• 200826 Mar

    Update by David: Opera has now released a public WinGogi build of Opera passing the DOM test and with pixel perfect rendering. We don’t believe we have passed the test yet, as there are performance issues with a couple of tests. This puts Opera and Safari neck and neck. It is fantastic to see both the Apple and Opera teams taking this test seriously and pushing each other to improve our standards support. Who ever wins the race doesn’t really matter, the main point is that there are now two engines with the required standards support to pass Acid 3. This could be a shot in the arm for both CSS3 and SVG.

    Update by David: Although the Acid 3 test was updated to allow WebKit to pass by the letter of the test, they still seem to fail by the spirit of the competition. One of the sections of the Acid 3 test checks SVG Animation. WebKit have added the interface to allow the test to pass, but not fully implemented the feature it was testing. See Jeff Schiller’s blog for more information.

    Update by Joost: While Opera might have been the first to pass the test in laboratory conditions, for which I applaud them and I hope they release it ASAP, you can download a WebKit nightly right now and enjoy the full 100/100!

    Update In the last few minutes (while I was eating lunch) the final 2% was reaching, making Opera the first browser to reach 100%. There are still some rendering issues, but things are well on track to passing the test. A preview build will be released on Opera Labs shortly. Thanks to our developers in Scandinavia that have been working into the evening to reach 100%.

    Acid3 in Opera

    Safari has been making great gains in its Acid3 score in recent weeks, currently residing on 96%. Opera however has come out of the chasing pack and moved from 77% in the latest weekly release of Kestrel, to 98%100% in the latest internal builds. As part of this it also includes the long requested CSS3 HSLA and RGBA support, and Web Fonts.

    These improvements wont be included in a weekly Kestrel build any time soon. Opera, like Mozilla, are at a stage on our development process where we are closing in on a release, and thus regression testing and stability are critically important. This work will most likely (although not confirmed) go into a post Kestrel release, in case it causes regressions and the like. There will probably be an experimental alpha release showing this improved support in the not too distant future.

    You can skip to the end and leave a response.

  • Comments

    • 01.

      Everyone say YAY for HSLA!

    • 02.
    • 03.

      […] from 77% in the latest weekly release of Kestrel, to 98% in the latest internal builds” (src) Spread the […]

    • 04.

      Great! Let’s wait for a build…
      Also hope to see more than 77% in Kestrel.

    • 05.

      can you post a screenshot of this internal build on the acid3 testpage?

    • 06.

      Isn’t anyone else getting a bit bored with these “look we can pass the tests in an internal build, but none of this will be released any time soon” anouncements?

      I’m impressed with features that are in real browser installations on real peoples computers. I want features that can be actually used, passing tests surely means little unless the features are solid enough to make it into the wild.


    • 07.

      WebKit is at 98/100 now…

    • 08.

      I think opera just want to be the first, they wont let safari take it before them. But I couldnt agree with you more Ed, Its useless if they dont actually put it in the browser (Even though when they do it will be almost as worthless since they will be the only 2, with their very little market share.)

    • 09.

      6: even if the annoucement came with a useable snapshot build, you’d still need sites that use code for this and internet explorer to finally die so advanced css features spread..

      also this is the first time we saw an announcement about an internal release, the rest from every single browser vendor came with usable preview builds

    • 10.

      Who are these “every single browser vendor”, csulok? Sounds like there are a lot of them :D

    • 11.

      Can’t help but agree with Ed Everett David :) I’m just as psyched as you are over the speed of development, but show us some of the stuff :) (or at least email ME that build ;) )

    • 12.
    • 13.

      Great job Opera devs!
      And congrats, hopefully we’ll see as much as possible of these fixes in 9.5.

    • 14.

      Thanks Tim. We were talking to the other Tim and we were on the verse of 99% at that stage. Great to see us reach 100% while I was at lunch.

      Joost/Ed: It is unfair to demand such a build instantly. We have basically just checked in these changes, and there is no way a public build can be made that fast. This site is about future technology, and we’ve just added a couple of CSS3 properties, so I thought it would be best to announce it as soon as we have it. This is post Kestrel work, as we are deep into our development testing. These kind of changes require a lot of work that could cause regressions that will take many QA hours to ensure they are up to release quality. Safari were very fast recently at adding Acid3 as they have already released their major release and don’t have to focus on the stability as much at this stage, like Mozilla and ourselves do. The build with these improvements will be released as soon as possible however. Some fixes may be included in Kestrel but it depends how risky the changes are. HSLA and RGBA are quite a big change for example as colour has to be represented with the alpha channel internally instead of just RG and B channels. This changes code in a lot of modules.

    • 15.

      Wow, what a race it turned out to be, even if we didn’t know that Opera were actually working on Acid3 until today.

      Glad that Opera beat the webkit team to it. The Mozilla guys have plenty of time to get Acid3 right, probably until summer/winter 2009 (Firefox 4).

    • 16.

      I think it’s also worth noting that for the Acid2 test, it took over a month after Safari claimed to pass it for a public build to be released. I don’t think many people complained about that at the time.

    • 17.

      But Fyrd if you look carefully “internal” Opera doesn’t pass Acid3. There is no space between coma and ‘a’ mark in a sentence “To pass the test, a browser must use its default settings”. There is a strong condition to pass the test: “the final page has to look exactly, pixel for pixel”.

      Also other condition is: “the animation has to be smooth” If there is no public version of “internal” Opera how you can claim that Opera pass the test if cannot check smoothness of the test?

    • 18.

      @sentio: I don’t see where Fyrd said Opera passes Acid3. Opera does not pass the test and we haven’t made any claim saying so.

    • 19.

      Tim, to be fair, most people don’t notice that there are other strong conditions despite 100/100 score. My question to Fyrd was rhetoric. It points this misunderstanding what passing Acid3 really is.

    • 20.

      I would like to reiterate Tim’s comments.

      I think too much attention is being given to the score percentage itself, rather than the smooth rendering and pixel-for-pixel likeness to the reference rendering (although judging by the screenshot David provides, it looks pretty spot on), that is required to pass the test.

      Even though the internal build passes all 100 tests, Opera are obviously aware of some additional issues that need to be ironed out- to quote David, “There are still some rendering issues, but things are well on track to passing the test”.

      @Ed: As I’ve mentioned before, neither David or Tim suggest that Acid3 has been passed; David has merely indicated that all 100 tests that are part of Acid3 have been passed. If the guys over at Webkit release a nightly that passes all three criteria, before Opera release a public build incorporating Acid3 support, then I would deem Webkit to be the winner.

      Still all to play for, I think.

    • 21.

      […] It’s always nice to be first. […]

    • 22.

      Rather than get caught up in who earns bragging rights for reaching the milestone first (and I think the first browser to make a release with a 100% score should get that!), what we should be celebrating is that browser makers are taking standards and interoperability seriously and that rendering quirks are becoming a thing of the past. We all win when that happens.

    • 23.
    • 24.

      @Ed Everett I totally agree with you. It seems that many of the browsers are on the cusp of releasing a new version and we’re all just impatient to see the results

    • 25.

      GASP! Judging by that screen shot somebody is using Windows Vista. It will be a few minutes before the tar and feathering begin. Well, JK, I actually have Vista at home, don’t have too many problems I didn’t see in XP.

    • 26.

      WebKit is now at 98/100 btw, on a downloadable nightly ;)

    • 27.

      […] (via CSS3.info. More at OperaWatch.) […]

    • 28.

      How about Acid 1 test pixel-to-pixel rendering?
      Seems to me that only IE and Firefox pass this. Safari, Opera, iCab also passes, but not pixel-to-pixel because of rounding differences:
      Opera, Safari and iCab use flooring
      Internet Explorer, Mozilla use rounding

      Firefox 3 does best job for now…

    • 29.

      sentio: comment 21 proves your point exactly…
      Anyway, it’s obviously all about the publicity. Kind of sad, but still good at the same time, as Peter pointed out.

    • 30.

      Congrats on the 100% achievement. However, I’m still a WebKit junky since I can use the latest WebKit build as my everyday browser, so I’m enjoying 98% now. :D

    • 31.


      What are you talking about? Comment 21 proves nothing. It’s not an official Opera site (anymore) since Daniel left Opera, and it’s a misunderstanding on his part. None of the blame is on Opera since Opera Software never claimed to pass Acid3. In fact, the desktop blog specifically points out that there is still work left to do.

      You need to stop being dishonest and making claims about Opera Software that are demonstrably false.

    • 32.

      Congratulations to the Opera Desktop Team and core developers!

    • 33.

      A little question about Acid3. It seems to me to be pretty strange to have a test where 100 out of 100 is not a pass.

      If rendering issues can still fail it even though a ‘max’ is reached, shouldn’t the test be re-designed to encapsulate those failures in a reduced score? 100 out of 101?

    • 34.
    • 35.

      @Mike S: We will need to retest Opera, but just because the test changed doesn’t mean that Opera won’t score 100/100 on the updated test.

      @Andrew D: The pass conditions for the Acid3 test are pretty clear (they’re written right in the test). The N/100 portion is assessed via JavaScript/DOM. The rendering portion of the test can’t be assessed that way.

    • 36.

      @Tim Altman

      Heh, written in the test but changing as we go.

      A test that requires pixel by pixel comparison seems kindof perverse doesn’t it? Surely there has to be a better way of doing it? Would the changes that are being tested using per-pixel differencese not be possible with using 10pixel or something that becomes instantly obvious that it has failed if a visual comparison is the only way of doing it?

      (I’m never going to read the Acid3 test to find the pass conditions; just commenting that in general a test that relies on such acute attention to determine pass or fail seems to me to be a bad test)

    • 37.

      Sitting here at my cushy desk, I was able to DOWNLOAD and USE the nightly build of Webkit, and it, currently, in a browser I use day to day, gives me 100. Sorry to say, but bragging rights go (yet again, like with Acid 2) to Webkit and the Webkit team…

    • 38.

      WebKit achieves Acid3 100/100 in public build
      “With r31342 WebKit has become the first publicly available rendering engine to achieve 100/100 on Acid3.”

    • 39.

      None of this is useful until a full release has been made. What good is having a nightly/Alpha release which passes the Acid3 test if we then have to wait for 6 months until people start using it? Calm down, everyone, and wait for someone to actually release this.

    • 40.

      @David Storey: I’m not asking for the build now and I fully appreciate all the work that’s being put into making this stuff happen. It’s just that what makes a difference to the web developers (and the web) are the products that are on people’s computers. I look farward to being able to use these features, but I reserve my excitement for actually using them.

      My criticism was not so much of the browser developers, but more the web developer community that is hyping a vacuous race to pass Acid tests in builds that aren’t available to the public. Lets collectively get excited by proper releases and increases in the alternative browser market share – this would be stuff that actually lets us build better websites.

      @ All browser vendors (even MS): Good work recently :-)

    • 41.


      I don’t know what sentio had in mind, but that wasn’t what I was implying, I wasn’t accusing Opera of anything. I was just highlighting the fact people would easily misinterpret the facts.

      That said, now that you say it, it’s true that it sounds a lot like Opera tried their best to *make it look like* they passed the test first. If they truly wanted to avoid any misunderstanding, they could have easily done so in the article, but it looks like they preferred to keep it vague, for people to draw their own conclusion.

      That’s a lot like what the IE team did with the Acid2 test, I think. They didn’t literally claim IE8 passed the test (which would have been a lie, since the default rendering mode back then was the one of IE7), but cleverly implied it without saying it.

    • 42.

      Ed: We have to pass it first in a development build before it gets included in a public release :) This is just stage one of getting what you wish for.

      Stifu: I’m not sure I like your allegations. When and where did Opera try their best to make it look like we passed the test first. In every post I know from an Opera employee, we’ve made it clear that we were the first to reach 100% but haven’t passed the test. There is no debating that. I’m sorry if people can’t read correctly.

    • 43.

      I wonder why Opera felt the need to pull this stunt. Safari will still be the first publicly released non beta/nightly browser to pass while Opera 9.5 won’t even come close to this WinGogi alpha’s score as everyone from Opera says Acid3 won’t block the final release of Kestrel. Also it seems funny the Opera employees feel the need to defend themselves when they think they did nothing wrong.

    • 44.

      @Anon: I strongly disagree with your suggestion that this was a stunt by Opera.

      As I’ve mentioned before David merely pointed out the fact that an internal build has passed all 100 DOM tests, NOT that they’d completely passed Acid3. He sums it up quite nicely,”I’m sorry if people can’t read correctly.”

    • 45.

      Hi David,
      One thing your automated test fails to test, so I don’t really know if it’s in Acid3 or not, is the :lang() selector in xml-based doctypes.
      I use xhtml 1.1, so that “lang” attribute no longer exists and the CSS3 specifications explicitely ask implementors to use the value of “xml:lang” instead. Sadly, I haven’t see any browser supporting this yet…

    • 46.

      I’m sorry if my comments came off as insulting… but I just thought that if the Opera team truly wanted to avoid any (predictable) misunderstanding, they could just have simply added something like “Although we reached 100%, that doesn’t mean Opera pass the test yet, as we still have to fix this and that” in order to clear things up for everybody.
      Among all the people excited about the Acid3 test, commenting on it etc, how many actually understand what it’s really about, and what’s required to actually pass it? From some random sites/articles I read, many are clueless… So, it wouldn’t hurt to explain things clearly.

      That said, I have to agree with Anon about Opera people being overdefensive. Wut kinda jumped at me to defend Opera while I wasn’t attacking them at all in the first place, he only gave me the idea to do so. :p

    • 47.


      “it’s true that it sounds a lot like Opera tried their best to *make it look like* they passed the test first”

      You are lying on purpose, aren’t you? Here is what Opera actually said:

      “There are some remaining issues yet to be fixed, but we hope to have those sorted out shortly.”

      Why do you feel the need to constantly lie and misrepresent Opera’s statements?

    • 48.

      […] It’s an exciting time to be a web developer, as all four major browsers have released / are releasing new versions with extended CSS & HTML support. However, as Opera and Webkit race to be the first to score 100% on the Acid3 test, a lot of people are getting caught up in the excitement and turning this into some kind of pissing contest. […]

    • 49.


      You’re getting the wrong idea. I actually like Opera and all, and have not been known to troll and bash them for no reason (I actually promote it on my site). And yes, I must have forgotten about the fact they said there were remaining issues, so sorry about that.

      However, you pointed out earlier that this person who claimed Opera passed the test was an ex Opera employee, something I didn’t know. And although this person worked in the development of a browser, and so should know better, he still couldn’t even tell Opera didn’t pass the test. If such an experienced person makes such a mistake, what do you think “less knowledgeable” people will do?

      So yeah, the blame still clearly can’t go to Opera for that kind of misunderstanding, I have to admit. But I was also bothered by how overmediatized Acid3 is, and how Opera tries so hard to be the first just for the publicity (I can’t blame people who call that kind of stuff “PR stunts”), while much more serious bugs (that have a real impact on the web as it is today) are left hanging for months or years. To clear that up, yes, it’s good to pass the Acid3, and standards matter and all, but there are more urgent priorities to me.

      Maybe you’d think the same could be said about Safari, and that may be true… but from my experience, Safari seemed less flawed than Opera, making concentrating that much on the Acid3 “more justified”, maybe.

    • 50.

      David, it is true that our SVG animation implementation is pretty incomplete. Obviously we don’t consider it done. But it is not our fault that the test passed with barely more than turning on the temporarily disabled code that we already had.


Advertise here?