Working together for standards The Web Standards Project

Acid3 nearing completion

By Kimberly Blessing | February 5th, 2008 | Filed in Acid3, Browsers, CSS, DOM

If you’re a fan of the Acid browser tests, you already know that Acid3 is in the works. It’s now in a “final review” state, so please check it out and submit your feedback.

Skip to comment form

As noted previously, Ian Hickson has been hard at work on the Acid3 browser test. We’re happy to report that his work is nearly done!

Acid3 is a complex test compared to Acid1 and Acid2. The Web is increasingly becoming a platform for application development, so Acid3 tests many of the DOM2 and ECMAScript specifications, in addition to markup languages, CSS, SVG, and others. A complete list of specifications tested will be available when the test launches, along with a guide to the test.

Due to the complexity of the test, we’re asking for a final round of feedback to ensure its fairness and accuracy. Please review the test and submit your comments here or .

Your Replies

#1 On February 5th, 2008 10:29 am Chris Hester replied:

Try it in IE7. Oh dear…

#2 On February 5th, 2008 10:32 am Steve Lucas replied:

Doesn’t work in Safari, although I have Javascript and Java enabled.
Message: “Scripting must be enabled to use this test.”

#3 On February 5th, 2008 11:15 am iamshimone replied:

Part of what made Acid2 so neat was the end result – a smiley face that alluded to the Acid test itself. Acid3 is a bit too clinical – where’s the fun?

#4 On February 5th, 2008 2:34 pm Jeremy Gillick replied:

I get a picture of 2 cats in Firefox2/Mac — which obviously didn’t pass.

Right now Firefox2/Mac is at 50%. It fails 50 out of 100, but then I guess I could say it passes 50 our of 100 too. :)

Great work on the test! JavaScript has needed a good Acid test like this. Thanks.

#5 On February 5th, 2008 11:55 pm Krish replied:

56/100 for Firefox 3 beta and 59/100 for Opera 9.5 Beta

#6 On February 6th, 2008 3:27 am Davide replied:

I think it’s time to test DOM2 & C. in XHTML with application/xhtml+xml MIME type. No more document.write and so on!

#7 On February 6th, 2008 10:52 am Steve Lucas replied:

Any idea why this doesn’t work in Safari.
Error Message: “Scripting must be enabled to use this test.”
I have Javascript and Java enabled.

#8 On February 6th, 2008 4:56 pm Michael Kriskovic replied:

I agree with iamshimone. I’d like to see a smiley face that goes from full on frown to full on smile as the percentage goes higher! :-)

#9 On February 8th, 2008 1:14 pm La domo de karotoj replied:

Acidtesto tria alproksimiĝas al kompletiĝo

Ĝi estas en la stadio “fina kontrolado”. Vi povas sendi komentojn (en la angla) kaj vidi la nunan staton de la testo.

#10 On February 9th, 2008 5:34 am ethana2 replied:

Keep up the great work!

epiphany (gecko): 59, no color
FF3b2 (gecko): 56, linktest fail [not using default settings]
konqueror, webkit epiphany, safari in WINE, and opera not tested.

#11 On February 10th, 2008 12:04 am kreed replied:

WebKit/Gtk trunk is at 79%. ^_^

#12 On February 10th, 2008 1:41 am solipsism replied:

WebKit r30109 (2008-FE-09) is at 78/100.

Since Acid3 isn’t completed yet, how many can possibly be passed at this time?

#13 On February 10th, 2008 7:49 am iJoost replied:

CSS for paged media could do with better support in browsers. As it is, not even de hints for page breaks seem to do anything in Firefox. Sad. :-(

#14 On February 11th, 2008 6:16 pm jdehan replied:

I really liked the simplicity of acid2. While acid3 is neat with the animation, I feel it could definitely be more fun. Though looking at the code that doesn’t look too easy and I appreciate the hard work.

#15 On February 13th, 2008 2:03 am ethana2 replied:

Firefox 3 beta 3 (gecko)
59/100, linktest failed.

It would be nice to have a comparison page for various browsers, separated by engine:

trident, presto, webkit, gecko, as represented by
IE, Opera, Konqueror and Safari, Firefox

You know, make this a very direct competition. It would be a really great thing for the world if FF3 passed ACID3 before its release. I think that should be part of what we push for.

#16 On February 15th, 2008 12:07 am Yani replied:

Well I was hoping to test Safari B3 for Windows… only it crashes before the first pages is even loaded. ;)

So I guess that doesn’t pass the distilled water test even.

#17 On February 17th, 2008 2:38 am ethana2 replied:

We’re ready when you are, guys. The second this test is final, I’m getting it into the FF3 bugzilla.

We’re going to need an ACID4 you realize, right? Find an organization that abides by open standards, like Google or Novell, and see if they’ll sponsor you or something. For the future!

We also need a test suite for webpages to see if they’re well written and compliant and valid and all that. If myspace or facebook or someone is using code that sucks, it’d be great to be able to measure that. Death to quirks mode!

#18 On February 19th, 2008 1:55 am mediter replied:

Safari Version 3.1 (4525.9)

passes 73 out of 100

linktest failed

#19 On February 19th, 2008 1:57 am mediter replied:

sorry, forgotten to include OS info.

Mac OS X 10.4.11

#20 On February 20th, 2008 12:36 am ethana2 replied:

I tried it in IE7 on XP.

It wasn’t recognizable, but I think the number it got to was 5.

#21 On February 20th, 2008 1:54 pm Vo0Ds replied:

From my WinXP Home SP2 machine

Firefox 3 Beta 3: 61/100
Firefox 51/100
Opera 9.26: 46/100
Safari for Windows 3.04: 39/100
Internet Explorer 7.0.5730.11: 12/100 – i had to load an activeX control and wait a few mins
Internet Explorer 6 (multiple IE): 11/100

#22 On February 21st, 2008 3:27 pm Greg replied:

SVG is mentioned. It would be great to include the canvas tag in the test as well if it hasn’t been already. A circle with a radial gradient would be great. Semi-transparent stop points in the gradient would be bonus. Adding a shadow to the circle would be a double bonus.

#23 On February 21st, 2008 4:59 pm ethana2 replied:

Greg: just throw them the last inkscape splash ^_^
…although i don’t know if inkscape has made any extensions on top of the w3c .svg specification.. hmm.. I don’t even know whether I hope they did or not. I guess I just mean I hope .svg isn’t holding them back..

/anyway/, could we have an ETA for the final release or something? I’m starting to get tired of checking this site every day…

#24 On February 21st, 2008 6:27 pm Peter replied:

webkit nightly – 85/100
firefox 3b4pre nightly – 67/100
opera 9.5b – 60

#25 On February 22nd, 2008 2:22 am Pyro3d replied:

New WebKit nightly – 87/100

#26 On February 23rd, 2008 7:06 pm ethana2 replied:

So what happens if the latest webkit and gecko pass the test before it’s… out?

..Because if the data Peter and Pyro3d posted is correct, it’s looking like that may actually happen.

#27 On February 24th, 2008 6:28 pm ryan replied:

ie 7 12

#28 On February 25th, 2008 12:11 am ethana2 replied:

My firefox3 nightly gets 66, but it’s probably not default settings.. I was hoping to see a 1 or 2 point increase over Peter’s results..

I would estimate that many of the subtests are now passed by all three viable browser engines. IE, of course, is still a joke. Could we have an update on the final review? Is it almost ready to be declared done?

#29 On February 25th, 2008 4:19 pm GranitW replied:

ethana2, they are correct, Webkit is already at 87, and I can see by the time Safari 3.1 or 3.2 comes out it’ll be near 100.

Webkit is fast in development and a new build comes out about 3 times a week.

#30 On February 26th, 2008 12:14 pm Daniel replied:

I can also assert that Webkit is at 87/100, which, in fact, surprised me due to the low score (about 37/100) of the most recent release of Safari. Nightly builds of Firefox score about a 66/100, and nightly builds of Opera score about a 65/100. IE 7 scores about 12/100, and, just for laughs, I took the test in IE 6. It scored 4/100, I kid you not.

I’m glad that Acid3 is testing scripting compliance. Several major sites that are based on AJAX (Yahoo Mail, for instance) redirect Safari and Opera, even though they work (and work well). Yahoo Mail, in particular, sends Opera back to the original mail interface (as in the original interface from when Yahoo Mail was launched). If you change the user-agent, both Safari and Opera handle these AJAX environments. Since Acid3 is testing JavaScript, etc., perhaps more sites will start to be friendly towards Opera and Safari.

#31 On February 27th, 2008 11:38 am Simon replied:

I agree with Daniel (Feb 26/ 12:14pm) – Long enough have websites been limited to FF and IE only. I hope that the Results from Acid 3 will help change people’s views about Safari.

I’ve found Webkit a very useful browser with amazing support on their IRC, and will continue to use it no matter how many ‘ew’s I get from people who know no better.

So Thank You for the Acid Tests: For helping browsers to become more complaint, and showing people that even the big browsers don’t always get it right, therefore motivating them to improve.

#32 On February 28th, 2008 2:42 am ethana2 replied:

You know you’re a nerd when you come back to the same page every day getting impatient waiting for the release of a web standards compliance test page.

Will some kind fellow please email me when this test is made final?
[email protected]
~thank you very much, in advance.

#33 On March 1st, 2008 4:20 pm Kevin Chen replied:

Now, try it in Opera 7. It passes 25 tests fron Acid 3. If a browser 2 years older than IE 7 passes more tests on Acid 3, there’s a huge problem.

#34 On March 1st, 2008 5:04 pm ethana2 replied:

It seems the test is done. A bug has been filed on Firefox 3 via launchpad.

I just found out Ian works for Google. Man, they’re just awesome all around, it’s amazing.

I think it’d be great to have a full use test for each major script form– an svg page that uses the full svg spec… an HTML5 page that uses the full html5 spec, etc. The web is constantly advancing, constantly evolving.. and the ACID tests are a great way to hold it all together. The fact that the IE team even recognizes their existence says a /lot/.

Keep up the great work, guys: lead us into the future.

#35 On March 6th, 2008 5:47 am Jayce replied:

Webkit r30790 gives 90/100.
Mac OS X 10.5.2

10 to go…

#36 On March 7th, 2008 6:56 am Barney replied:

WebKit r30868 safari 3 beta
Windows XP

88 / 100

IE8 Beta 8.0.6001.17184

17 / 100

#37 On March 7th, 2008 7:09 am Barney replied:

Sorry I have just run the test again with WebKit r30868 safari 3 beta and am now getting 90 on windows XP.

#38 On March 22nd, 2008 9:51 pm Paulo Ricardo replied:

Firefox 3.0b5pre 2008032204 nightly – 71/100

Unfortunately, Firefox 3.0 final don’t will beat the 100/100 rank. It’s expected for 3.x or 4.0.

#39 On March 23rd, 2008 10:23 pm Someone replied:

though i did not trace too deep into the code, it seems in the step 0 (erasing the javascript requirement) you expect a indentation to appear as a text node, which is absolutely not correct and not standard behavior… the standard states clearly that white characters like line feeds and whitespaces/tabs used for indentations are ignored in the dom tree and thus a indentation should never appear as a text node between tags (btw this is a bug that firefox has for way too long and even before firefox in mozilla, and does not seem to get corrected forever, or they don’t want to correct it, making us to put no indentation between tags where we want to use some dom functions like nextsibling), if you are expecting it you may want to re-read the w3c specifications and correct it

then again i repeat i did not went too deep so i can be wrong about you expecting that blank text node, but sure it does look like that

another thing i see not very well is the inclusion of svg, yes, it is standard but not something that all browser have by default, which is very very unfair, the tests should be done with technologies that every decent browser has by default, or at least warn about having the appropiate plugin before taking the test

#40 On March 26th, 2008 12:35 am alcatholic replied:

Reposting from the first Acid3 posting:

OS X.4.11

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

Safari 3.1 gets 75/100 linktest failed

They’ve also just introduced an optimization into Webkit nightlies for pageloading that is very sweet on script heavy pages.

“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.”

#41 On March 26th, 2008 4:09 pm Jean C replied:

51/100 with Orca Browser 1.0 RC3
53/100 with Kamelon 1.1.4
58/100 with Firefox 3 Beta 1

#42 On March 27th, 2008 3:41 pm Davide replied:

Latest Webkit nightly build gets 100/100!

#43 On April 3rd, 2008 12:07 am Kyle replied:

Firefox 3 beta 5 gets 71/100 linktest failed

#44 On April 25th, 2008 10:47 am Guillermo Garcia replied:

Firefox 3 Beta 5 @ Ubuntu 8.04 : 71/100 (.98 sec)

Return to top

Post a Reply

Comments are closed.

All of the entries posted in WaSP Buzz express the opinions of their individual authors. They do not necessarily reflect the plans or positions of the Web Standards Project as a group.

This site is valid XHTML 1.0 Strict, CSS | Get Buzz via RSS or Atom | Colophon | Legal