Web Browser Grand Prix: Chrome 20, Opera 12, Firefox 13

I didn't thoroughly read the article, so I'll just get that out of the way first hand.

Chrome probably is the best performing browser these days, no arguments with that. I never really did like the overall feel of the browser (nothing against it, just preference), so I never really started using it.

I used Firefox for years. I was a diehard. When Chrome came out, I stuck with it because I felt I was loyal. However, it eventually got to the point that the memory leaks and other issues pissed me off so much I set out to try something else. I've been running Opera for about a month now, and I couldn't be happier. I feel it is a very similar experience to Firefox, however without any of the issues that I experienced. In my humble opinion, Opera and Firefox should be swapped on those standings.

I'll be sure to take a bit more of an in-depth look once I get some downtime. Just thought I'd give my two cents.
 
It's remarkable how they made something quantitative and turned it subjective, but still used numbers to describe the results. Their numeric scoring system is completely flawed. On top of it, the title of your thread implies those numbers are the scores. Really, they're the version numbers that were tested.
Jesse B said:
Chrome probably is the best performing browser these days, no arguments with that.
Click to expand...
This assertion is absurd -- and I mean the "no arguments" part, not the "Chrome" part. Taken as a whole, you disagree with yourself: if there's no arguments, why is Chrome only "probably" the best-performing browser, and not unequivocally the best?

The arguments around a claim of "best performing" start with deciding what performance really means.
 
Last edited:
Yeah it doesn't really matter, they're all pretty similar. I too have been a have recently left Firefox for Chrome, mainly because of the recent Flash update which crashed the browser shortly after starting. They really need to make some deep changes to Firefox, there's little isolation between add-ons and tabs, also a massive memory leak (it isn't isn't rare for it to be using 1, even 2 gigs of memory).

Chrome doesn't suffer from these issues and it has many of the same add-ons as Firefox has, so I'm quite happy with it.

Personally I believe Safari is the fastest browser. It's also very smooth when scrolling or refreshing (it hides the rendering very well). But then it doesn't come with the extensions I have become accustomed to.
 
Last edited:
Wiseguy2001 said:
Personally I believe Safari is the fastest browser. It's also very smooth when scrolling or refreshing (it hides the rendering very well). But then it doesn't come with the extensions I have become accustomed to.
Click to expand...

And safari came dead last in the article...:)
 
At work at a web design firm, and all the other employees use Chrome, but I still use Firefox.

Chrome is faster for sure, but I like Firefox because of certain extensions that I use.
 
mikeblas said:
IThis assertion is absurd -- and I mean the "no arguments" part, not the "Chrome" part. Taken as a whole, you disagree with yourself: if there's no arguments, why is Chrome only "probably" the best-performing browser, and not unequivocally the best?
Click to expand...

I said probably as I haven't done any in-depth performance testing myself that would indicate so, but from casual use would believe that Chrome had the best support for modern HTML5/CSS3 features, and seemed to load pages faster. I'll try to word my response better next time, didn't put a whole lot of time into that post.
 
mikeblas said:
It's remarkable how they made something quantitative and turned it subjective, but still used numbers to describe the results. Their numeric scoring system is completely flawed. On top of it, the title of your thread implies those numbers are the scores. Really, they're the version numbers that were tested.
Click to expand...
Due to version inflation (started by Chrome), we have these ridiculously high #s, such as 20 for the youngest browser in Chrome. :eek:

Basically, the article tested the latest and greatest versions of the 5 most popular Windows 7 desktop/laptop browsers. I'd like to see them carry this out for other platforms, such as Linux, Android (where I currently use Dolphin, but may switch back to Firefox), iOS, etc.

I know some Firefox fans are steadfast that FF 3.6 was the last good browser Mozilla made, but I've been happy with the progression. Granted, the footprint has gotten larger, but the user experience is good. It does require relatively modern hardware. FF will bring a single processor CPU to a crawl if loaded with Flash-heavy tabs, for example.

The only browsers I use are IE and FF and I much prefer FF. For certain work-related apps, I have to use IE and just deal with it.
 
I like these comparisons that Toms does. Yeah its not earth shattering or anything but I think they make for cool reads and its interesting to see how these browsers make changes here and there.
 
I am still using FF 9 as ea. time I upgrade, there are far more problems than it's worth. My current problem w/ FF9 vs. say FF 4 or 5 is that the book mark file, "places.??" gained so much in size, that it's not possible. And compares this file vs. bookmarks.html, it is much bigger.

So say my file size on Places.?? is 15 MB, without changing much or adding say just a few new bookmark, eventually, that file could be 20MB. My old bookmark.html is only a few hundred K, and the no. of bookmarks are about the same
 
20 megs? Why is that unacceptable? Are you worried about running out of floppies?
 
Didn't read the article, don't find Tom's Hardware to be a reputable resource anymore.

Other than that folks tend to use what they like and what they feel has the best features and support they need. I use FireFox for reasons that I cannot state or I will get banned from H. Also for Firebug, FireShot, and MeasureIt. I don't care that it currently uses up over 800MB of ram with the tabs I have open. That's what I bought the ram for, for it to be used.

I don't use Chrome because every time I click on a link it sits there for 5-10 seconds saying "Resolving Proxy" even though I am not behind anything that would justify that. I can't be assed to look into how to fix that. If a browser doesn't work right out of the box I lose interest.

I don't use IE because it doesn't have the one extension I value most in FF.

I don't use fringe browsers because they are just that.
 
mikeblas said:
20 megs? Why is that unacceptable? Are you worried about running out of floppies?
Click to expand...

I don't like that file expand the way it did. Even when I first transfer that file from bookmark.html to place.?? , it's only 8.5Mb, with adding about 100 bookmark, that file ballon to 20MB. I sens something isn't right

Anyhoo, the only small problem I just have w/ FF 9, is when I click the "Check" button under yahoo mail, it's not working anymore, I wouldn't mind upgrade to ver. 14, but in that thread, a lot of people is complaining about it. I hate to fix 1 thing and create another problem
 
Happy Hopping said:
I don't like that file expand the way it did. Even when I first transfer that file from bookmark.html to place.?? , it's only 8.5Mb, with adding about 100 bookmark, that file ballon to 20MB. I sens something isn't right
Click to expand...
You're not making any sense.
 
Love these browser discussions. There are some pretty religious defenders out there for their particular browser. Including myself probably. I've given all the browsers a try, and always wind up back at firefox for whatever reason.

Chrome is great, but the task manager process spam drives me away every time. Firefox definitely performs worse, and I restart it every few days, but it has the best interface IMO and the best extensions. I heard the firefox devs may be switching to a multi-process model as well sometime in the future... :rolleyes: If that happens, I will again re-evaluate my browser options. Perhaps Opera?
 
A few things to note about Opera:

12.00 was tested.

With 12.02 that was just released, for the 32-bit version, Opera reverted from out-of-process plug-in handling back to in-process plug-in handling due to lots of users experiencing performance and stability problems. The 64-bit version still uses oopp so it can load 32-bit NPAPI plug-ins and because it doesn't suffer from the problems as much as the 32-bit version did.

12.50 (not finalized yet but builds are already at http://my.opera.com/desktopteam/blog/), already has a shit-ton of core updates. Among other things, it adds support for the HTML5 fullscreen API (although the spec has been updated again so Opera will have to update its implemnentation) and ICC color profiles. Opera's websockets implementation was also updated to match the latest rev of the spec and support is now turned on by default. Opera 12.50 builds have new rendering delay defaults (that are tweakable via opera:config) that make perceived loading faster.

Due to problems with a lot of GPU drivers (mostly intel), performance not being good as other browsers' HWA and no way to use GDI (instead of DirectWrite) for fonts when using DirectX mode, Opera's hardware acceleration is turned off by default, along with webgl (except in newer builds). But, you can turn them on in opera:config (harware acceleration works great for many).

Enable WebGL (0 = off, 1 = on if driver supported, 2 = force on)
Enable Hardware Acceleration (0 = off, 1 = on if driver supported, 2 = force on)
Preferred Renderer (0 = OpenGL, 1 = DirectX

(Opera's default software renderer is already pretty damn fast for a software one and blows other browsers' away.)

Also, Opera 12.50 now supports windows (vista and newer I believe) Direct3D Warp for getting Direct3D 10-11 out of DirectX 9-capbable GPUs via the CPU. It not going to be as good as full HWA, but it *might* be faster than Opera's software renderer, which might be a good fallback if you GPU driver sucks or if your GPU is a little older.

See opera:gpu in Opera for info.

In short, stable versions of Opera are always quite a ways behind (the latest snapshots are way better and more on par with other browsers). And, Opera is careful about what they enable by default (lots of things can be tweaked for better performance/standards support).

I really love Opera's UI. It's so responsive. I can get things done way faster in it.
 
I'm looking at their analysis table, and it really doesn't make sense to me how they decided some traits are just so much more important than others. For example, how do they figure that 'Page Load Time' is more important than 'HTML5 Hardware Acceleration'? How do they figure that 'Standards Conformance' is more important than 'Memory Efficiency'?

Certainly they all seem somewhat important, and I really don't understand how they can decide one is just so more important than the other. I mean, sure, a lot of web pages use Javascript, but that's not really at stake here. What their results are saying is that having a somewhat faster Javascript engine (even though all of the browsers have seemingly sufficient Javascript performance for modern web use) is more important in a browser than something immediately more tangible, like having somewhat faster startup time or somewhat better memory efficiency. Certainly the tests could receive a little bit of weighting, but to say one thing is twice as important as another (by awarding twice as many points in each standing) seems a bit designed and arbitrary.

And what's with this whole 'categorizing' of performance? Why not just keep the numbers from the tests and weight them appropriately? Why put it up to someone to decide which of 4 discrete categories to put each browser in, when you could simply leave it up to the data to put them exactly where they belong? If browser A gets a somewhat higher score than browser B, but both of them were outscored by browsers C and D, and if browser A and browser B both end up in the same category, then browser A and browser B both get the same number of points even if browser A did better in the test. It's like they're purposefully making their results less precise for reasons outside of making a good test.

I honestly feel like they could come up with any results they want with that same data just by adjusting the analysis table to their liking.
 
FF 15.0 (just released this week) addresses the memory issues that some people were complaining about. :cool:
 
Here's the update: Web Browser Grand Prix: Firefox 15, Safari 6, OS X Mountain Lion

TomsHardware.com said:
Today we're breaking out the Hackintosh for our first-ever Web Browser Grand Prix on Apple OS X 10.8 (Mountain Lion). How will Chrome 21, Firefox 15, Opera 12.02, and Safari 6 stack up against each other, and to IE9 and the rest of the Windows 7 browsers?
Click to expand...

Read the link to see who wins on both Win and Mac platforms.
 
That means I care about the conversation, not about the review. Hint: nobody else responded.
 
Someone needs a hug.

I agree that the review is kind of worthless. However, it's always great to show how bad Safari is on Windows. ;) (I do like Safari on OS X)
 
The other thing that was pretty obvious was that the Windows performance was significantly better than Mac OS performance for those browsers that are on both platforms.
 
You must log in or register to reply here.
Back
Top