the tabs, FF is still using 1.5GB of memory. It'd never release the memory until the whole browser was closed. Chrome and IE both give back most of what they take as you close tabs.

And no, I'm not a big fan of Google products either as they have no idea how to make a proper interface and always want to think their products are smarter than you. But Chrome isn't too bad, and I don't bother signing in with any Google account.

 

[ronch79]

I just downloaded and tried the new Opera browser (Webkit), and I must say it doesn't really feel like Opera at all. In fact, it's oozing with Chrome-ness. It's obvious Opera wants to make its new Chrome clone feel like the old Opera but it just doesn't. Heck, it doesn't even have the familiar items in the options section ("Preferences"), Speed Dial does not feel like the old Opera's speed dial, and I can't even find the smooth scroll feature, which makes scrolling choppy. In the old Opera Smooth Scrolling is enabled by default.

I don't really run into a lot of website incompatibilities with the old Opera, and if this new Opera trades in the old Opera's feel for near-100% website compatibility, well, I'm not sure I won't uninstall this new Opera and go back to the old Presto-based Opera.

 

[Taintedskittles]

Friends who ask me why I don't use chrome. Will get a forward to this article in a heart beat now lol. Been using Firefox since forever. I like the interface, add-on's (persona), & now bragging rights to speed.

 

[nevilence]

IE is S**t, no surprise there =P

 

[tweedledee]

I rarely had it crash, I just hated the memory leak, or something similar to it. Open a few tabs, no problem. Open a lot and suddenly it's using 1.5GB of memory. Close off the tabs, FF is still using 1.5GB of memory. It'd never release the memory until the whole browser was closed.

It's just not true, you should just give FF some time to collect garbage. After minute or two, Firefox will release most of its used memory. I've just closed all my tabs (around 50 or something), and memory usage gradually went down from 1.5Gb to less than 500Mb in two minutes. It's still more than memory consumed on start-up, but the memory was not leaked -- it's just can't be released due to heap fragmentation. It will be reused when you'll open new tabs again (leaked memory on the other hand means memory that can't be released and can't be reused). Giving that 500Mb is not much for any remotely modern PC, and the fact that FF consumes massively less memory than Chrome, I just don't see a problem.

 

[ElDani]

Giving that 500Mb is not much for any remotely modern PC, and the fact that FF consumes massively less memory than Chrome, I just don't see a problem.

While not everyone is able to use or own a computer with 8GB RAM or more - which is about what's needed for some serious multi-tasking - I can absolutely agree with your point.

The maximum memory efficiency after a number of tabs were closed, doesn't play as large a role in the browser or system performance as people would expect. At least that's the way I see it. It's perfectly all right to slam a browser for memory leaks, but Firefox got rid of most of them in the post 3.5 release stream. I haven't experienced a memory leak in months now and even that was caused by an extension and not the browser itself.

If Firefox' only criticism should be, that the browser doesn't recapture the minimal use of memory it had after a fresh start with only 1 tab, then I don't care! To me it's much more critical, that Chrome takes double the memory for the same amount of tabs while they're being used. Here's an example statistic from me with both figures taken from Chrome's accurate about:memory page:

Firefox 22: 5 tabs active, 87 unloaded = 357MB RAM
Chrome 28: 4 tabs active = 774MB RAM

The browsers are directly comparable in their use of extensions, as Adblock Plus, Ghostery, Stylish and Grease/Tampermonkey are loaded in both browsers, Firefox even runs half a dozen more.

I'd much rather restart Firefox manually every now and then - which takes me less than 15 seconds, as tabs remain unloaded when the session is restarted - because Chrome's only salvation on a lower class computer would be to add more RAM to the computer.

 

[bedex78]

Why isn't Safari in the test??

 

[Onus]

Iirc, Safari performed poorly enough in prior Browser Grand Prix that it was probably weeded out in a "qualifying round."

 

[RedJaron]

It's just not true, you should just give FF some time to collect garbage. After minute or two, Firefox will release most of its used memory. I've just closed all my tabs (around 50 or something), and memory usage gradually went down from 1.5Gb to less than 500Mb in two minutes. It's still more than memory consumed on start-up, but the memory was not leaked -- it's just can't be released due to heap fragmentation. It will be reused when you'll open new tabs again (leaked memory on the other hand means memory that can't be released and can't be reused). Giving that 500Mb is not much for any remotely modern PC, and the fact that FF consumes massively less memory than Chrome, I just don't see a problem.

Yes, yes it is true, though thank you for calling me a liar. And yes, I do know the difference between a true memory leak and simple fragmentation. This was typical behavior for FF 3.6 in 2010. I worked a year as web page QA and FF was the company's officially supported browser. Memory usage was constantly at 1 GB or more, and no matter of waiting for garbage collection helped. The only thing for it was to close the whole browser.

And I disagree that 500MB isn't much for a modern PC. Considering that's on top of everything else that's running, a simple office PC with only 4GB RAM ( though 2GB of RAM wouldn't surprise me on some laptops, ) that can quickly start approaching memory limits. Especially if FF starts hoarding memory again.

 

[smeezekitty]

The memory leak was fixed around firefox 15 or so.

 

[Someone Somewhere]

Why isn't Safari in the test??

Iirc, Safari performed poorly enough in prior Browser Grand Prix that it was probably weeded out in a "qualifying round."

Safari for Windows is no longer available. Or you could ask why they aren't running a linux system, and running Konqeror/ReKonq.

 

[grumpigeek]

No 64-bit Opera 15.

Goodbye Opera.

 

[stupidlogin]

One thing sorely missing from this as from so many other browser comparisons is a memory benchmark. I routinely have a TON of open tabs, so whether the browser cobbles up some 90% of my available RAM or not is a lot more relevant for overall performance than a couple of percent points in speed difference. Add to that the time it might take to swap in a page when switching tabs and even the fastest rendering browser comes down to a crawl if it can't keep a tight lid on its memory consumption...

 

[Someone Somewhere]

I suggest looking at page 9: Memory Efficiency.

 

[stupidlogin]

I suggest looking at page 9: Memory Efficiency.

Okay, how did I miss that...!? When I read the article on my tablet yesterday, I could've sworn it ended with Hardware Acceleration. I blame bad connectivity...

 

[xizel]

Só id like too see these testes in multiple setups. For example I have a amd e-450 and chrome really struggles. Also noticed that since I sold my gpu and using the on board chip in my 2500k chrome really struggles with flash playback. With and without hardware acceleration.

 

[omendata]

Could we possibly see Maxthon added to the list next time - its the best browser by far next to Firefox and Chrome.

 

[mikeynavy1976]

Well, I posted earlier, and after this article decided to give Firefox another try. I tried 22, Aurora (24a), Nightly (25a), and Nightly UX (25a). None was as responsive, quick to load pages, or ran as smoothly as Chrome (I use the dev channel and my wife uses stable or beta). My results about application loading was also the opposite of the results of this study. Firefox took much longer to open (cold or hot) than Chrome. While IE opens instantly, Chrome isn't far behind, and Firefox took several seconds. While Chrome doesn't have the customization that Firefox has, speed and responsiveness are the most important things for me. I'm guessing that the test setup makes a big difference. I have a 3-year old Lenovo W510 with a 1.7GHz quad core i7 and 8 GB RAM.

 

[_zxzxzx_]

I've always used Firefox and this proves I made the right choice in sticking to it.

 

[Someone Somewhere]

One thing I'd like to see is a comparison between them on SSDs and HDDs - I'd imagine there's some difference in how the internal file structure is laid out, so some might use more random I/O vs sequential.

 

[spotify95]

Chrome didn't win the performance tests because of it's unusually long start up times in Windows 8. You've even commented on previous WBGP's, that Chrome was much slower on Windows 8 than Windows 7. However, Chrome has never experienced this before on any other OS, so either Google need to do something about Windows 8 performance, or there may be a bug in Windows 8. (I can't say because I have Windows 7, I've just read the previous articles). Additionally, Chrome has always been rather poor at page load reliability, thus explaining that one.
However, something that I must add: despite Chrome winning by a long shot in the "page loads", it still tanks because of the significance of the 40-tab load time (which, as discussed, Chrome has bad loading times under Windows 8 as a whole). Maybe the significance between the two tests could be balanced out a little, for the next installment?
http://www.tomshardware.com/reviews/web-browser-chrome-25-firefox-19,3459-4.html
In the last article, Chrome was over a second and a half slower in Windows 8 than Windows 7. So until Google sort out their Windows 8 performance, Firefox will ALWAYS win the WBGP articles.
(Note: I am not biased towards either browser, I use Chrome as primary and Firefox as secondary; I just wanted to make a point about Google's lousy W8 performance. Hopefully Google will sort it out)