Working together for standards The Web Standards Project

It’s been three years since we told browser makes that we want to see them smile, but now we wanna hold their hand.

Acid3 goes beyond the CSS tests implemented by Acid2 and tests a browser’s DOM Scripting capability, as well as continuing to probe visual rendering of CSS, SVG and webfonts. Further information can be found on the Acid3 page, in the press release, or you can just go ahead and take the test.

We know that work is already underway based on the Acid3 previews, but time will tell which browser is the first to pass all 100 tests fully, and with default settings. This is going to be interesting.

Your Replies

#1 On March 3rd, 2008 6:29 am Jon Tan replied:

Nice work, folks! All the best to the vendors in their quest to pass the test.

For a quick glimpse, I just Flickr’d a few screenshots of FF 2.0.12, Opera 9.26 and Safari 3.04 taking the test.

#2 On March 3rd, 2008 8:18 am David Naylor replied:

Wow, is it really three years since Acid2 was released? Well I guess quite a lot has happened since then so it is plausible…

#3 On March 3rd, 2008 10:09 am Jason LeMoine replied:

An interesting and optimistic note: the latest version of Webkit (r30707) receives an 87/100.

#4 On March 3rd, 2008 10:21 am Robin replied:

Pity iCab won’t be first to the finish this time, but never mind.

Dave Hyatt: any chance of a series of blog posts explaining the fixes that go into Webkit (à la Acid2)?

#5 On March 3rd, 2008 1:46 pm alex_mayorga replied:

Mozilla Firefox nightly build 3.0b4pre gets a decent 67/100 over here (all gray scale). Hope they’ll pass this one by the release of 3.0 final.

#6 On March 3rd, 2008 4:45 pm Maury replied:

So what browser are you guys using to get a 100% score?

#7 On March 3rd, 2008 5:11 pm WaSP Member faruk replied:

Maury, the 100% score is a reference rendering, not an actual browser result (yet).

#8 On March 3rd, 2008 5:19 pm Sander replied:

alex: There’s absolutely no chance of Firefox 3 passing Acid3. Gecko 1.9 is effectively feature frozen, and major architectural changes will undoubtedly be necessary to complete certain of the missing parts of Acid3, which just won’t be accepted at this late stage. (Though part of the current high score of Firefox is due to it picking up a semi-risky change at a late stage to enable dynamically updating selectors, which was needed for a blocking bug, but also made a difference for Acid3.)
Moreover, the Mozilla backend is due for a major revision (much larger than usual between releases), namely Mozilla 2. Those architectural changes will undoubtedly go into the same Mozilla 2 work. However, due to the nature of the work happening there, it wouldn’t surprise me at all if a Firefox release to benefit from this work wouldn’t be ready for another 2+ years. (There could very well be a Firefox 3.1 (or 3.5 or even 4) release in between, but that would be a Firefox release from the same Gecko 1.9 code, which is guaranteed not to pick up the necessary architectural changes.)

Opera (and Safari?) nightlies are having scores comparable to current Firefox 3 betas, too, so really, which browser is the first to pass the test will mostly depend on fortuitous release schedules. (I very much suspect it’s way too late in the game for Opera 9.5 as well, so there it’d be Opera 10; don’t know what things are like for Safari…)

Maury: no browser currently exists which gets a 100% score, obviously. However, the specifications are precise enough (and Ian Hickson who created the Acid3 test is knowledgeable enough) that it can be deduced how the test should render. (Although there’s always a tiny chance that a flaw in the test is found at some stage and that it will need to be updated; this happened at some point in Acid2, too, iirc.)

#9 On March 3rd, 2008 6:02 pm Maury replied:

I see, thanks for the clarification.

#10 On March 4th, 2008 12:29 pm Lars Gunther replied:

I have been authoring the Acid3 page on Wikipedia, together with a few others.

Now that the test is final, we should update the images of how it renders in current stable browsers. If anyone could help submit some, with Creative Commons License, it would be appreciated.

In the article we have references to the tracking bugs for Gecko, Webkit and KHTML, if anyone wishes to follow the progress. Even though Webkit especially has thrown themselves with abandonment into fixing Acid3 related bugs, I predict it will take a while for them to get 100 %. They need to add SMIL, which today is a non-existent feature.

As for Mozilla, they are focusing more on ECMAScript 4 and Tamarin, so although they may be in a lead of sorts scripting-technology wise, I predict that both Safari and Opera will pass Acid3 ahead of Firefox. Let us also remember that Acid3 is not the final word on standards compliance.

Seeing Webkit go from worst in class when it comes to the DOM and ECMAScript, to having a really good scripting engine is good news of course (no MSIE is not worst in class – it is not in the class at all…)

#11 On March 4th, 2008 12:58 pm Taylor replied:

Even if firefox doesent pass first it renders stuff accepable. Imagine how long it’ll take for IE to pass

LOL

#12 On March 5th, 2008 6:28 pm David Masover replied:

Konqueror seems to pass the Acid2 test, but crashes horribly when attempting Acid3. I can’t wait for a Konqueror with Webkit…

#13 On March 6th, 2008 4:04 am Steve replied:

Webkit r30790 is already at 90/100.
It looks like they’re working fast on this…

#14 On March 9th, 2008 12:00 am Ryan replied:

Steve, you beat me to it. Is this Steve Jobs? jk Webkit will have this by mid-April I think.

#15 On March 10th, 2008 8:37 am Mark replied:

Lars, where would you like the submissions of images showing rendering? Directly on the Wikipedia entry, the talk page or elsewhere?

#16 On March 13th, 2008 1:37 pm Sean Burke replied:

FireFox Minfield for 3/13 is now getting 70/100 so even just after the beta 4 release they are still moving along as well :)

#17 On March 17th, 2008 8:18 pm sd87 replied:

Firefox 3 Minefield for 3/17 has 71/100! Hopefully they pass Acid 3 by the FF3 release.

#18 On March 18th, 2008 9:25 am satcomer replied:

The WebKit now gets a steady 93 on the Acid 3 test. I can’t wait for the final standard. Maybe someday all browsers will adhere to standards so everyone could have an equal experiences on all web pages.

#19 On March 18th, 2008 10:25 am Mark Howson replied:

Safari 3.1 is getting a steady 74… highest score for a production browser, I think.

#20 On March 18th, 2008 1:52 pm Riku replied:

Opera 9.50
Windows 9841 build
18. March 2008

77pts.

#21 On March 18th, 2008 4:55 pm Aulo replied:

Firefox 3.0 beta 4 steady 68pts.

#22 On March 19th, 2008 8:08 pm mediter replied:

Safari Version 3.1 (4525.13)

Mac OS X 10.4

75/100, but still link test failed

#23 On March 23rd, 2008 1:06 am Jeff replied:

WebKit as of March 22nd 2008 nightly build is at 95/100 now! Looking great and WebKit has by many magnitudes a faster javascript, dom and svg implementation versus Gecko and IE.

Go WebKit!

#24 On March 23rd, 2008 7:17 am RZPogi replied:

M$ doesn’t care much about acid3 but IE 8 beta is passing

#25 On March 26th, 2008 12:50 am alcatholic replied:

Reposting:

OS X.4.11

Webkit Nightly Build (r31307) (with Safari 3.1 installed, but not used) now gets 96/100

They’ve also just introduced an optimization into Webkit nightlies for pageloading that is very sweet on script heavy pages.
http://webkit.org/blog/166/optimizing-page-loading-in-web-browser/

“The latest WebKit nightlies contain some new optimizations to reduce the impact of network latency. When script loading halts the main parser, we start up a side parser that goes through the rest of the HTML source to find more resources to load. We also prioritize resources so that scripts and stylesheets load before images. The overall effect is that we are now able to load more resources in parallel with scripts, including other scripts.”

#26 On March 26th, 2008 2:40 pm Nao replied:

Looks like Opera is officially the first to pass the test ;)
http://my.opera.com/desktopteam/blog/2008/03/26/opera-and-the-acid3-test

#27 On March 26th, 2008 7:11 pm DWalla replied:

Apple’s current nightly WebKit build is now at 98/100 on the Acid3 test.

#28 On March 26th, 2008 7:44 pm Fernando replied:

Opera Rocks!!!!

THE FIRST ONE THAT PASSES THE ACID3 TEST
http://my.opera.com/desktopteam/blog/2008/03/26/opera-and-the-acid3-test

#29 On March 26th, 2008 10:34 pm alcatholic replied:

Congrats to Opera!

Webkit also just made 100/100, and there is a stable, public release.

http://webkit.org/blog/173/webkit-achieves-acid3-100100-in-public-build/

Congrats to the Webkit crew!

#30 On April 1st, 2008 6:17 am ethana2 replied:

Firefox3 beta5 comes out soon! I can’t wait to see what it scores…

Here’s hoping for a firefox 3.0 release that passes acid3 completely!

#31 On April 2nd, 2008 3:38 pm Tristan Young replied:

Firefox 3 Beta 5 scores 71/100 on ACID3, which is better than Beta 4, which was 67 if memory serves me.

I was hoping to see the various squares turn different colors, rather than shades of gray. I hope this will be fixed in time for the official release.

There are also a few artifacts (2 small red squares, white X against a magenta background).

I’m glad to see improvements being made. It’s only a matter of time.

#32 On April 10th, 2008 6:09 am Robert Orso replied:

I’m not convinced that Mozilla will get a 100/100 score in the next couple of years. While the Gecko-Engine improves from version to version, the ACID3 test checks, among others, the correct implementation of “CSS2 (@font-face)”.

This includes the support of the “src” parameter to specify a dynamically loadable font-file.

You did note that this is a CSS2 feature, did you? A specification that dates back to 1998.

As Netscape released the Source of the browser back in 1998, the community had to remove the support for dynamically loaded fonts due to copyright-issues. It’s a long lasting and unresolved bug (https://bugzilla.mozilla.org/show_bug.cgi?id=52746, https://bugzilla.mozilla.org/show_bug.cgi?id=70132) with no hint of progress in the mean time.

The licenses of the font vendors obviously require a “closed source” implementation and the Mozilla license obviously prohibits just that – the utilisation of a closed source library. A classical deadlock.

We had the 10th aniversary of this unfortune happening lately.

Maybe the OpenType specification and a set of public domain fonts can solve this issue one day.

#33 On April 10th, 2008 12:56 pm Resimler replied:

I like this post.Hopefully they pass Acid 3 by the FF3 release.

#34 On May 14th, 2008 8:43 pm Marte replied:

Epiphany 2.22.1.1, which uses Webkit as the rendering engine, completely pass the test.

Post a Reply

Comments are closed.