Acid3 Test Unleashed, Murders Every Current Browser

Just a few months after the announcement that Internet Explorer 8 successfully passed the Acid2 standards compliance test, the Web Standards Project (WaSP) announced last Monday that it unleashed Acid2's successor, Acid3. Created to identify flaws in the way a browser renders its web pages, WaSP's Acid tests throw down the gauntlet with difficult-to-display graphics written to accentuate browsers' quirks. When the original Acid test was released in 1998, it helped reign in browser inconsistencies and insured that Internet Explorer, Netscape, and others handled HTML code according to specification – making web designers' lives easier and ensuring the web rendered consistently in the future.

Currently, no known browser is able to correctly render the Acid3 test, which displays an animated, incrementing score counter and a series of colored boxes with some description text. Bloggers have already assembled galleries of browsers' failing test results, with most of today's browsers scoring between 40 and 60 on the test's 100-point scale. The results shouldn't be too alarming as the Acid tests have always been forward-looking in nature, and are designed to measure standards to aspire to, as opposed to what's current. Also note that more than six months lapsed between Acid2's release and Safari 2.02's announcement that it was the first to pass Acid2.

News Source: DailyTech

Report a problem with article
Previous Story

AMD Reveals Upcoming Plans

Next Story

Computer viruses as they look in 3D


Commenting is disabled on this article.

Oh darn Maxthon only scores an 11/100 but then do i realy care NO i dont i'd rather use maxthon than fire fox anyday

Can't even read it in IE 7. Firefox is 50/100. Safari for Windows is 38/100

Opera is the highest (Opera 9.5 latest build I have is something like 9815)


I'd check the Firefox preivew except I've got too many plugins that rely on Firefox 2.

I guess the CSS isn't compliant according to the W3C validator, but that's because the validator doesn't support CSS3 and the test is also poking at browsers handling of CSS bugs. The HTML is all nice and compliant for all of the acid pages.

The latest Opera beta for Mac OS X scores a 65/100. We'll see what the next build brings.
The latest Webkit build for OS X scores an 88/100. Weird.

guys, help me
acid3 crashes my internet explorer 7
im running windows xp x64 and the 32-bit IE crashes but the 64bit passes
does anyone know whats going on?

Interesting bit of trivia: Webkit nightly from 2-3 nights ago scores an 86/100. That's 4 extra points in a few days, so they're well on their way to passing the test.

So, how do they know it works if there is no browser that can render it correctly?

Do they have some secret browser that "works"?

(tmaxxtigger said @ #13)
So, how do they know it works if there is no browser that can render it correctly?

Do they have some secret browser that "works"?

I assume they simply code it to adhere to exact standards, though as you point out there is no way to ensure there isn't a bug in their code. I would hope that they have gone over it with a fine tooth comb before publicizing it though - that would be quite an embarrassment if they found a bug in the test because they had no browser that could properly test it in advance!

The tests have been public but unreleased for months and the browser developers (at least Mozilla and Apple) have been able to point out problems with them. A few issues have been found and fixed as a consequence.

(Orlando Rays said @ #11)
Any idea if the Mac versions of Firefox, Safari, Camino and Opera score differently?

In theory they should score the same since they use the same rendering engine on all platforms, but that's obviously not confirmed or tested.

(johnorien said @ #9)
Firefox 3 Beta 4 (rc) scores 66-67
weird, different scores depending on reload

Yes I also got different scores on reload while using a WebKit nightly... I wonder if it behaves different if some of the data is cached. But then I would think a reload *should* re-load all of the data.

CSS3 is not even a W3C Recommendation. Why test it at all?
Why include CSS3 tests in Acid3. What if the final specification is different?

Latest versions:
FireFox: 59
Opera: 60
IE: 17
Safari: 90

Webkit wins, for test I used the lasted Beta versions (nighty builds)

(Orlando Rays said @ #6.1)
Funny, Safari 3.04b only gets a 39 for me.

Indeed, he said he was using nightlies - as am I. Using the latest WebKit nightly, which Safari 3.1 is supposed to include when it is released in the coming months, I also received a 90/100. As someone else (below I think) pointed out refreshing gets different numbers.... odd.

To try this out go to

As was mentioned somewhere else here, the WebKit nightly gets positions almost exactly like the reference (very slight spacing differences). The colors of the smaller boxes and one in the middle are grey though and not the bright colors they are supposed to be.

Either way, it looks like Safari might end up being the first to pass acid again. I would guess that if it's already scoring 90/100 there isn't much more work to be done to get a perfect mark, at least in comparison to IE, Firefox and Opera maxing out at 60/100.

(david13lt said @ #1)
Latest versions:
FireFox: 59
Opera: 60
IE: 17
Safari: 90

Webkit wins, for test I used the lasted Beta versions (nighty builds)

win32 Firefox 3 nightlies have hit 67 for weeks. They pause briefly on 18.

Here on my work computer, I have IE7, Firefox and Opera 9.50b.

Firefox and IE7 both mangle it, and fail miserably. Firefox scores a 50, while IE7 scores a 12. Opera at least looks like the rendering in structure but has no color, and scores a 60.

Safari 2 was the first to pass, then Konqueror, which uses the KHTML engine (which Webkit used as their base, and modified it to their liking (gotta love open source)), made changes to their engine using the code from Webkit. Next, Opera passed, and now the betas of the upcoming major revisions to both Internet Explorer and Firefox pass the test.

One thing that puzzles me slightly... How can they be 100% sure that a fully compliant browser, when it comes out, will actually pass?

I mean, if no current browser actually passes the test anyway; kinda makes it hard to debug the test itself!

Well, Microsoft have only said they have made IE8 so that it passes Acid2, I haven't read anything about Acid3 support yet, for IE8.

I know what your saying, FloatingFatMan. They can be sure because there is a way to code it. If there is a way to make it in code why shouldn't the browser be able to render it.

Because there are specifications for (X)HTML, CSS, Javascript that define how they should operate, and that's how they know whether something passes or not.

The Acid tests are meant to be complex and standards-compliant. They are going by the standards specified by the W3C in order to generate the code that should look a particular way. I think the authors intentionally poke at flaws in particular browsers so that the image does not render correctly. If all browsers render the same thing on a set of rather complex code, then that means the work of joe-web-developer will be so much easier because he doesn't have to worry about how things look across all web browsers.

In the end, the best web designs are the simple ones IMO.