Acid3 test results for IE8 and its competitors

Posted 14 April 2009 by Russ in Other stuff. Tags: browsers, Web Standards.

To quote The Web Standards Project, Acid3 is the third in a series of test pages written to help browser vendors ensure proper support for web standards in their products. Acid3 is primarily testing specifications for “Web 2.0″ dynamic Web applications. Also there
are some visual rendering tests, including webfonts.

The term ‘Web 2.0′ has been around for a few years, but I think it’s fair to say that browser vendors have been struggling with the gauntlet of standards compliance thrown down by the web development industry. So Acid3 – launched in October 2008 – is going to be a bar that in fairness, is going to be too high for most current browsers to pass. With IE8 now launched however, it’s not a bad time to stop and see how it fairs against its competitors in the latest Acid3 test.

The Acid3 test reference

The Acid3 test performs a number tests using JavaScript to try and provide some kind of indication of the level of support for a whole host of functions, including support for CSS2.1 and CSS3 specifications, DOM2, data handling and SVG support and you can find a full list of test details here.

I’m not going into detail about the test results here, I’m merely showing how the automated tests for each browser stack up. I’m also not testing on a multitude of platforms – just Mac OSX (Leopard 10.5.6) and Windows XP. As Lars Gunther explains, simply achieving 100 out of 100 is not a pass as such as there are other, less absolute measures of success, so treat this as an early indication of how the big vendors are fairing.

Ready then? OK let’s go. For a browser to pass the Acid3 test, they should render the test exactly as per the reference image below* – pixel perfect – with the browser’s default settings. The rendering progress should also be smooth, which perhaps is a little subjective but the X/100 is the basic benchmark we’ll be looking at for these tests.

* This is scaled down slightly so the test fits the width of this page’s content column but the results are still accurate.

So why do we care?

I’ve worked with younger developers who haven’t had much exposure to legacy browsers and stories of working with Netscape 4.01, 4.7 and IE 4 and 5 pretty much fall on deaf ears for those guys. But for those of us who have worked with these early browsers, standards can only be a good thing.

With the advent of Firefox, Safari and Chrome and the continuing efforts of Opera, Microsoft has had no choice but to sit up and take notice of the calls to improve their browser product and although the tests above appear poor, the level of standards built into IE8 is far better than IE6. Will IE ever be at the bleeding edge of support for technology like CSS3 and HTML5? Probably not, but with the majority of market share, most of us need to develop our sites to work with the lower standards compatibility of IE, so seeing any improvements to these scores can only be a good thing.

And who knows, in a rapidly changing environment (Apple recently announcing that Safari has achieved a full pass that extends beyond the automated test), maybe we’ll start to see more pressure applied on Microsoft from businesses striving to deliver the best sites without having to conditionally code for browsers that aren’t up to the job.