What does Acid3 mean to you and me?

So, last week two browser vendors proudly announced that their rendering engines now achieve a 100/100 score on the Acid3 Browser Test: Opera (Opera and the Acid3 Test) and Apple (WebKit achieves Acid3 100/100 in public build).

Getting a 100/100 score does not mean that the browser has completely passed the Acid3 test, since there are other criteria as well - the animation has to be smooth and the final page has to be a pixel perfect match of the reference rendering. Despite that, it’s great news to see browser vendors in a battle to implement standards first. Too bad the biggest two in terms of market share - Firefox and Internet Explorer - didn’t take part in the Acid3 race.

What I’m wondering is if, how, and when, this will help Web designers and developers like you and me. How long will it take for the other vendors to catch up enough that the standards that are tested by Acid3 can be used reliably? And what parts of the Acid3 test checks stuff that we really can’t wait to use?

What’s your thinking on this?

Posted on April 1, 2008 in Browsers, Web Standards


  1. Even if IE and FF joined the race, the install base would still take quite a while to catch up.

    A lot of people still are afraid to install updates, even if Windows moans at you once in a while or is on by default.

    There are probably bugs in the implementations somewhere too :P

  2. The Acid3 test is a bit of a false positive; as has been noted elsewhere, the WebKit team fixed certain parts of the SMIL standard which were just enough to pass Acid3 but not enough to be useful (that said, they are working on the rest).

    While I love the fact we now have Web Fonts, the fixes aren’t of much use to me day-to-day until ALL browsers implement them; that’s the point of a standard, isn’t it?

    I think that it’s a good thing that aiming to pass the test causes competition between browsers, but the actual passing of the test is less important.

  3. April 1, 2008 by Mathew Byrne

    What I’ve always wondered is what the use of the Acid test is when the test itself is nowhere near representative of a real-world site and how the technology will actually be put to use.

    Not only that, the Acid test is specifically geared toward badly supported features and specific browser bugs. It doesn’t in any way prove the absence of bugs.

    Another thought: does anyone really think that these browsers would really be pursuing getting 100/100 if there wasn’t a number score (ala Acid 2).

  4. Eric Meyer wrote a blog post “Acid Redux” which gives some interesting perspective on this issue:


  5. To me, AcidX tests means neutral judgement of standards compability. On the other hand, one could also expect markup validation to be proof of best practice web development …

    Either way, I think it’s better with one test partly unreliable than none at all.

    Besides, it is the end user, not the developer, which is the important target.

    Both IE and Firefox teams have other things which are more critical than high scores in the acid3 test (IE focus on CSS support, Firefox focus on speed and less memory usage).

    No end user will cry about acid3 races if they twice as fast performance and saves 500MB of RAM in Firefox 3. No end user will complain about acid3 when IE8 makes CSS developers around the world happier.

    (with that, I say that I expect Firefox to pass acid3 quite easily as soon as more critical fixes is done. IE have their own troubles with backwardcompability and slow upgraders, which proparly will slow everything up, making acid3 reaces pretty much useless.)

  6. I don’t think Acid3 means very much. Where Acid 1 was a true test of one (major) element of standards support, Acid 3 isn’t. While passing is certainly better than not passing, it doesn’t mean much apart from that those browsers support those bits that acid 3 tests. Which, despite seeming numerous, is in no way complete.

    Acid has become a trophy for bragging rights, rather than a test for quality. Because vendors can race ahead and implement only the bits that are needed to pass the test, and everyone assumes that they will therefore be better browsers for designers and developers. But passing Acid 3 doesn’t mean they support any of the other bits we might want or need (look at the pathetic SVG implementation in Safari for an example of just how little of that particular spec got implemented in order to pass).

    Acid has become a PR exercise when it ought to be a core test suite (emphasis on ‘suite’).

  7. Because ACID3 is mostly a joke. read Eric Meyer’s post on the subject. Passing ACID 3 means nothing, fixing things just to pass the test is not playing it fair. As far as FF and IE are concerned they have better things to do.

  8. Unfortunately, as you said, IE & FF are not in the race. That’s why Acid3 means nothing. If they were, I’m quite sure browsers would render the things more equally. However, technologies involved in Acid3 are not those we use everyday… But isn’t it because browsers are not compliant with it ?

    I just hope IE8 will be very standard compliant, then it could enter the race and pull FF with (it would be very bad for FF to be less compliant than IE). If they do this, perhaps we can use html5 and css3 in a near future (or not).

  9. It’s a good thing.

    Firstly, it promotes a bit of competition amongst the browser vendors as to who can be the first to complete the challenge.

    Secondly, this particular test was aimed directly at existing browser bugs, meaning they can’t rest on their laurels. After all, there’s a well-publicised test suite to show how well their rendering and scripting engines really perform.

    It also brings some extra PR to the “alternative” browsers and does a great job of pointing out just how far behind the curve certain major players really are.

    As for the point that Webkit enabled just enough SMIL to pass the test — what did people expect, full support overnight? I expect there’ll be better support by the time the next major revision of Safari is released.

  10. Its great that browser developers are moving towards implementing this increased functionality. Unfortunately at the moment the standard is the exception as I still have IE6 showing up on my site traffic.

    Maybe in 10 years I can start using some of the newer cool features available today and curse the lack of support for stuff that wont be widely available for another 10 years after that.

    If you talk to average internet users (take my parents for example) they really don’t understand that there are alternatives to internet explorer. As far as they are concerned that is the only way to get onto the internet.

    The only way I could get them to use firefox was by describing the extra features from a users point of view. When I tried to explain the advantages in terms of what a web site designer can do their eyes glazed over and I could see that they thought all browsers must be the same inside as the internet is the same etc.

    What I am trying to say is that its all very well and good for web people to lord the implementation of CSS 3 in cutting edge browsers, but for the majority of users that don’t know and don’t care, they just want it to work.

  11. Sorry to be banal here, I don’t use Opera but Safari/ webkit and FF. I can’t use gmail in Safari. There are 60+ bug issues. Since I live in Google Safari is not much use to me. Eric Meyer has a good post about why Acid3 is not actually all that amazing.

  12. My first thought when I read your heading? Oh no, Roger, please do not add to the confusion! Neither Presto nor Webkit passes the test. A score of 100/100 is not enough. Pixel perfect rendering is not enough. There is also a speed test which all browsers currently fail. And Opera as well as Apple admit they do. They have never claimed to pass the test. Please do not spread this misinformation further!

    May I humbly ask that you change your blog post to reflect this reality?

    As for the test, everyone who knows something about standards know that there exists a plethora of formal test suites. Many of which are published by W3C. They are however not really “sexy” and do not get the attention they deserve. Browsers vendors know about them and they are working towards passing them. In the end that is much more important than public tests like Acid2 or Acid3.

    But in teaching the general public about standards these test serve a purpose too. I am surprised to see even Eric Meyer writing as if he did not know that there are formal tests still being written for the W3C, since he stopped doing that.

    When will the test start to matter? Opera and Safari support media queries right now. Immensely useful. Anyone ever heard of progressive enhancement. We do not need to wait for IE 10 or 11!

  13. April 1, 2008 by Roger Johansson (Author comment)


    They have never claimed to pass the test. Please do not spread this misinformation further!

    Well, both Opera and Apple representatives need to express themselves a bit more clearly then to avoid any such misunderstandings. Or perhaps I need to read more carefully. Or both :-). Either way, I can assure you that I did not intend to spread any misinformation.

    May I humbly ask that you change your blog post to reflect this reality?

    Of course. I changed “pass the Acid3 Browser Test” to “pass the rendering part of the Acid3 Browser Test”. Is that clear enough?

  14. To you and I, presently, Acid 3 means very little. However, with the movement towards HTML 5 and getting all browser developers on the same page, I expect it’ll have the same impact as Acid 2 (more far-reaching [CSS] standards compliance). Which means, since Acid 3 specifically address ECMAScript and DOM capabilities, that eventually, when it’s actually adopted and all the major players pass, it’ll make our lives easier on that front (just as Acid 2 positively affects CSS support), and more consistency for those on the other side of the monitor.

    That’s how I see it, anyway. I’m sure it’ll be a long path to travel, but it’s all for the future and, someday, all for the good.

  15. Hi again Roger. Maybe I sounded a bit harsh. I have been contributing to the Acid3 article on Wikipdia - trying to write about the test, what is being tested, how it works, etc. Almost everyone else wants to write about the race - and we have had a torrent of people who have claimed either Webkit or Opera passes the test, citing blogs. The official statements are quite clear, I would say.

    The problem with the test is that the speed criteria has no visual clue. It looks like it passes. One has too click the A and read the comments to see that it does not.

  16. April 1, 2008 by Roger Johansson (Author comment)

    Lars: No problem! I like getting the details right, so thanks for correcting me on this. I updated the post again, and I hope there is no misinformation left now :-).

  17. To be fair, Firefox 3 was way too late in the development process to possibly be bothered with Acid 3. IE8, well they are just happy passing Acid 2, so I can’t imagine we should hold our breathe for Acid 3 success any time soon. Either way, tweaking your browser just to pass a relatively meaningless test is poor form. Work to support the standard for that reason in itself, not just to pass a test just to look good in the eyes of a few nerds.

  18. As a way of comparing standards implementation, the Acid tests are useful and it gives browser vendors something to aim for.

    Real world though? IE6 is still used by around a third of web users, so it won’t help web developers for several years to come. The same can probably be said for HTML5: it’s going to be a while before we will see widespread adoption.

  19. Acid3 is nice and dandy but until IE6 has significant market share even just a h2 + p { } can’t be used…

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.