Home > Software Reviews > Browser Comparison 2010

Browser Comparison 2010

UPDATE: Since there were many browser release such as Safari 5 and Chrome 8, this post is somewhat outdated. I will update it as soon as I return from my overseas trip. *sorry for the delay*

Hi guys. Since the last update that I gave you, I have been quite busy as hell. Got a submission this week, as well as mid-term exam on the same day. Best bit? It’s all about C++ programming which I have left several years ago. I gotta admit that I’m probably being too spoiled using object-oriented modern languages such as Java that has its own garbage collector, but what do you know, after all these years I finally understands the beauty of C++.

Anyways, enough of the (mid?) weekly update. I was planning to give these results on March but weren’t able to do so. Nowadays there are so many browsers out there, and by browsers I meant web browsers. Some of them you might not know of, such as Konqueror (Kubuntu users probably will smack me), or Midori. Some, you might have it installed in your computer. But here is the thing: of those widely known browsers, which one is the best?

Well, to tell you the truth, it really comes down to your own personal experience (NOT personal preference, though, which I will explain a bit later on). Some browsers might have their own strengths, such as Firefox with its plugins, Chrome with its simplicity, Opera with its damn good user experience, and so on. But the thing is, they all have to be able to do the most important thing: delivering the web content to the user.

JavaScript Execution Time

Let’s face it, the world’s most visited website is facebook (pun SEVERELY intended). Whether you realize it or not, most social interaction sites such as facebook, twitter, linkedIn uses AJAX for processing request-response interaction. To put it in a simple way, everytime you clicked on that ‘Share’ button on facebook, or tweeting, you use JavaScript to send the text you’re about to share, and the other people will get the text using JavaScript as well. It all happens in the background.

The test, measuring the JavaScript execution time using the SunSpider JavaScript benchmarking test, is intended to measure the time it takes to fully executes the most used JavaScript components. The test itself has an error latency of less than 10%, which in my opinion its more-than-average accurate. Faster JavaScript execution means faster user response time, meaning your status will be updated straightaway the moment you click the ‘Share’ button.

JavaScript Execution Time

DOM Selection Speed

No, I’m not talking about the DOM in the Gundam universe, but rather, the Document Object Model, a representation of objects used in the browser language (HTML, XHTML and its derivatives). Most of the web 2.0 applications such as the online Photoshop, or even this WordPress dashboard uses DOM as the object model. As such, DOM is in a way a standard in web development. The test measures the DOM selection speed by using jQuery using the SlickSpeed Selector test. It is worth noting that using Dojo instead of jQuery to simplify the interaction between AJAX and HTML pages is faster, however it only runs perfectly in Google Chrome. The faster the selection speed is, the more responsive the web application would be.

DOM Selection Speed

CSS Rendering Speed

As with AJAX, most of us would already seen this before. The reason why facebook have a similar theme in every single language, the reason why we can change themes in twitter, and the reason why we even have sidebar in My Sanctuary (it’s on your right, really), is because of CSS. Cascading Style Sheets define the graphical elements to be displayed such as your fonts, your font color, the background image, and even, the position of those things. CSS has gone way way beyond the first time I learned CSS. It’s becoming too deep, therefore I had to stop and focus myself to learning what I did best. Anyways, the CSS rendering speed defines the response time of a web page. The tests are taken using the Nontroppo CSS test and measures the page load times of a heavily CSS-ed webpage.

CSS Rendering Speed

Page Load Time

In conjunction of the previous test, the next test measure the page load time on a website. Of course everyone would want the fastest browser that can render a webpage instantly without even thinking or waiting for it. However, given the location, broadband speed, network topography and site traffic, the results may vary one another. Therefore it’s probably best if these considerations are taken into account. The tests are measured using the Numion Stopwatch for loading the Yahoo! main page, using a wireless broadband network in Melbourne.

Page Load Time

Acid3 Test

This is where all browser is put into real test. Remember the primary goal of a web browser? Displaying a web page to the user correctly and according to the web standards. The reason web standards are there is because the worldwide web, like information and communications technology, is growing exponentially everyday. New inventions are made and old ones might or might not be discarded. Without a certain standards, this would be a chaos.

A bit of history lesson, the worldwide web emerged with the first generation browsers. At that time there were no standards. People at that time wouldn’t even imagine that the world would be just like the world we’re living in now. As such, they did not bother to create a standard. Guess what, the worldwide web revolutionized the whole world (if I could, I would surely vote it as the most revolutionizing invention ever), and at the same time, Microsoft dominates the PC market, thus the dominance of Internet Explorer.

Acid3 TestAs you can see from the test result, Internet Explorer fails the Acid3 test. IE is known by us web developers as infamously notorious for breaking or failing to compel with web standards. The reason is because it was the dominant browser in every single PC in the world (save the Linux and Mac users), so that rather than abiding to the newly established web standards, it would rather create its own standard. That is until Firefox kicked in, then IE starts to step-by-step complying to the web standards.

Most of the browsers nowadays passed the Acid2 test, but not Acid3. Some claimed that the test measures rarely used components of DOM and JavaScript, but hey, if you ask me, that component must be there for a pretty damn good reason.

Some Additional Info

Of course the test is not considered valid if it is tested on different machine. That’s why I tested all of it under my Throne Drei. Below is the info for the machine and the versions of the browsers that I used. Keep in mind that all browsers are extensions-free. That means no plugins and extensions installed in every browser.

Additional InfoThe Verdict

It is interesting to note several things that I encountered when I did the test. First of all, under my machine, Internet Explorer DID actually starts up really fast. Like blazingly fast. In fact it is even faster than Chrome. I suppose it has something to do with my 64-bit operating system since it utilizes RAM better than 32-bit. And although we might say that IE is not web standard compliant, when it comes to using a standard static (although in theory, a fully static website is almost extinct) HTML pages, IE renders it perfectly. This might be on the fact that most web developers uses IE for testing purposes (it is still by now the world’s most heavily used web browser in the browser market).

Another interesting fact is that, Chrome 3 actually performs BETTER in these tests (as seen by previous tests in other websites). I guess the inclusion of extensions in Chrome 4 really put things to the ground after some hype that the Chrome caused in recent years.

Safari, with a decent results in these tests could be the browser you might want to choose. But note just this single thing. As with IE, Opera, and Chrome, putting up search string in the URL bar takes you straightaway to the search results of your default search engine. In Firefox, you would be taken to the related website, or if there’s no related website, you’ll be taken to the search results as well. In Safari, however, apparently their developers are just too lazy to put a redirection page. You would see an error page which prompts you to click to redirect to the search results. Disappointing, I must say.

Opera, on the other hand, performs very VERY well in these tests. Opera 10.5 really did a good job and is a big improvement over Opera 10. However, there is one slight thing that I dislike. It sometimes, just sometimes, renders a webpage incorrectly. I know it’s not on every website, but it’s just not right for a browser that emphasizes on user interface.

How about Firefox? The used-to-be IE competitor and rival seems to be pretty good in these tests. I couldn’t agree more that Firefox is INDEED better than IE overall, but again, as a web developer, I couldn’t help but feel that Firefox is way too BLOATED these days. I used to trust Firefox and use it for my main browser, but once, just once it betrayed me. For all those web developers, mind these, Firefox do NOT support synchronous AJAX. Added with the slow startup time, those are enough reason for me to leave Firefox be. I’m just hoping Firefox did not make the same mistake as IE did, creating its own standards.

So what’s my verdict then? As I’ve told you beforehand, it really comes to your own personal experience rather than personal preference. For me, I prefer the Safari sleek Apple-ish design, but my experience told me otherwise. If you want a good user interface, go for Opera. If you want plugins and customizable browser, go for Firefox. But if you’re a web enthusiast, never ever ever EVER use IE as your main web browser. As for me, the simplicity and the omnibar of Chrome is more than enough for my daily usage. So, what’s your choice?

jovee~

  1. April 23, 2010 at 1:01 AM

    Lynx can be useful when you broke your system and can only access the terminal/command line and no gui.

    • April 23, 2010 at 1:11 AM

      true. I’m considering of trying out lynx, but I read it has some issues concerning SSL. but who the heck will go over to ebay and buy something with lynx. ;)

  1. May 23, 2010 at 10:04 PM
  2. July 7, 2010 at 7:58 PM
  3. July 29, 2010 at 8:53 PM
  4. August 3, 2010 at 4:14 AM
  5. January 7, 2011 at 4:11 PM

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: