Web Browser Grand Prix: The Top Five, Tested And Ranked

[chimney rock]

Another down point of the other browsers is bars not going away when you don't need them, when I got to my bookmarks I want to click on the one I want and it to then go away and not hang around like an old senile duck like firefox and opera do shrinking my viewable area.

Have you actually tried Opera, at all, or are you just making this up? I have the most minimal browser bar imaginable here (10.50 beta) with just the tabs and an address bar, without even a title bar. Everything else is accessible via a right click or pressing the O top left and navigating the menus.

That is when I'm not browsing full screen.

 

[chimney rock]

@mitch074: I totally agree with what you said about Opera's UI. I've always felt that the thought process behind their UI design was "how can we make it as different from the other browsers as possible" instead of "how can we make a usable UI". But I gotta change my mind with 10.50. The new UI actually makes sense and takes advantage of screen real estate. Check out 10.50, it's worth the price But I would have to agree with you for every other version of that browser.

Have to agree - 10.50 is the most amazing browser I have ever used - and I've used scores of them...

 

[1kpc]

[citation][nom]leviramsey[/nom]@tomtompiper: there's an easy way to close that popup: switch from Author Mode to User Mode (there may be a dropdown somewhere in the stock UI; if not View | Style | User Mode will do the trick). The page re-renders without all the original formatting and you can close the popup... when you switch back to Author Mode, the popup is gone.The linked page also does not validate...[/citation]

Why go through all the pain...I want my browser to be intelligent enough to do all these things automatically ..!!

 

[chimney rock]

Always thought Opera 10.5 felt pokey loading web pages compared to Firefox and Chrome. This review confirms it.

Baloney.

10.50 is only recently released - as a beta - so "always thought" is simply trolltripes

 

[chimney rock]

Its like comparing a Mercedes Benz that has grown pretty well with age but the market around it has come up

If Internet Explorer is a car, then surely a Trabant, not a Mercedes.

 

[Tomtompiper]

"The linked page also does not validate..."


It would be an ideal world where everybody stuck to standards, but this is the real world and they don't. It would appear that FF seems to be more tolerant of badly coded pages. As speed is not so important to me I will stick with FF and my add-on's for the present, but I have every intention of trying Opera again when 10.5 becomes available.

 

[Mousemonkey]

If Internet Explorer is a car, then surely a Trabant, not a Mercedes.

That's a bit cruel, Trabant's aren't that bad.

 

[the_man]

[citation][nom]adamovera[/nom]@the_manI have gone back into the test environment and run Acid3 on Chrome and Opera ... Chrome takes around 50ms to clear Test 26 and 2 attempts to clear Test 69. Opera takes about 120ms to clear Test 26.[/citation]

Great that you take the effort to test it properly for Opera and Chrome. It is appreciated. It would be great if you do Safari too. With two fails on timing for Chrome and one fail on timing for Opera, it is probably unfair to declare Chrome the sole winner. What about making Safari, Opera and Chrome share the first place, ignoring the minor timing problems? It would be great if you have the personal strength to update the article.

 

[Guest]

I guess I'll continue to use firefox because it has lower memory usage, loads pages faster, and has more stable addons!

 

[SlipUp]

I guess I'll continue to use firefox because it has lower memory usage, loads pages faster, and has more stable addons!

Yeah, except both Chrome, Opera and Safari are faster than Firefox...

 

[Tomtompiper]

[citation][nom]SlipUp[/nom]Yeah, except both Chrome, Opera and Safari are faster than Firefox...[/citation]

The post you replied to mentioned three things and you replied to only one of them , speed? And you got that one wrong as the article clearly stated that FireFox won the page load test hands down. By the way last time I looked both referred to two things not three.

 

[Sushi Warrior]

Real world tests are ALL that matters, this has WAY too many synthetic tests with questionable results.

 

[stve]

Opera has always been far faster going backwards or forwards on a tabs history.
For a real world test open twenty tabs & then 10 links on each tab
and then back through the tab history to the original page & forwards through the ten links on each tab & just to make it interesting close each browser half way through each test & then restart the browser & finish the test.
Also add 10 or twenty extensions to bring the other browsers up to what you get with Opera as default ( I'm guessing it won't speed them up probably slow them down)
Another good test use a dial up connection Opera would ace that with it's turbo mode.
A tip for anyone trying out Opera for the first time right click on any toolbar then customize appearance & you can have any toolbar on or off & in any position i love the way you can have the panel spring out clicking on the edge of the browser.
That big red button for the menu that saves space : get rid of it right click remove if you need the menu just use the alt key.
& if you like to tinker type opera:config into the address bar.

 

[Guest]

I've been using Safari for the past 8 years (and Firefox as a secondary). As soon as I jumped into Chrome I knew I couldn't go back (just like when you switch from PC to Mac

 

[venjude]

Dear Sir,

I think your test is worthless.

Who cares if Chrome loads Google the fastest? IT SHOULD. GOOGLE MADE IT.

Why should that count in the testing? IT SHOULDN'T.

Also, your test shows that Firefox only uses more memory than the others with just one tab open. Firefox uses less memory than all the others with multiple tabs open which is a more realistic test than all the single page(tab) tests that make up the bulk of your testing.

Also, Firefox's ad blocking programs actually block ads. At least one other browser doesn't actually block them even though it says it does.

ALSO, also, why weren't the different features of the different browsers compared?

Raw testing based on easily stilted things is crap. Just pure crap.

And, seeing as your site features GOOGLE Adsense, you certainly didn't try hard enough to be unbiased.

TRY AGAIN

 

[Guest]

I like almost everything about Chrome: the speed, the lightweight design, the interface, the sandboxing, even the extensions. The only reason I don't use it as my primary browser is the lack of a master password to protect my saved site passwords. That is a HUGE privacy/security oversight on Google's part. I also don't like that it takes more system resources to run than Firefox, but I could live with that if I felt my passwords were secure.

 

[Guest]

Speaking of Browser testing, anyone else finding that the annoying ads in the upper left corner of TomsHardware cover ~1/4 of the text at the top of the page?

 

[Syndil]

I'm pleasantly surprised by the overwhelming majority of positive comments for Opera. I was under the impression that Opera was one of the best-kept secrets on the web, but apparently it has a large and well-deserved following. I've used Opera since version 3.62. I've tried other browsers either out of curiosity or at the suggestion of others, but none hold a candle to Opera. The author needs to give Opera another chance at real-world browsing.

 

[Guest]

You should also ask yourself, "How much did Google pay Tom's to craft the article this way?" Google isn't above paying for results, and Tom's has become more and more biased towards the money over the many years I've been here. Chrome has problems, just like all the rest.

 

[Guest]

Guys, why do you consider memory usage relevant in a SPEED test? These tests distort the end results. Memory usage can be a way to compare browsers, but it is a different issue from speed. Actually you know that you can optimize algorithms different ways, which means you usually consider if speed or memory usage is more important, cause you can't have both!

 

[synth0]

The 1st ever "Best Software Award" for a software product, disregarding the overall end-user Experience and Security of that product is a bit questioning, to say the least.

Even disregarding the lack of Security and User Experience tests, in my opinion, it's not just the wining quantity that should count.

For instance: how can you say Chrome is better than Firefox (for instance), if the latter beat Chrome in Memory Usage, Page Load Times and Flash/Java/Silverlight tests?

Is Startup Time/Javascript/Acid3/Dom/CSS more important that those?

Just to be clear i find the article alright, but the conclusion is lame and that award is just puzzling.

For you to give such an award, you need the product to be flawless, secure, have an awesome user experience and most importantly - to be dominating in each major test.


/puzzled

 

[Guest]

Well, nice test, some thoughts to it from an Opera user's point-of-view:

- the release of Opera 10.5 was hurried because of the ballot screen deadline. It's not good, but I can understand that, at least, they have a good reason (better than with 9.50). That could lead to some strange problems.

- The new JS engine (Carakan) is more RAM consuming. Opera uses dinamic memory management, if you have more RAM, it takes more RAM. I have 3GB-s and Opera consumes about 250-300MB almost constantly. I don't care... On the other hand, Opera is very scalable in this matter, think of the Mobile version which uses (almost) the same core. Opera is indeed very good for older machines.

- V8 benchmark should be disqualified as it heavily biased toward Chrome. It is now, and it was even more biased when it came out (Chrome was about 10 times faster than anybody else).

- It should be noted that SunSpider should be run with C'n'Q disabled(!) as its short tests fakes the results. Or you should go with its newest iteration, the 0.9.1 (see here: http://www2.webkit.org/perf/sunspider-0.9.1/sunspider.html )

- Dromaeo might have some bugs. Opera is superfast there in one complex benchmark, the Regular Expressions. The final 10.5 locks up, but earlier development snapshots worked just fine, and the results were the same. You can see my results of the Regular expression test here (not with Opera 10.5 final, but pre-alpha): http://m.blog.hu/ma/magyaropera/image/carakan/opera105_dromaeo_regular.png , and all the other here: http://m.blog.hu/ma/magyaropera/image/carakan/opera105_dromaeo_detailed.png (big pic!)

- Peacekeeper does not take Complex Graphics benchmark into account because IE is not capable to run it. I find this to be ridiculous. Opera wins that test clearly: http://m.blog.hu/ma/magyaropera/image/carakan/opera105_peacekeeper_detailed.png (for me, Chrome played better in total)

- Chrome 5 is much faster in some cases than Chrome 4

 

[adamovera]

[citation][nom]the_man[/nom]Great that you take the effort to test it properly for Opera and Chrome. It is appreciated. It would be great if you do Safari too. With two fails on timing for Chrome and one fail on timing for Opera, it is probably unfair to declare Chrome the sole winner. What about making Safari, Opera and Chrome share the first place, ignoring the minor timing problems? It would be great if you have the personal strength to update the article.[/citation]
Well, there were 0 fails for any of the three. Chrome finished the test first, Opera second. I did run a few tests on Safari, it's a solid third and in no way a contender like Opera, on this test, or in the conclusion. That's what should change, Opera and Safari don't tie, Opera is second and Safari is third. Also, when a browser takes too long to finish a test but reports 0 fails, it hasn't failed, that just shows it's weak points. Chrome was slightly slow on two tests with it's overall finish time lower than Opera, who is significantly slower on a single test - bringing it's total finish time higher than Chrome.
But remember that I already threw Opera a bone by counting the completely wrong-looking Dromaeo JavaScript test (I really hope that someone more familiar with Dromaeo follows up on that). The Acid3 test doesn't change the relational placement of Chrome to Opera. And throwing out the mem test (the other test in question) does not bring Opera a win, it simply takes away a loss for both browsers (Opera was 5th and Chrome was 4th). I would have to rule in favor of Opera once more for a tie, twice more for a win. Though in retrospect, perhaps Opera deserves the 'recommended buy' award. We give this award to products in two situations: 1) when a product is a good value, but still not a 'best of' in their segment - for example, a budget videocard that has high performance relational to price; and 2) When there is a clear winner but the finish is very, very close. Honestly, I would like to make some new awards for the free software products that I cover. The crown is just going to change hands too often with free software and I don't want to dilute the meaning of the awards for our hardware reviews. But yes, if I were to make any changes, I would give Opera 10.50 the 'recommended buy' award (the 2nd situation) and drop Safari to 3rd in the Acid3 tests (Ch 1st, Op 2nd, FF 4th, IE 5th).
It's not a matter of personal strength at this point (I already had to make 2 changes while the story was live), I'm not sure if I actually can change it now. But thanks for the feedback, it will be utilized for the next one. The full Acid3 results will be relayed and I will re-work the mem test, or probably just create another mem test to look for usage spikes over time to include along with the current one.

 

[adamovera]

@synth0
See the second half of the second paragraph of my response to the_man. I would love to look at the more subjective things like UI and extensions, but I would do that in stand-alone reviews for each browser individually. How does one compare those features across the browsers? Better yet, how do you weight subjective findings with the raw quantifiable data of benchmarks. Believe it or not, if I said a single word about how I like one UI over another, I'd have 20x the negative comments. I decided to go old school TH on this one and let the benchmarks write the article. But I see your point on the award and I'll make moves to differentiate from our regular hardware reviews (who typically keep the award much, much longer than constantly-changing free software).

 

[adamovera]

@cousin333
Thank you for the links. I don't think I want to do anything about Peacekeeper (sounds more like an IE problem), but I'll use the newer SunSpider next time (I used 0.9 for this article). I may want to leave the Dromaeo JavaScript Test out next time, unless they fix it by then. I don't know why Opera 10.50 devastates in that single test, but it skews the final result BIG TIME and considering that it didn't finish another portion of the test at all, you'd think that would balance it a little, but no. We'll just have to wait and see on that one.