Acid3 Test Released

An anonymous reader writes ""The Web Standards Project has announced the release of Acid3, the latest test designed to expose flaws in the implementation of mature Web standards in browsers. 'By making sure their software adheres to the test, the creators of these products can be more confident that their software will display and function with Web pages correctly both now and with Web pages of the future. The Acid3 Test is designed to test specifications for Web 2.0, and exposes potential flaws in implementations of the public ECMAScript 262 and W3C Document Object Model 2 standards.' Screenshots at the Drunken Fist site show the success of Safari 3 (which originally scored 31, but is now Scoring 87/100) IE6, and IE7 (massive fail, of course)'." There are additional discussions of the new test happening around the web.

Comments

Bad day for IE8

[Anonymous Coward]

It gets... 17. Heard at Microsoft "ACID3? We just passed ACID2! AH CRAP!"

Re:Bad day for IE8

[Anonymous Coward]

Yeah, well, Mozilla is also just passing ACID2 in FF3 as well.

And does not pass ACID3.

Re:Bad day for IE8

[Jugalator]

At least that beta doesn't crash. When I ran on a recent Opera 9.50 beta build, it counted, stalled, stalled, crashed. ;-)

Re:Bad day for IE8

[zuro]

actually, i just tested IE8 beta on acid2, and despite what they say, it fails

Re:Bad day for IE8

[Your.Master]

The test is screwed up again, probably because so many people are rushing to test IE8 in it. No browser passes Acid2 from the original site at this time. IE8 passes Acid2 mirrors, like the other browsers.

Re:Bad day for IE8

[jimbojw]

Hey now, at least it passes the reference rendering, which is more than I can say for some browsers (*cough* lynx *cough*)

Re:Bad day for IE8

[SanityInAnarchy]

My Konqueror passes Acid2, and crashes instantly on Acid3.

I'm told Webkit fares better -- can't wait for a Webkit-powered Konqueror.

Re:Bad day for IE8

[Beale]

Safari3 passes Acid2, no?

Re:Bad day for IE8

[dwater]

My webkit-based browser fails immediately. It doesn't crash though, as does FF3.0b3 on Mac.

Re:Bad day for IE8

[RiotingPacifist]

yeah its like they wrote the code before apple went of and used improved it (just a shame apple are terrible with commits and that's why Darwin & the initial khtml fork effectively died)

Re:Bad day for IE8

[erikdalen]

Here's a shot of MSIE 8 Beta with the ACID3 test:

http://www.nada.kth.se/~edalen/msie8beta1-acid3.png

It shows some really strange stuff during the load and warns you that the page wants to run the MSXML Add On.

Re:Bad day for IE8

[JackieBrown]

My Konqueror (4.0.2 from the Debian experimental repos) did not crash but it did fail.

It did better than my firefox 2 (to be fair I did not bother to disable the extensions.) Konqueror had 62/100

Re:Bad day for IE8

[99BottlesOfBeerInMyF]

At least that beta doesn't crash. When I ran on a recent Opera 9.50 beta build, it counted, stalled, stalled, crashed. ;-)

What OS? Opera 9.5 beta works fine for me on OS X and gets 59/100. The only things that crashed for me were Shiira on OS X and Konquerer 3.5.2 on Kubuntu.

Note, the best score I'm getting is from Safari 3.0.4 with a nightly Webkit on OS X, with a score of 86/100.

Re:Bad day for IE8

[hmallett]

Lynx spoils you. I prefer wget.

Re:Bad day for IE8

[elrous0]

Well, it takes time for the folks at the "Web Standards Project" to think of new ways to fuck up Internet Explorer while still working with more geek-fashionable browsers.

Re:Bad day for IE8

[freyyr890]

wget? Hah! Real die hards telnet directly to the server and pass the HTTP commands themselves.

Re:Bad day for IE8

[alanhogan]

Same here: build 4506 of Opera 9.50 beta on OS X gets 60/100.

Re:Bad day for IE8

[cbiltcliffe]

Real die hards telnet directly to the server and pass the HTTP commands themselves.

Telnet? Man, you kids have it easy. We used to compose individual network packets manually by coding assembly on the fly......

Re:Bad day for IE8

[Anonymous Coward]

Assembly? We had to write out the 1's and 0's on the wire directly!

Re:Bad day for IE8

[oliderid]

Assembly? We had to write out the 1's and 0's on the wire directly!

Writing? I sing the bytes stream just like my good old zx81 tape drive.

And just in time

[drspliff]

for IE8 to be released, finally claiming ACID2 support out of the box...

Now, I wonder if their PR people will slip up and accidentally write "ACID3" just to get any sort of news...

I would check out the screen captures, but...

[The Ancients]

Error establishing a database connection

That was fast. Even for slashdot.

Re:I would check out the screen captures, but...

[emilper]

yes, looks like the acid tests are failing the slashdot test.

Too late for IE8?

[riceboy50]

This might be coming onto the scene a little bit too late in order to complain about the upcoming IE8 not passing the test. It's a shame that this didn't come out last year.

Re:Too late for IE8?

[CastrTroy]

If they actually implemented the standards well, they wouldn't have to worry about specific tests, they would just do well on them by default. Now I don't think that any browser does 100% on Acid 3, but I think a lot of them do fairly well.

Re:Too late for IE8?

[Goaway]

None of them do "well". The test is specifically designed to break them all.

Re:Too late for IE8?

[edwdig]

If they actually implemented the standards well, they wouldn't have to worry about specific tests, they would just do well on them by default.

Have you ever tried reading the HTML/CSS specs? They're huge and often vaguely worded. There were often sections that just weren't intuitive, and the only real approach to implementing them was to just figure out what other browser did and copy it. The specs were created by people who have no intention of implementing them themselves, and it really shows.

Re:Too late for IE8?

[Myopic]

I don't think 100% correct is the standard of doing "fairly well".

Re:Too late for IE8?

[mrslacker]

The specs were created by people who have no intention of implementing them themselves, and it really shows. You're talking about OOXML right? Oh, wait.

Re:Too late for IE8?

[RiotingPacifist]

Too late for FF3, IE8 is only just about to hit betas so it stands a chance, but FF3 is still fairly competitive for a browser that has done specific for the test ( it gets ~50 more than IE)

Re:Too late for IE8?

[beoba]

[citation needed]

Re:Too late for IE8?

[Plutonite]

Yeah well 50% isn't fairly well either (FF3 score). These tests are a difficult philosophical question.

Firefox

[BlowHole666]

success of Safari 3 (which originally scored 31, but is now Scoring 87/100) IE6, and IE7 (massive fail, of course)'." Umm what did Firefox get on this? What about Opera? If you are going to report something why not report all the facts. You listed three browsers where are the other two+ ?

Re:Firefox

[brunascle]

the test is here.

i'm getting a 50/100 in Firefox.

Re:Firefox

[Anonymous Coward]

Confirm the 50/100 on firefox 2 and I get 46/100 on Opera 9.25

Re:Firefox

[Jugalator]

Ooh, my Opera 9.50 weekly actually didn't crash this time. Maybe the test was changed, or something in Opera did.

Anyway, Opera 9.50.9807 receives a 65.

Re:Firefox

[caerwyn]

Interestingly, I'm not getting an 87 with Safari 3.0.4- I'm getting a 39.

Re:Firefox

[wile_e_wonka]

In firefox 2 the test shows a red picture of a cat.

Very interesting. I also got a 50/100. The rectangles are all messed up--they're gray, stacked vertically, and each rectangle is the width of the test. The test name also has no shadow.

Re:Firefox

[Cap'n.Brownbeard]

Same score for me using FF 2.0.0.12 on WinXP.

Re:Firefox

[ergo98]

Umm what did Firefox get on this? What about Opera?

One the nightly build of Firefox 3 (built in VS 2008, running on Vista x64), it yields 67/100.

Having said that, is it really such a quantitative, numerical test?

Re:Firefox

[potatog]

firefox 3b3 32bit (gentoo x86-64) scores 57
rectangles are gray but well placed
some visual garbage: there are two red squares and 'X' on purple background
vertical borders of two rectangles are broken
no text shadow

Re:Firefox

[CrazedWalrus]

I was wondering the same thing. Isn't it FF3 that just began rendering ACID2 correctly?

Besides, I see these as a process or goal -- giving the browser makers something concrete and visual to shoot for, as well as an easy way for users to judge the quality of their browser of choice. If the thing was just released, I'm not really surprised that many of the browsers don't pass it completely. Now a year or two from now is a different story, after the browser makers have had some time to address the issues the test points out.

Re:Firefox

[Naughty Bob]

Acid 2 showed the same picture with firefox, though I thought it was a rabbit.

When these Acid tests are 'testing', is anyone else reminded of Fry playing the holophonor, without worms?

Re:Firefox

[bunratty]

The test consists largely of 100 JavaScript tests designed to throw an assertion on failure and return a certain value on pass. The score is how many of the tests out of 100 pass. You can see which tests failed by clicking or shift-clicking the A in Acid3 after the test completes. In the sense that each test can relatively independently pass or fail (although some tests depend on previous tests), yes, it is a quantitative test.

The other part of the test is rendering the Acid3 text with shadow and the colored rectangles. By seeing how the Acid3 test fails in many other browsers, you can see that it can also render X, Fail, and a picture of a cat on failure of some rendering tests, typically in red so they stand out.

Re:Firefox

[c00rdb]

Firefox 3 nightly build got 67/100.

Re:Firefox

[Random BedHead Ed]

Hey, me too. Are you also using Netscape Navigator 4.01?

Re:Firefox

[Anonymous Coward]

I'm getting 404 in firefox.

Re:Firefox

[oopsilon]

I got a 65 in the SeaMonkey (2.0) nightly. Build string for the nightly:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b5pre) Gecko/2008030501 SeaMonkey/2.0a1pre

The Firefox nightly should be pretty close to that since that's also using Gecko 1.9b.

For reference, SeaMonkey 1.1.8 scored 49.

Re:Firefox

[Anonymous Coward]

That was one of the issues with acid 2 - it was just a picture, and not everything shows up perfectly in the render (like the eyes changing on mouseover). However, even acid two was just testing sixty some individual CSS rules. Acid 3 tests 100 and actually tells you the number of mistakes (and which ones, I bet, the test is down now) in addition to the picture.

Since CSS is just a collection of rules (that's what a standard is) it isn't hard to test each one (or some selection thereof) and give a count of how many are followed correctly and how many fail.

Re:Firefox

[bunratty]

Assertions aren't thrown, exceptions are.

You're right... many of the Acid3 tests use assert and assertEquals to test that results are as expected, and these methods throw an exception when they aren't.

Re:Firefox

[uzytkownik]

57/100 in Firefox: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9b3) Gecko/2008020513 Firefox/3.0b3
49/100 in Epiphany (on Gecko 1.8)

Re:Firefox

[Carbonis]

I got a 49/100 on Firefox 2.0.0.12 On Minefield (firefox3 nightly build) 3.0b5pre, I got 67/100

Re:Firefox

[dal20402]

It's misleading for the summary to say "Safari" gets 87/100 when the version of Safari that does that is not yet released.

Run current WebKit nightlies to get the high score now. The changes will be in the upcoming Safari 3.1 release.

Re:Firefox

[caerwyn]

So I did finally notice. I only managed an 85 on the nightly WebKit that I grabbed, however, leaving me to wonder about the discrepancy there.

Re:Firefox

[Spy der Mann]

I got a 100/100 on netcat... but man, finding the 100 under all that bunch of code sure was difficult!

Re:Firefox

[Ken_g6]

For some reason, I got 51/100 on Firefox 2.0.0.12 on the second try.

The failed tests are as follows (shift-click on "A" to see yours):
Note: Test 61 expected *two* spaces, got just one.

Failed 49 of 100 tests.
Test 0: expected: pre-wrap, got: normal - found unexpected computed style
Test 1: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMDocumentTraversal.createNodeIterator]
Test 2: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMDocumentTraversal.createNodeIterator]
Test 3: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMDocumentTraversal.createNodeIterator]
Test 4: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMDocumentTraversal.createNodeIterator]
Test 7: Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIDOMRange.cloneContents]
Test 8: Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsIDOMRange.setEnd]
Test 9: expected: Hello Wonderful KittyHow are you?, got: - toString() on range selecting Document gave wrong output
Test 10: expected: result, got: - toString() didn't work for attribute node
Test 11: Index or size is negative or greater than the allowed amount
Test 12: collapsed is wrong after insertion
Test 22: expected: 5, got: 14 - wrong exception for createElement('0div')
Test 23: expected: 5, got: 14 - wrong exception for createElementNS('null', '0div')
Test 27: e2 - parent element doesn't exist after waiting
Test 30: Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsIDOMEventTarget.dispatchEvent]
Test 31: capture handler called incorrectly
Test 33: expected: 1, got: 0 - whitespace error in class processing
Test 35: expected: 0, got: 1 - :first-child still applies to element that was previously a first child
Test 36: expected: 0, got: 1 - :last-child matched element with a following sibling
Test 37: expected: 1, got: 0 - :only-child did not match only child
Test 38: expected: 0, got: 1 - adding children didn't stop the element matching :empty
Test 39: expected: 1, got: 0 - :nth-child(odd) failed with child 0
Test 40: expected: 1, got: 0 - part 1:0
Test 42: expected: 1, got: 0 - rule did not start matching after change
Test 46: expected: uppercase, got: none - case a failed (index 1)
Test 47: expected: none, got: auto - cursor none not supported
Test 51: expected: 6, got: 5 - wrong number of rows
Test 54: expected: HIDDEN, got: hidden - input control's type content attribute was wrong
Test 60: attribute not specified after removal
Test 61: expected: te st , got: te st - class attribute's value was wrong
Test 67: when calling removeNamedItemNS in a non existent attribute: no exception raised
Test 68: Unpaired surrogate handled wrongly (input was 'text', output was 'ext')
Test 70: UTF-8 encoded XML document with invalid character did not have a well-formedness error
Test 71: expected: null, got: - internalSubset wrong (first test)
Test 72: expected: 20, got: 10 - change failed to take effect
Test 73: expected: 10, got: 0 - click event handler called the wrong number of times
Test 74: getSVGDocument missing on element.
Test 75: anim.beginElement is not a function
Test 76: expected: 0, got: 100 - Incorrect animVal value after svg animation.
Test 77: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMSVGTextPositioningElement.getNumberOfChars]
Test 78: Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIDOMSVGTextPositioningElement.getRotationOfChar]
Test 79: An attempt was made to create or change an object in a way which is incorrect with regard to namespaces
Test 82: unexpected 1 in t3
Test 84: illegal radix 0
Test 88: \u002b was not considered a parse error in script
Test 92: expected: function Object() {\n [native code]\n}, got: functio

Re:Firefox

[KDEWolf]

I guess it was something Opera did. I'm having also no more crashes on the test or when closing the browser. Also it seems much more responsive, possibly faster. Getting 65 as well, using build 9789 of Opera 9.50b.

Re:Firefox

[Jarjarthejedi]

CURSE YOU!

I'm getting a 49/100 in my FF :(. I wonder what that last point is since the only extensions I have running that should interfere are security related ones (as in, don't abuse my security ones which shouldn't be used by Acid3). Perhaps it's a user-setting somewhere in the about:config or something...

Re:Firefox

[littlegothlost]

For the record, Firefox 3beta3 scores 58/100

Re:Firefox

[FF0000 Phoenix]

This discrepancy seems to be happening with a lot of browsers. If you click on the text "ACID 3" itself [in OmniWeb at least], a window pops up and gives a list of the tests that failed and how, such as

        Test 25: failed to raise exception

It is probably more useful to compare these details to find out the discrepancies really are.

Re:Firefox

[Nimey]

Gods know with all the misleading or wrong summaries and headlines here, people shouldn't believe a word until they RTFA.

Yeah, I know, but I've been here longer than most of you.

Re:Firefox

[caerwyn]

Good point. Running now gets me a 90 with the webkit nightly, and examining the earlier failure shows at least one test (69) which explicitly states may be a networking issue- something that seems highly likely given how overloaded the acid3 server was earlier in the day.

Re:Firefox

[dwater]

> I'm getting a 49/100 in my FF :(.

Ha, even Netscape Navigator 9 gets 50/100, and that's not even a supported product any more!

Re:Firefox

[carlmenezes]

hehe...proves the nightly gets better i guess...66/100 here with grayscale squares :)
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9b4) Gecko/2008030317 Firefox/3.0b4

Re:Firefox

[rawg]

I'm getting 66/100 in Minefield.

Re:Firefox

[nneonneo]

3.0.4 is fairly old. Get the latest WebKit nightly (http://nightly.webkit.org) and you will see it do very well. The nightly seems to be increasing its score by one each day, as far as I've observed, so by next week, it will have passed all the tests if it keeps up. Now, if they could only get ruby text to work - properly...

Re:Firefox

[SEE]

I've seen a statement that Acid3 (and Acid2 before it) is deliberately designed so that no browser available at the time of release can render it correctly, specifically so that it serves as a challenge to all browsers and to avoid the appearance of playing favorites.

Re:Firefox

[dagamer34]

I'd say that the latest Safari nightly is favored pretty heavily coming in at 90/100.

Re:Firefox

[John Whitley]

With Firefox 3.0 beta 3, I upped that to 59/100 by turning off AdBlock Plus on the test page.

Re:Firefox

[99BottlesOfBeerInMyF]

Umm what did Firefox get on this? What about Opera? If you are going to report something why not report all the facts.

Well, it seems most people are too lazy to run the numbers. I wrote a script to run the sunspider javascript test, so here's Acid3 as well:

Mac OS X 10.5.2

• Safari 3.0.4 - 39/100
• Safari 3.0.4 with Webkit nightly - 86/100
• Opera 9.5 beta - 59/100
• Firefox 2.0.0.12 - 50/100
• Firefox 3.0 beta - 67/100
• Camino 1.5.5 - 50/100
• icab - 4.0.1 - 39/100
• Shira 2.0 - 26/100 then crashes

Kubuntu 6.0

• Konquerer 3.5.2 - crashes
• Konquerer 3.5.2 with Webkit nightly- crashes
• Firefox 1.5.0 - 50/100
• Firefox 3.0 beta - 67/100
• Opera 9.5 beta - crashes
• Epiphany 2.14.3 - 50/100

Windows XP SP2

• Internet Explorer 7.0.5730.11 - 11/100
• Firefox 2.0.0.12 - 50/100
• Firefox 3.0 beta - 59/100
• Safari 3.0.4 - 39/100
• Safari 3.0.4 with nightly Webkit - crashes
• Opera 9.26 - 46/100
• Opera 9.5 beat - 49/100 then crashes

Re:Firefox

[99BottlesOfBeerInMyF]

Oops, "Opera 9.5 beat - 49/100 then crashes" should obviously be "Opera 9.5 beta 49/100 then crashes."

Re:Firefox

[zerocool^]

Besides, I see these as a process or goal -- giving the browser makers something concrete and visual to shoot for, as well as an easy way for users to judge the quality of their browser of choice.

Bullshit. The Acid tests have become the SAT's of the browser world. People *think* it's a measure of how standards compliant your browser is, but all it *really* is, is a measure of how well your browser does on the acid test.

That's it, nothing more, nothing less. The acid test is incredibly nit-picky and it's possible to even argue some of the decisions the guy has made about how things should be rendered (i.e. questioning his interpretation of the standard). And it's important that if your browser fails the acid test, but looks fine when you surf fark/slashdot/cnn/myspace, then whothefuck cares? The browser exists to deliver content to you; not to make some jackass feel happy that his CSS and Ajax code is the hardest thing to render in the known fucking universe.

I really, really wish people would get over the Acid tests; perhaps in favor of "the CNN test", or the "does it work with my proprietary intranet badly-coded webapps?" test. If it passes these, then just roll with it.

~Wx

Re:Firefox

[wheany]

Then again, if all the browsers perform the same according to any arbitary test, you can trust those features working the same way in your proprietary webapp.

Re:Firefox

[Bogtha]

I really, really wish people would get over the Acid tests; perhaps in favor of "the CNN test", or the "does it work with my proprietary intranet badly-coded webapps?" test.

But those kinds of "tests" only test new browsers. The developers behind CNN's website and all those proprietary intranet applications bodge their code to make them work in existing browsers already, so nothing in existing browsers is being tested. At most, you could say that they test the popularity of the browser.

Sure, the Acid tests aren't particularly interesting to users, but in case you haven't noticed, developers are pretty important to the web too, and the Acid tests are highly relevant to them.

Re:Firefox

[CrazedWalrus]

Bullshit nothing. The problem is and has been two-fold:

1. "Standards compliance" was a stick to beat browser makers with, but it was an unquantifiable stick. There was never any metric to say *how* un-compliant a given browser was. ACID tests, for all their imperfections, are a tangible meter stick. This stick is otherwise known as a "benchmark". Benchmarks always have problems, but we've got to agree on one, and then strive to make it the best benchmark possible.

2. Very little motivation on the part of some browser makers. With a concrete benchmark to point to which can give a relative idea of where the browser sits in the scheme of things, developers can be pressed to improve their products due to industry pressure to meet the benchmark.

If you think benchmarks are ever perfect, read /. when they review a new GPU or CPU, and see how people think the benchmark was biased toward AMD or nVidia, or database benchmarks where people say they didn't tune the thing correctly, so Postgres should have beat MySQL.

As for the "CNN Test", now you're talking about a biased test. CNN Developers are going to develop their site to look best in the most popular web browser. That means the "CNN Test" will heavily favor IE, and all other browsers will fail to some degree simply because they're not IE. You're essentially making IE the meter stick, and that's setting a pretty low bar. The ACID benchmarks attempt to set a higher bar by coding to the standards, and not give a damn which browsers can actually render them.

Re:Firefox

[Anonymous Coward]

How do you get a +5 informative by copying the link from TFS?

Re:Firefox

[Red Alastor]

You'd get 51 if you disabled adblock.

Firefox 2.0.0.12

[poetmatt]

I get a 51/100 with firefox 2....wonder how 3 will do.

Re:Firefox 2.0.0.12

[The MAZZTer]

3b3 gets a 61. Opera 9.5 is the best I tested at 65. Safarai 3.0.4 for Windows got a 39. IE7 got a 12 and also managed to mangle the page the most.

Re:Firefox 2.0.0.12

[elFarto the 2nd]

58/100 with Firefox 3 beta 3.

Re:Firefox 2.0.0.12

[thornomad]

Woo hoo! IE 6.0 displays this just fine: http://acid3.acidtests.org/reference.html Read 'em and weep Firefox!

Re:Firefox 2.0.0.12

[imess]

58/100
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b3) Gecko/2008020514 Firefox/3.0b3

Re:Firefox 2.0.0.12

[Jugalator]

Opera 9.5 is the best I tested at 65. It better be good, since Håkon Wium Lie, Chief Technical Officer of Opera Software, worked together with Bert Bos to develop the CSS standard.

I'm not sure how many actually knows this. *shrug*

Re:Firefox 2.0.0.12

[Jeffrey Baker]

That's interesting. I wonder why the Linux build of Firefox 3 Beta 3 scores 61 while the Windows build scores 58.

Re:Firefox 2.0.0.12

[diegocgteleline.es]

Firefox 3 beta 4 gets 66. I've heard that webkit snapshots get even more.

Re:Firefox 2.0.0.12

[ne0n]

I just got a 0/100 in Firefox 2.0.0.12. How did you rig yours to get so far? I even disabled NoScript completely for this one..

Re:Firefox 2.0.0.12

[ne0n]

funny thing, I closed the tab, opened the same address and got a 50/100. We even have a cat on the page now.

Re:Firefox 2.0.0.12

[poetmatt]

With all the details I can surmise:

Firefox with the following addons:

Noscript
NukeAnything Enhanced
Restarter
ShowIP
IE Tab
Gmail Manager
Ebay Toolbar (the one those college kids made)
DOM inspector
All In One Gestures
Adblock

I as well had to let noscript through on the site and then refresh.

options :

advanced tab:
advanced encryption - SSL3 and TLS 1 checked
security tab: warn about addons
site forgery - using downloaded list
remember passwords (my own preference ofc, I only use it for ones I can afford to compromise)

Only warning on: when viewing a page with low grade encryption

content tab: all 4 boxes checked, all advanced things checked under enable JS

and that it I think. I don't think I've changed the default that much on everything.

Oddly, at home on this config I cannot install addons, I get that undefined error shit. So no idea what the diff is, but this one runs a-okay.

Re:Firefox 2.0.0.12

[emilper]

unless people from Mozilla or Safari or IE were forbidden to participate, I think it's a good thing that Opera had somebody in the team that developed the standard, and maybe the others should have sent somebody, too.

Re:Firefox 2.0.0.12

[kamikaez]

This is a javascript / DOM test, not CSS.
It was developed by Netscape, so I guess they will do best.. Oh well..

Re:Firefox 2.0.0.12

[Petrushka]

Just for interest: IE6 gets 4. That makes me feel all warm and fuzzy.

Re:Firefox 2.0.0.12

[thc4k]

yeah i got 65 bith the latest snapshop of opera, but it had one huge lag at 54, i thought it had crashed

Re:Firefox 2.0.0.12

[monkeyman_67156]

The new 3b4 release candidate gets 66. Take that Opera! ;)

Re:Firefox 2.0.0.12

[Quarters]

It better be good, since Håkon Wium Lie, Chief Technical Officer of Opera Software, worked together with Bert Bos to develop the CSS standard. I'm not sure how many actually knows this. *shrug*

About one more than actually care, to be honest.

Re:Firefox 2.0.0.12

[nneonneo]

For IE7, the trick is to wait about 30 seconds to a minute. Then, it will start counting again.

Presumably, this is because it fails a ton of tests in the middle, so it *looks* like it stopped counting.

It eventually gets to 12.

Re:Firefox 2.0.0.12

[99BottlesOfBeerInMyF]

3b3 gets a 61. Opera 9.5 is the best I tested at 65. Safarai 3.0.4 for Windows got a 39. IE7 got a 12 and also managed to mangle the page the most.

Your numbers are quite different than mine. I scripted all the browsers/OS's I had handy from the sunspider javascript test last week and ran them on Acid3. The results are here.

Where do you get a nightly of Opera? I ran the beta version they have up tonight and got 59/100 on OS X and it crashed on Linux and Windows XP. In any case, the best number I got was Safari 3.0.4 with a week old nightly of Webkit on OS X, which got 86/100. The Firefox 3 beta also did well getting 67/100 on OS X and Linux (but only 59 on Windows for some reason). Other people have gotten slightly better numbers for both using a more recent nightly of Firefox or Webkit.

Re:Firefox 2.0.0.12

[99BottlesOfBeerInMyF]

58/100 with Firefox 3 beta 3.

I get 59 on Windows and 67 on OS X and Linux. Some people report a bit better with the nightly's.

Re:Firefox 2.0.0.12

[Anonymous Coward]

Err... wha?

Acid 3 is mostly Hixie, who's currently at... Google, I think. Previously Opera/Netscape.

Aside: yes, he did ask for browser developer feedback (in particular, to look for things that will fail in under-development versions of Mozilla and Safari; presumably, Opera too.) Apparently new things that only failed in IE wasn't interesting. Reference: his blog.

Re:Firefox 2.0.0.12

[Parrais]

Interestingly, in IE6, if I click the link to go to the reference rendering, then hit the back button, the score increases to 5.

Is that a slight improvement, or damning evidence that IE6 can't even render a page the same twice?

Firefox

[grim4593]

I got a 50/100 at http://acid3.acidtests.org/

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.12) Gecko/20080207 Ubuntu/7.10 (gutsy) Firefox/2.0.0.12

Link to the actual test

[I kan Spl]

Why does slashdot keep linking to dead blogs?

The actual test is http://acid3.acidtests.org/ here.

Re:Link to the actual test

[The Ancients]

Slashdot linked to a well known blog because they had screenshots of several major browsers to save the readers some time, or show results for those who don't have access to other browsers.

I visit drunkenfist quite often, and this is the first time they've been down. Somehow I think the two are related

Re:Link to the actual test

[jez9999]

For some reason, acid2. and acid3. respond, but acid1. times out. *shrug*

Re:Link to the actual test

[Kyokushi]

They didnt link to the actual test, and it was slashdotted already. Putting it on top will just create additional burden for those poor guys!

Nothing like including a link to the actual test..

[inotocracy]

.. http://acid3.acidtests.org/

Re:Of Course IE will fail, ACID test is biased...

[setagllib]

Yes, biased towards conforming with open international published standards, rather than to any specific vendor's implementation. It just happens the best of the best web browsers try to conform to the same standards, scoring much higher than Microsoft's offering which is deliberately designed to break from the standard to ensure lockin.

Link to the actual test

[robkar]

http://acid3.acidtests.org/

I'd read TFA but...

[Fozzyuw]

My browser won't render the page properly.

Latest Safari nightly scores...

[The Ancients]

90/100.

Getting pretty close.

Re:Latest Safari nightly scores...

[Jeffrey Baker]

Not bad. Unfortunately I couldn't get the webkit nightly to load the page, probably for unrelated reasons. The latest trunk build of Firefox gets 66, which is an improvement from 61 in Firefox 3 Beta 3.

Re:Latest Safari nightly scores...

[digitalcowboy]

I got it to load twice in the latest nightly of WebKit just now and scored 85/100 both times. That's pretty impressive, in my opinion and one of the reasons I use WebKit as my primary browser these days.

Unfortunately, the server is still timing out when I try to view the reference image.

Re:Latest Safari nightly scores...

[kozmico]

IE 8 Beta 1 gets 17/100. Download of IE Beta 1: http://www.microsoft.com/windows/products/winfamily/ie/ie8/readiness/Install.htm

Re:Latest Safari nightly scores...

[Bogtha]

I wonder if there will be a release version of Safari that passes Acid3 before a release version of Firefox passes Acid2?

Re:Opera

[wile_e_wonka]

I just tried it on Opera 9.5 Beta, build 9755. I got a 60/100. Then I tried again and got a 61/100. Then a 60/100 on a third try.

All of the rectangles are grey (two different shades), the test name is red and does not have a shadow, and there is an x in the upper right hand corner.

Re:Of Course IE will fail, ACID test is biased...

[moderatorrater]

Not quite. When none of the browsers are getting 100/100 and the only browser to get over a 60 is a safari beta, I think it's safe to say that it's a test designed so that every browser will fail. That's the point: they're giving solid targets to browser developers and giving a concise score to everyone else so that they know where the browsers stand in the next generation of web tech.

So, I guess what I'm saying is that complaining about it being designed so that IE would fail is like saying that American Gladiators was designed so that my 8 year old brother would fail. Sure, it has that effect in the end, but the fact that he's under-equipped for such a competition isn't American Gladiators' fault.

Failure

[mrbah]

The Acid tests would be a lot more productive if they were oriented more towards the practical non-compliance issues than obscure ones. A back-asswards JavaScript implementation or a horrible box model is more of an issue than the inability to display base64 images encoded directly into the page markup. Total compliance is great, but it's much more pragmatic to get the fundamental issues fixed first.

Re:Failure

[Qzukk]

than the inability to display base64 images encoded directly into the page markup

Trust me, that one could actually become fairly important, depending on how large the image could be. I've got scripts that produce reports where right now it takes several seconds of database grinding just to give me the table of data and then several seconds again when the browser hits the script in the <img> tag to get the graph, if I could calculate the table once and produce the graph at the same time and insert the graph into the HTML itself, it reduces the wait time to get the report.

Re:Failure

[prockcore]

You can. Use the canvas tag.

Re:Failure

[BZ]

The Acid3 test was trying to test things that:

1) Are actually fully specified somewhere as of a few years ago
2) Are failed by at least one of Safari, Opera, and Mozilla.

Turns out you have to really scrape the bottom of the barrel to find those. The more practical things that are actually covered by specs there's decent compat on.

Re:Geek version of a measuring contest?

[gerbalblaste]

No this is a measurement of compliance to international standards.

Firefox 3 beta 3

[corychristison]

57/100 :-( I really was expecting more.

Re:Firefox 3 beta 3

[corychristison]

... and 64/100 for Firefox 3.0b5pre ("Minefield")

Re:Firefox 3 beta 3

[BZ]

Why? The test expressly picked things that one of Opera, Safari, and Firefox would fail, preferably more than one, and tried to balance the number of tests each would fail.

Put another way it looked really hard for things to test that would give browsers low scores.

There's nothing to say that the things it tests are necessarily useful. Some are, some are not.

Konqueror

[kevmatic]

I haven't seen anybody answer konqueror yet!

I tried it in Konqi 3.5.8 with Gentoo. It asked me what I wanted to do with "empty.txt" then segfaulted. Anyone fairing better?

Re:Konqueror

[Anonymous Coward]

That crash is fixed in 3.5.9 (which does 41)... 4.0.2 does 61, 4.1-pre 63. Early 4.0.x versions do less.

Re:Konqueror

[Prefader]

3.5.6 in Kubuntu 7.04 crashes after loading up the page. I don't get any questions about "empty.txt".

I don't know if I'd call that "fairing better", but at least it's different.

Re:Konqueror

[Caesar Tjalbo]

Little better. 3.5.8 gave 41/100 but didn't look at all like the reference and 3.5.9 had 51/100, but looked identical. I had to copy the score from behind a picture of a dog (?) to even see it. I got the 'empty.txt' question and answered to open it; it didn't but also didn't segfault.

FWIW: sidux 2.6.24.2 amd64, KDE 3.5.8 & 3.5.9, Qt 3.3.8b

Re:Geek version of a measuring contest?

[Poromenos1]

No, it's measuring both dinki with the same ruler and saying "You must be this long to procreate" (not to scale).

Re:Of Course IE will fail, ACID test is biased...

[Ambiguous Coward]

Comment on your analogy:

I think the mere fact that American Gladiators is considered tv-worthy indicates that we, as a nation, have failed. Also, sorry about your brother. That was really brutal when they knocked him into the pool.

Just sayin'.

-G

Slashdoted

[fluch]

Conclusion: ACID3 test didn't pass Slashdot test. Too bad.

Re:Slashdoted

[The Ancients]

It's ok here. I've even scripted 700 machines here to reload the page repeatedly just to ensure it isn't.

Re:Slashdoted

[meiao]

It hasn't been slashdotted yet.

I tested a few browsers

[nxsty]

I tried the Acid 3 test in Firefox 2.0.0.12 , Firefox 3 (recent nightly build), IE7, Opera 9.26 and Safari 3 (windows beta). The best was Firefox 3 which almost rendered the page correctly. The worst was IE7 (as expected). Safari was probably a little bit better than Opera and Firefox 2 but it's hard to tell.

Re:I tested a few browsers

[99BottlesOfBeerInMyF]

I tried the Acid 3 test in Firefox 2.0.0.12 , Firefox 3 (recent nightly build), IE7, Opera 9.26 and Safari 3 (windows beta). The best was Firefox 3 which almost rendered the page correctly. The worst was IE7 (as expected). Safari was probably a little bit better than Opera and Firefox 2 but it's hard to tell.

I ran the tests on a few more browsers and OS's. The results are here. The upshot is Safari+a recent Webkit on OS X is in the lead, followed by Firefox 3 on OS X or Linux.

Re:Geek version of a measuring contest?

[RzUpAnmsCwrds]

CSS3 isn't an "international standard", it's a draft specification.

My results of Acid 3 test

[TimSSG]

On Windows XP SP2, ACID 3 test results 57/100 for FireFox 3 Beta 3 I am guessing 17/100 on IE 8 Beta; it was very hard to see the top of the 1 and the 7, but the print preview should a 17 on it. Tim S

Re:Of Course IE will fail, ACID test is biased...

[dgatwood]

Right now, no browser can make it to 100 even if somebody had everything working. The servers appear to be falling over and timing out returning content.

Results in major browsers

[Lobais]

See http://browsershots.org/http://acid3.acidtests.org/ for the test in 75 different browsers.
Wikipedia: http://en.wikipedia.org/wiki/Acid3 also lists the results for the developversions of browsers:
Webkit: 87
Firefox: 67

Firefox

[zulater]

I got a 100/100 on the reference image.

Re:Of Course IE will fail, ACID test is biased...

[moderatorrater]

no browser can make it to 100 even if somebody had everything working With an attitude like that I don't expect any browser to ever make it ;)

IE8 Cheats ACID2!!

[feld]

IE8 doesn't pass Acid2! I think it cheats!

Check it out quickly guys!

http://www.webstandards.org/action/acid2/ PASS

http://acid2.acidtests.org/ FAIL

The only thing different between these tests is a 404 link on about line 130 of the source. Is IE8 cheating?!!!

Re:IE8 Cheats ACID2!!

[Anonymous Coward]

The eyes for the test at acird2.acirdtests.org is cross-domain, which is causing it to fail because of boundary trusts.

Re:IE8 Cheats ACID2!!

[prockcore]

No. The test got screwed up. Everything fails that second one. Including FF3 and Safari 3.

Re:IE8 Cheats ACID2!!

[Anonymous Coward]

This was already explained, and the behavior should have been updated as per a previous slashdot story.

IE8 fails when it runs in quirks mode, and passes when in standards mode. Before it would run in quirks by default, and only change behavior when it visited certain key sites, or sites had a tag.

That url would be one of those "key sites"

However if the previous slashdot story is true, IE8 should eventually operate in standards mode by default, so it will pass both.

Re:IE8 Cheats ACID2!!

[ejtttje]

Both work for me with Safari 3...

Re:IE8 Cheats ACID2!!

[Bodero]

Why Isn't IE8 Passing ACID2?

Basically, it fails because of XSS on the other sites.

Re:IE8 Cheats ACID2!!

[CastrTroy]

Maybe Safari is cheating? If the test is screwed up, and Safari still passes, they must be cheating.

Re:IE8 Cheats ACID2!!

[LingNoi]

FF3 Beta 3 passes both..

Re:IE8 Cheats ACID2!!

[hixie]

The test isn't screwed up, it's a bug in IE8.

Re:IE8 Cheats ACID2!!

[Crayon Kid]

I think you mean "cross site addressing" ie. loading content from several websites on the same page. It is NOT XSS, which means "cross site scripting". No scripting involved here.

All I see (and you can too, if you follow the comments and explanations on that page) is Microsoft doing things awkwardly, as usual. HTML inside OBJECT tags should be no different from HTML inside an IFRAME, yet Microsoft felt the need to use ActiveX and thus raised a security problem. Which they've handled inappropriately, by blocking, instead of falling back to simple rendering.

Good old ActiveX. Still messing things up, two browser generations later.

If one day I'll be coding a website and I find myself having to work around IE8 quirks just as I had to work around IE7 and IE6 quirks... Where's the difference? Hopefully there'll be less quirks to worry about, but at the end of the day I've still wasted time catering to IE whims.

Re:Of Course IE will fail, ACID test is biased...

[gsnedders]

Actually, trying to make IE fail it wasn't an aim: the aim was to include tests that one of Firefox, Safari, and Opera fails. If IE happens to fail them too, so be it.

IE8

[nxsty]

I just installed IE8 beta 1 and while it's better than IE7 it's still worse than the rest.

Thunderbird ;]

[Przemo-c]

Thunderbird 2.0.0.12 with thunderbrowse scores 52/100 ;]

download links for IE8 beta!

[ionix5891]

http://www.microsoft.com/windows/products/winfamily/ie/ie8/readiness/Install.htm

Re:download links for IE8 beta!

[Przemo-c]

http://adam.pra.to/public/pics/omg.jpg

100% score on IE8

[spacemky]

I got a 100% score rendering Acid3 on IE8! All I had to do was add the following line to the top of the page!

<meta http-equiv="X-UA-Compatible" content="IE=8" />

Once that meta tag is there, all web pages look just as they're supposed to! I'm so glad Microsoft finally fixed this whole compatibility fiasco.

Re:100% score on IE8

[Tacvek]

I got a 100% score rendering Acid3 on IE8! All I had to do was add the following line to the top of the page!

<meta http-equiv="X-UA-Compatible" content="IE=8" />

Once that meta tag is there, all web pages look just as they're supposed to! I'm so glad Microsoft finally fixed this whole compatibility fiasco.

The good new is that they have fixed this. They have now decided that the presence of a doctype will Enable the latest "Standards" Rendering mode by default. The older "Standards" modes can be requested using the HTTP header or meta-tag. "Quirks" mode will be activated in the event that no doctype is available. See the blog post http://blogs.msdn.com/ie/archive/2008/03/03/microsoft-s-interoperability-principles-and-ie8.aspx

My guess is that Microsoft figured out that high level of standards compliance can work fine as a default. (Since the more standards complaint browsers have few problems with the most sites.) The oldest sites will still work fine because they would use quirks mode. Newer sites can fix themselves, or use the tag. For intranet applications adding the HTTP header or meta tag can often be added easily enough that it is just not worth hurting the Internet to spare companies the effort of adding the header or tag to their broken internal apps. This leaves only unmaintained sites (which will not get fixed or updated) that claim standards compliance (via a doctype) but actually rely on IE bugs. Those will break, but on the other hand, those sites are almost certainly broken in other browsers too, so no real loss.

Web 2.0?

[Tatsh]

Someone again try to explain to me the definition of web 2.0, and don't tell me flash.

I personally think it's the move of the entire web (the content that matters) to valid XHTML, CSS, etc (of course everything is controlled dynamically by PHP/Perl/whatever you want). I also hope there can be an open standard soon to do the same functionality that Youtube's Flash container that runs on everything and that everyone agrees upon. Silverlight is obviously closed and so is Flash. We need an open source mid-quality (and high-quality) video player that loads quickly and is OS-independent, just like Flash. I think that is all that is missing in this 'Web 2.0'.

Re:Web 2.0?

[Anonymous Coward]

Here you go: the definition of Web 2.0.

Re:Web 2.0?

[Gideon Fubar]

Caesar is all things to all men.

Srsly, tho.. Web2.0 is (supposedly) about cutting out the page refresh in between client actions and content updates. AJAX, SOAP, etc. are obvious examples. Of course, i've heard all types of other descriptions, to the extent that i'm no longer sure of that definition, as inaccessible to the public as the idea was anyway.

As for the movie player, it's not a particularly good idea to write something like that in a serverside-rendered language.. i guess it'd be possible to write one in javascript, but i haven't looked into it.

Re:Web 2.0?

[jonbryce]

Web 2.0 is an ajax powered, mash-up paradigim, leveraging user generated content, rich internet applications and software-as-a-service business models. Web 2.0 will transform the connected experience by providing a social networking experience together with a highly personalised content delivery platform.

Re:Web 2.0?

[lwsimon]

XHTML is not intended to replace HTML. "Moving the web to XHMTL" is a HORRIBLE idea. XHTML, served properly (which it very rarely is) breaks entirely with a single error. XHTML should only be used if you have a specific, quantifiable reason to do so - and it must be served as such. Otherwise, you're only giving the browser funny-looking HTML with a bunch of slashes on the end of tags.

Re:Web 2.0?

[insertwackynamehere]

I like XHTML because it feels cleaner and I write to standards. What other reason would you use XHTML?

Re:Web 2.0?

[Tatsh]

To your post, "What?!" I write in XHTML and I write to standards (I always make sure my pages validate). I think the idea behind XHTML is for other devices besides web browsers to be able to get data from web pages. The Doctype at the top of the document tells the browser how to treat the page, and as you might already know there are thousands of pages that do not even have a doctype specified. I see the future of the web as being parseable, and XHTML allows a lot more than regular HTML where you can forget about tag ordering and even some tag endings (what's the parser to think then?). This is why a lot of programmes cannot yet just "get" data from whatever web site (because it's not easy to parse the page). XML is generally easy for a computer to parse through, and XHTML is just the HTML version of that. I think it's great and future-proof (even W3C says XHTML makes sites "future-proof", meaning ready to parse through easily, I think).

As you can see, my post is all about parsing data. This is SO important. Parsing data can be also through binary means or by guessing (which is how browsers engine's often render HTML), etc. If every page is XHTML, then there will not be any problems parsing data whether it's a browser, or just a programme getting some small bits of data from a page (like say a weather programme getting the temperatures from a weather site by parsing through the page and knowing what to look for).

Not to mention...

[rsborg]

The latest webkit (Safari) nightly is just amazingly fast.

Faster than FF3 beta 4, much much faster than FF2 or IE7.

Perhaps....

[acvh]

....there is a need for people who actually develop browser software to use this. but, of course, what it is really being used for is. "my favorite browser scored X", even when there are maybe six people here who know what's going on in the test. something tells me that there is little correlation between an Acid3 score and how well a browser displays a page.

On my first day of Tests and Measurements the prof asked, "What does an IQ test measure?"

The correct answer, of course, is "how good someone is at taking IQ tests."

Re:Perhaps....

[bunratty]

Yes, there is almost no correlation between how well a browser does on Acid tests and how well it renders pages on the web. The purpose of the Acid tests is to break the chicken-and-egg problem of web development. The web developers tend not to use features unless all popular browsers support them. On the other hand, the developers of the web browsers tend not to add features that are not used by web developers. Without anyone willing to go first, the implementation and use of new web standards stalls.

The purpose of the Acid tests is to break this logjam by using these new standards in a very public way so that web developers will be motivated to implement them. The "my browser does better than your browser" posturing is a bit immature, but as a side effect it popularizes the faults of browsers and motivates the browser developers to fix them. Then, the web developers use the new features after they are well supported.

Re:Perhaps....

[ivan256]

The ideal here actually is that if a reasonable number of mainstream browsers scored 100 on the test, web developers could use all of the features the test exercises and have a reasonable expectation that their page will display correctly for end users.

The test is about making life better for web developers, and about making the web more interoperable, instead of having sites which jump through browser predicated hoops, or restrict users to "IE7.0 or newer on 32-bit Windows" or the like. Thus having your favorite browser, and your least favorite browser score well is in the best interests of all web users.

Re:Perhaps....

[HeroreV]

It seems to me that the purpose is just to be fun, to draw attention to web standards. There are several test suites that are much better at testing this stuff, but not many casual users are going to see how well their browser does in those. Browser implementers are more likely to pay attention to standards if many of their users are looking at a test of those standards.

regular test suites -> not fun
Acid tests -> fun

That's the difference.

Re:Of Course IE will fail, ACID test is biased...

[Goaway]

Actually, it's tests that one, or preferably more than one, of Firefox, Safari, Opera or IE fail.

Unfair browser bashing?

[Sebastian Reichelt]

While I welcome tests for standards compatibility, such test scores are not really a solid ground for browser comparison. If they wanted an unbiased indicator, they would need to publish compliance tests before parts of the specification are implemented in browsers, not after the browsers have been released to the public. This way I strongly suspect they are actually testing against specific browser flaws, which means they can design the test so each browser achieves whatever score they want. I'm not saying the scores are wrong; I think they are simply not meaningful.

BTW, I think we would be better off if ECMAScript wasn't part of the "Web standards" at all. Most of the time, it is used in places where that is completely unnecessary. As a client-side scripting language, browser incompatibilities are no surprise. Some browsers don't support it, and some people would like to turn it off.

Consider this: Almost all websites are used to deliver text, images, and downloads (i.e., some sort of content), and some have forms for interaction. No browser seriously has a problem with any of that. They may not make the pages look the way they are supposed to, but that is secondary. On the other hand, websites that depend on client-side scripting are effectively hiding their content and making it available only to specific browsers. That's why ECMAScript is harmful to the web.

Re:Unfair browser bashing?

[bunratty]

You're exactly correct that the Acid tests test specific browser flaws. They are testing exactly the flaws that plague web developers. That way, when all popular browsers pass the Acid tests, web developers don't need to work around the flaws in each different browser. We all benefit by getting web sites with fancy new features that work in all browsers. The scores are not meaningful, but are a way to motivate the developers of web browsers to fix their flaws so they're not embarrassed by a low or non-passing score.

Re:Unfair browser bashing?

[BZ]

Really? All the issues Acid3 tests plague web developers? Ranges inside Attr nodes, for crying out loud?

Re:Unfair browser bashing?

[bunratty]

Yeah, for every item tested by Acid3, I am sure there exists a web developer who would have benefitted from the issue being fixed. If no web developers wanted a feature, it would not have been written into a web standard. If no web developers ran into a bug, they would not have known to test for it.

The bottom line is that the OP is correct. The scores are not a meaningful indicator of "goodness" of a browser. They simply indicate how many tests each browser passed. The purpose of the test is not to make browsers look good or bad, but to ensure web browsers support the features requested by web developers for Web 2.0. As the description for the Acid3 test states, Acid3 is primarily testing specifications for "Web 2.0" dynamic Web applications.

Re:Unfair browser bashing?

[BZ]

> If no web developers wanted a feature, it would not have been
> written into a web standard.

This is quite false for the DOM. There were many organizations involved with it that had nothing to do with the web, and large parts of the DOM are aimed at generic XML workflow tools and pretty useless for web developers.

> If no web developers ran into a bug, they would not have known to
> test for it.

It's easy to generate some sorts of tests using a random-number generator, actually...

> The bottom line is that the OP is correct.

That the scores don't mean much? Yep.

> but to ensure web browsers support the features requested by web
> developers

Nope. Some of the stuff covered by Acid3 falls into this category, but some is just there to pad the total number of tests out to 100. It took a while to find those last few dozen tests to pad out the number after the ones that cover things people really care about had been written...

> As the description for the Acid3 test states,

Yes, Acid3 tests aspects of the DOM. That doesn't mean all aspects of the DOM are useful. A lot of it really isn't.

Re:Unfair browser bashing?

[bunratty]

It's easy to generate some sorts of tests using a random-number generator, actually...

Yes, fuzz testing can be very useful at finding obscure bugs that no web developer would knowingly trigger. They're useful for finding edge cases where extreme bugs such as crashes and security problems can occur. That's not what Acid3 is testing, however.

Acid3 is pointing out deficiencies in current browsers, as the OP conjectured. But it is not pointing out those deficiencies for unfair or useless browser bashing. It is pointing the deficiencies out that make it harder for web developers to develop Web 2.0 applications.

Re:Unfair browser bashing?

[BZ]

> That's not what Acid3 is testing, however.

Are you sure? ;)

> But it is not pointing out those deficiencies for unfair or useless
> browser bashing.

No. It's pointing out some for good reason and some just to get the test count to 100.

> It is pointing the deficiencies out that make it harder for web
> developers to develop Web 2.0 applications.

You never addressed my concrete example: How does the interaction of ranges with the text node inside Attr nodes (which is explicitly tested by Acid3) affect Web 2.0 applications? What Web 2.0 applications would even want the performance penalty of having Attr nodes synthesized? Note that Acid3 will give you a pass on that test if you don't implement Attr nodes at all; it'll only fail if you implement them but don't allow range operations on their textnode. It's really pretty closely targeted to fail one particular UA.

firefox 3b5

[narfman0]

...Gets a 66/100

Re:Geek version of a measuring contest?

[Anonymous Coward]

CSS3 isn't an "international standard", it's a draft specification

Very true, but not quite the killer point you probably intended it to be; Acid3 does test some of the more widely implemented CSS3 features, but it's primarily concerned with JavaScript and the DOM, not with CSS at all.

Re:Geek version of a measuring contest?

[misleb]

It isn't Firefox vs. IE. It is IE vs. All other browsers that seem to be reasonably consistent with each other.

You shouldn't have.

[Millennium]

The Acid tests are not made to pat browser vendors on the back. They exist to show areas in the standards that still aren't covered well. Browsers should fail when the test is first released. that shows that the people making the tests have done their job right by covering areas that are still spotty when it comes to implementation.

Of course, there's the fact that the browser which made Web standards famous will be the very last non-IE browser to pass Acid2, despite having had over two years to fix the issues, when the iCab guy managed it in less than a month. But I suspect that the answers there are more political than technical.

How do the acid-test creators test the acid test?

[dskoll]

Seriously... if no browser gets 100/100, how do the test creators generate the reference image? And how do they know there are no bugs in their test? I'm genuinely curious...

What would be really useful....

[CodeShark]

Is if there was a way to not only get a copy of the acid test fails, but a quick list of which browsers fail which test and what that should mean. So that us legions of OS coders or even Mozilla, Opera, or Safari's own guys could get busy and fix it in their next releases.

Anyone have this or know some web location where it's happening?

Re:What would be really useful....

[The Ancients]

For Safari, buglist here, google docs spreadsheet here

Re:What would be really useful....

[RiotingPacifist]

click on the A

Re:Web 2.0? - My definition

[businessnerd]

Here's my definition of Web 2.0 (NOTE: Your definition may vary)

I refer to Web 2.0 as user generated content. I define user generated content pretty broadly as well. This isn't just YouTube videos and MySpace pages. This is the ability of the user to use the web to dynamically create content on the web, whether it's uploading a video to a public site, creating their own web profile, or typing a research paper using a browser based word processor. The user has the ability to go to a web site and then manipulate the data on the page dynamically to suit their needs. Things like Google Maps (and others) come to mind as well. In the past, you looked at static maps and clicked links to go to other maps. These tied together into different "views". But now, you manipulate a single map by dragging, zooming, placing "push pins", and changing the map overlay (satelite vs. non-satelite). Overall, I see the difference between Web 1.0 and 2.0 as the emergence of a web sites as applications. AJAX is often associated with Web 2.0, as it, as well as other new techniques, is what helps make Web 2.0 possible.

The term Web 2.0 is a pretty ambiguous term, though, so others may have a different definition. Early in the Web 2.0 hype, my mother asked me "I've been hearing about Web 2.0 this and Web 2.0 that, but what the hell is it? No one seems to be able to explain this to me!". To which I replied, "Web 2.0 is a bullshit buzzword made up to describe everything new that is happening on the web. It is mostly meaningless marketing speak. Treat it as such." I was right at the time. It was such a blanket term. But looking back, I think we can better say what it is, because we can now see what all of the things classified as Web 2.0 have in common.

Epiphany doesn't even come close

[r_jensen11]

Maybe there should be a slashdot3 test. First client is acid3

I scored 3/100 here.

Using Epiphany 2.20.3

Re:Epiphany doesn't even come close

[pizzach]

Actually, Epiphany should have botten 50/100 as it uses the same rendering engine as Firefox 2. Plus, I get 50/100 when I go to the page with 2.20.3. Maybe you stop the script before it finishes, thus lowering your score?

Re:Epiphany doesn't even come close

[FF0000 Phoenix]

I still don't have a score yet; I'm trying to work this one out by hand.

Re:Epiphany doesn't even come close

[r_jensen11]

I didn't stop it. I guess it simply timed out. I guess, therefore, that the servers that host acid3 only got a 3/100 on their slashdot test. Just running the test now gave me 50/100.

Re:Epiphany doesn't even come close

[pizzach]

Hm. The page being slashdotted is not out of the question. Also with how the page does all the tests all one by one over a relatively long period of time weird things are probably much more likely to happen.

My computer and connection are slow. So Acid3 causes Epiphany to ask me if I want to stop a unresponsive script every time I run it at around test 3 I think. It even happens occasionally on slashdot's scripts.

Re:Epiphany doesn't even come close

[99BottlesOfBeerInMyF]

I scored 3/100 here. Using Epiphany 2.20.3

What OS? Under Kubuntu 6.0 and Epiphany 2.14.3 I got a score of 50/100.

By the way, if you're tired of hunting through posts to find a given OS/browser, I ran a bunch with a script and posted it here.

38 Safari 3.0.4, 84 Safari Webkit Build 30790

[Wingsy]

38 Safari 3.0.4, 84 Safari Webkit Build 30790

Re:38 Safari 3.0.4, 84 Safari Webkit Build 30790

[appleguru]

I'm scoring 90 with the latest webkit build (30790) (in leopard). I was getting 87 with 30726.

Re:Web 2.0? - My definition

[Tatsh]

"Web 2.0 is a bullshit buzzword made up to describe everything new that is happening on the web. It is mostly meaningless marketing speak. Treat it as such."

I agree, and thought this ever since I heard the term, so I hereby propose abolishment of the term.

83 on the latest WebKit build for OS X

[Anonymous Coward]

All I know is I DRINK YOUR MILKSHAKE - I DRINK IT UP!

Seriously, I'm not at all certain why the score is so high relative to Safari 3. Curious ...

ff3 improved since beta3

[hdparm]

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9b4pre) Gecko/2008030204 nightly build scores 65/100 here

Re:How do the acid-test creators test the acid tes

[bkaul01]

If the test returns javascript errors and confirmations as we're told it does, they should be able to compare those results against the standards for what should have happened. It's a bit more tedious than having a reference browser, perhaps, but it would work.

I've always wondered

[Trogre]

How do they test these Acid tests to know they're working properly?

Re:I've always wondered

[Arimus]

They've got their own browser which is 100% standards compliant. However it can not be used for normal web browsing as it will make most web pages in existence today look a right dogs dinner as their browser is strict to say the least about how it renders pages, which means all the kludges for IE and every other browser will seriously break their browser.

Re:I've always wondered

[Kyokushi]

cant they release the source on that? it should massively help in fixing the other browsers....

Re:I've always wondered

[arrenlex]

[[citation needed]]

Re:I've always wondered

[Anonymous Coward]

You sir were trolled.

There is no 100% standards compliant browser. They make the reference rendering using pain staking brain power, and double checking the standards each and every time!

Safari...

[simpl3x]

gets a 39. Not great for the latest and greatest!

Camino comes in at 50, the same as Firefox, and Opera hits 46.

Ick!

Re:Safari...

[DavidinAla]

Actually, Safari 3.0.4 gets an 87. I just ran it.

Re:Safari...

[99BottlesOfBeerInMyF]

Actually, Safari 3.0.4 gets an 87. I just ran it.

Safari 3.0.4 for both Windows and OS X, get a score of 39/100.

Safari 3.0.4 for OS X with the nightly drop of Webkit (not stable) gets 86/100 (well it is a week old, I'm too lazy to grab the most recent one).

Safari 3.0.4 for Windows with the nightly drop of Webkit crashes on the test.

Re:How do the acid-test creators test the acid tes

[kryten_nl]

Magic

Just found it...

[CodeShark]

Shift click on the A in Acid, it gives you a list. My latest Firefox (2.0.0.12) got a 50, the latest Firefox 3.0 beta build did better Results momentarily.

Re:Geek version of a measuring contest?

[Spy der Mann]

CSS3 isn't an "international standard"

... yet. See, this is why IE fails. IE developers have refused to conform to under-construction standards, but when the standards are finally released (most of the times unchanged from the working drafts), oh too late! We already released the browser! Maybe the next version, too bad.

But we have to give credit to the IE8 team, I'm sure they're already beginning to work out the failures on CSS3. I just hope they're aiming for full standards compatibility and not just "standards covered in acid3".

Just found it for Firefox 3.0 Beta (3)

[CodeShark]

Passes 59 tests in Xp. Interesting Stats:

• 37 of the tests that fail in FF 2.X are also fails in FF 3.0 beta.
• Three of the fails are significantly different, not sure if this means progress or not
• 1 fail is a minor difference
• Firefox 2 passed a test (#69) that FF 3 did not
• and finally, FF 3.0 passes 8 tests directly that FF 2.0 does not.

• • 
    That said, I looked at a couple of the notes on Bugzilla for Firefox and they are already looking at the bug list... wonder who will be the fastest to fix the most....

quick testing

[reemax]

firefox 3.0b3 - 57
konqueror kde 4.0.1 - 60
opera 9.50 Beta 2 - 61

Re:Of Course IE will fail, ACID test is biased...

[djcapelis]

I believe the standard for the last 16 tests were webkit and firefox trunk must fail.

So IE or Opera failing was actually regarded as insufficient.

Re:How do the acid-test creators test the acid tes

[djcapelis]

The reference implementation is Ian Hickson's brain.

IE is a de-facto standard, and THAT's the problem.

[Spy der Mann]

Of course the Microsoft software will fail these tests as that is what ACID tests are designed for...

After thinking for a while, I had realized a really awful thought that has creeped into developers' minds. They ALWAYS design for two or more browsers!

if( ... ie ... ) {
} else { // Firefox and compliant browsers }

And that's just WRONG.

Of course IE fails, it doesn't implement the standard event handling methods in javascript. They don't care about "how things should be done", they only release THEIR specs on their webpage and say "we do things this way", and this is why we end up writing if's or adding a compatibility layer in our javascripts.

We have grown so accustomed to this problem that we take it for granted and thing "oh, that's just how things are". Hello!? there shouldn't be any if's in any webpage code... This is why standards were created in the first place!

Firefox 3 Nightly

[(eternal_software)]

Running the latest Firefox 3 nightly results in 66/100.

Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9b5pre) Gecko/2008030506 Minefield/3.0b5pre

Mobile Safari (iPhone)

[rsborg]

Gets a 38/100 for me.
Which is pretty on-par with Safari 3.0.4

I wonder what other mobile browsers get?

It'll be sweet when the latest webkit makes it's way down to the iPhone :-)

Re:Mobile Safari (iPhone)

[dwater]

Nokia Web (also webkit) on E90-1 doesn't work at all. I'm not sure why...it pops up a txt note with some 'fail' html in it.

odd.

Re:Of Course IE will fail, ACID test is biased...

[StonyUK]

I'm getting 66 today's FF3 build.

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b5pre) Gecko/2008030405 Minefield/3.0b5pre ID:2008030405

Re:How do the acid-test creators test the acid tes

[Anonymous Coward]

The FBI have their own browser, specially developed in house in collaboration with Bill Gates, Steve Jobs, Jimmy Wales, Linus Torvalds, RMS, Kevin Mitnick and Hugh Jackman (not many people realise this, but 'Swordfish' was actually a documentary on his life).

This browser is the absolute mutt's nuts of browsers. Not only does it render every web standards test 100%, it has done so since version 0.05b (now up to 3.1). By that I mean that version 0.05b will render Acid3 100/100 out of the box (and, I can assure you, Acid4 when it is released).

Anyway, the FBI don't release it to the general public as it would make the web-browsing experience too enjoyable and, when people stopped doing their business due to too much time spent online, then society as we know it would crumble (the US sub-prime disaster and Enron were directly linked to a copy which was smuggled out of the FBI and used by senior partners in influential financial firms). They do produce a *png for Acid3 though and email it back for the reference image.

So now you know.

W3C validator

[atomic-penguin]

If these acid tests are based on standards. Why is the only acid test that passes the W3C validator, the Acid 1 test?

1. Acid 1
2. Acid 2
3. Acid 3

Perhaps, I am missing the point of these acid tests. I'm not a web developer by trade, so I don't claim to be an expert on CSS. From personal experience, CSS has allowed me to use much less complex HTML in the little web publishing I have done. I never seem to get consistent results when I test my pages in different browsers. I hope that these "standards" Acid tests lead to greater compatibility across browsers.

Do these tests increase compatibility by pushing the envelope on new standards, or are they just a browser-war pissing contest?

Re:W3C validator

[DCstewieG]

Part of the test is to see how the browser handles invalid markup/css/whatever, which is why I don't see why ACID is so highly regarded. Developers who care about a browser following standards shouldn't care how it deals with incorrect stuff. I know I don't.

Re:W3C validator

[Millennium]

If these acid tests are based on standards. Why is the only acid test that passes the W3C validator, the Acid 1 test?

Because more recent Web standards include sections on how certain kinds of errors are supposed to be handled. These need to be tested just like everything else, but up until Acid2 many browsers weren't very good about that.

Remember, the point of Acid tests is to be a thorn in browser developers' sides: find areas of the standard that no one currently does well and test for them. Browsers shouldn't pass Acid tests when the tests first come out: that would be missing the point of the tests in the first place.

Re:W3C validator

[bunratty]

Of course developers should care how browsers deal with incorrect stuff. Most web pages today have incorrect stuff. If all browsers handle the incorrect stuff differently, they'll all show the pages in different ways. If we standardize how to handle incorrect stuff, web pages can look the same in every browser even if they are not correctly made.

Re:W3C validator

[BZ]

> Do these tests increase compatibility by pushing
> the envelope on new standards, or are they just
> a browser-war pissing contest?

Some of both. Last I looked at the ACID3 draft, it tested some useful things, and a whole bunch of pretty useless ones...

Re:W3C validator

[DCstewieG]

Sorry, I meant web developers. When I'm making a web page I care that the browser correctly interprets what I'm telling it. I don't care how it interprets the incorrect syntax that I'm not going to use.

I understand making everything incorrect work predictably too, but it seems like a waste to be worrying about that when there are so many cool CSS features unimplemented - across all browsers.

Re:W3C validator

[Anonymous Coward]

I don't care how it interprets the incorrect syntax that I'm not going to use. I agree, but the thinking is that if web devs test their work by iterative editing and preview and they hit upon that's broken syntax but right only then it's less likely to be portable to other browsers that handle the broken syntax differently.

However I completely disagree that we should be focusing on the 'cool' CSS features. Let's get the basics solid everywhere before we even worry about the frippery. Like you said, who cares about the stuff I'm not going to use?

Re:W3C validator

[jaques]

The Acid tests test for correct behaviour even with broken standards - so they deliberately include non-standard JS, CSS and HTML to check that the browser handles the non-standard data according in the way specified in the standards.

Hence why they don't pass validation tests.

Re:W3C validator

[Keyper7]

From the Acid2 page::

Note: Some 827 people (rough estimate, contents may have settled during shipping) have written to point out that the CSS used in the test is invalid. This is deliberate, as a means of exposing the ability of user agents to handle invalid CSS properly. I'm guessing it's the same case with Acid3.

Re:W3C validator

[Bogtha]

Developers who care about a browser following standards shouldn't care how it deals with incorrect stuff.

Some specifications (e.g. the CSS specifications) define error handling, so the Acid tests need to include invalid things in order to test the conformance to these parts of the specifications.

Part of the reason why error handling is defined is so that current implementations will have defined behaviour when they encounter code that is written to a future specification. When you write a stylesheet using CSS 3, it would be nice to have browsers that only understand CSS 2 behave consistently even though they don't understand the CSS 3 parts of your code.

For instance, if a browser that understands CSS 2 encounters a CSS 3 property it doesn't recognise, should it ignore just that property or the whole ruleset? Don't you think this is important to include rules like this for compatibility reasons? And shouldn't browsers be tested to follow those rules?

Re:W3C validator

[bunratty]

In a way, yes it is a waste. But when 90% of the pages on the web don't validate, it is a necessity. Your fellow web developers don't seem to be as diligent as you are, and they do care what browsers do with their buggy code, as do web users. It seems as if web developers such as yourself are the only people who don't care, and you're a tiny minority in the big equation.

Re:W3C validator

[Ixa]

> Why is the only acid test that passes the W3C validator, the Acid 1 test?

W3C CSS validator does not pass Acid tests 2 and 3. For example, Acid 3 is using HSLA color values, but the validator gives parser errors for those.

Re:How do the acid-test creators test the acid tes

[CaptainPinko]

They don't. They read through their spec carefully and presumably do it by hand. I believe with the first ACID test their was a bug with the reference image that was reported by someone implementing a browser.

Testing mentality: GRRRR!!!!!

[spaceyhackerlady]

I tried it with my old version of Firefox (dismal), but found the source interesting.

The whole philosphy of testing is not "I hope it works" but "I bet I can break it!". Good test cases are diabolical, probing boundary conditions and generally trying very hard to mess things up. Typical input data doesn't look like something any sane programmer would write.

...laura

Re:How do the acid-test creators test the acid tes

[Supergibbs]

I also wondered this, I assume they took the time to meticulously "render" it by hand, perhaps multiple times by different people (or groups of people).

In response to the other replies, I am not sure how you got from the parent that they thought the test was biased against IE. IE could have passed flawlessly and other browsers all fail but the test wouldn't have changed because it test the next generation web specs. The creators of ACID3 don't care how any browser does (well they want them all to pass of course, but they don't write the test for or against any particular browser).

The Acid Test Finally Relevant

[R3d Jack]

With everyone flunking Acid-2, I had to wonder if there was any point to the test. Now, with Webkit closing in on perfection, that pushes everyone to improve. I'm honestly amazed that quality and full compliance with standards actually matters to anyone. Long live OSS!

my browsers:

[rubah]

I felt bad that firefox3 beta 3 only got a 58/100, but then safari got a 39, IE5mac didn't even get started (the poor thing asked me what I wanted to do with empty.png and flailed sadly.) and camino hasn't even loaded it(and I don't think it will).

safari's is kinda funny; the last image that showed up was a picture of a cat or something xD

Here's to another few years of improvement! *toasts*

Re:How do the acid-test creators test the acid tes

[danfromsb]

Assuming that each of the tests are passed by at least one browser you would be able to check them individually. I don't know if they are dependent upon previous tests though, in which case this method would obviously not work.

ADDENDUM

[The MAZZTer]

IE8b1 gets a 17.

Browser rundown

[RiotingPacifist]

from posts here:
IE6 - 4
IE7 - 12
IE8 - 17
firefox2 - 50 (also epiphany2)
FF3.0b3 - 59 (mine)
FF3.0b5pre - 67
Opera 9.25 - 46
Opera 9.5 - 65
Safari 3.0.4 - 39(win) 34(mac) 38 (iphone)
Safari (unkown nightly) -90
Konqueror 3.5.8 - 0 (crash)
konqueror 4.0.2 - 61
konqueror 4.1-pre - 63
SeaMonkey (2.0) nightly - 65 (Gecko/2008030501)

so for next gen its safari >> Firefox > opera > konqueror >>>>>>>>> IE8
and for current browsers its konqueror > Firefox > opera > safari > IE7

Interesting that Opera is no longer the standards king.

Re:Browser rundown

[bunratty]

When was Opera ever standards king? Before Opera 7 I couldn't use Opera because it couldn't reflow web pages correctly to handle DHTML. The Mozilla and IE of the day could handle reflow just fine. I couldn't use Opera 8 because it didn't support XSLT, and one website that I frequented used XSLT. The IE and Firefox of the day could handle XSLT just fine. Opera's got good standards support, generally about as good as Firefox or Safari, but it seems like it's often playing catch-up in at least one area. But standards king? I think not.

Re:Browser rundown

[RiotingPacifist]

my apologies i dont know where i got that impression, but your correct it apears that opear was the 3rd non beta to pass acid2 (after konq & safari).

But as i have nothing but contempt for safari after http://apple.slashdot.org/article.pl?sid=05/05/12/1555240, im never going to accept that its the most standards complaint.

Re:Browser rundown

[99BottlesOfBeerInMyF]

But as i have nothing but contempt for safari after http://apple.slashdot.org/article.pl?sid=05/05/12/1555240, im[sic] never going to accept that its[sic] the most standards complaint[sic].

First, what does a supposed feud between the Safari and Konqueror teams have to do with which is most compliant? Second, you should probably have read the comments from the Konqueror guys before trying to demonize the Safari guys. According to them (KHTML team), several of them were unhappy with the code dump Apple published, and one of them made a forum posting to that affect... but none of them had asked the guys at Apple to give them a more granular version with better documentation and when they did, the guys at Apple went out of their way (read above and beyond the requirements of the license) to provide them with it. Then, the Konquerer guys started rolling in some of the code, but then decided they did not like some of the architectural decisions Apple had made, so they intentionally held off on integrating a lot of them.

Basically, the Konqueror guys were used to owning all the code instead of being one group of contributors. Apple was new to the scene to, and they had not yet established good lines of communication. Eventually, not only Apple, but also Google, Nokia, Adobe and several other players started contributing as well and the Konqueror guys decided there was simply too much good work being done by others for them to maintain a separate fork and jumped back into sharing a code base.

In fact, once a few people involved with Konqueror stopped complaining on forums only read by their team and some KDE people and actually contacted Apple, Dave Hyatt of Apple went through the code base commenting everything that linked to Apple specific libraries as well as calling out all the fixes for the Acid2 test with comments tagged specifically for the Konqueror developers. Several Konqueror developers had already made public statements that even if they never used Apple's code, Apple adopting it was the best thing that had ever happened to the project. They also made a few comments about the "nonsense" on Slashdot that misrepresented their views.

To put it simply, the only people upset at Apple about Webkit are people who are poorly informed.

Re:Browser rundown

[99BottlesOfBeerInMyF]

Hmm, what OS are you using for the ones you don't specify (Firefox, Seamonkey, and Opera)? Linux for all? In any case, your numbers are very close to mine.

Re:Browser rundown

[bunratty]

I don't know if Safari is the most standards compliant either. Out of the latest stable release versions of Firefox, Opera, and Safari, Safari does the worst on Acid3. On the other hand, out of the latest development versions of those browsers, Safari does best. It looks to me like Safari passed the Acid2 test quickly, and will pass the Acid3 test quickly, not because Safari is standards king, but because Safari developers work quickly on making Safari pass Acid tests. And remember, how well browsers do on Acid tests is not any kind of measure of how good the browser is.

Firefox, Safari, and Opera all have very good standards support, and some are better than others at one or another area of web standards, but no one browser stands out as generally better at standards than the others.

Re:Browser rundown

[RiotingPacifist]

I missed your post which is much more complete than mine, i compiled the numbers from various Slashdot posts. I personally only tested firefox3b3 and konqueror 3.5, which are both done on Linux (kubuntu 7.10).

I assume the numbers are similar because the OS should make little difference to cross platform code ( any difference is down to a platform specific bug), thats why i removed any OS details for everything but safari (where i found it interesting that the windows version performs better than the mac version).

Firefox + WebKit?

[sapphire wyvern]

So, WebKit is pretty clearly a very superior rendering engine for the web. Fast & accurate.

But I love my Firefox chrome & extensibility so very, very much. Safari for Windows is Not That Great by all accounts and I'm not aware of any other decent Windows WebKit-based browsers. Gecko's a pretty good rendering engine and I imagine it will continue to improve... but WebKit's reported performance is very tempting.

Would it be technically feasible to create a version of Firefox that uses WebKit as a rendering engine rather than Gecko, or is that a nonsensical suggestion? I'm not sufficiently familiar with Firefox's internal structure to know. I do know that the IETab extension allows the IE rendering engine to be embedded into Firefox, but the integration is not perfect.

Re:Firefox + WebKit?

[bunratty]

If you look closely at the results people are testing, Webkit wasn't doing so well when the Acid3 test was first becoming available. If you look at the most recent stable versions, Safari 3.0.4 scores only 39/100, but Firefox 2.0.0.12 scores 49/100. In that latest development version, Safari scores 90/100, but Firefox scores only 67/100. The Webkit developers are obviously deliberately fixing the bugs that Acid3 demonstrates. That great and all, but the high score doesn't mean it's an inherently superior rendering engine. It just means the Safari developers work quickly and hard to get a good score on the Acid tests, just as they did with Acid2. The scores browsers get on Acid tests really have no meaning except for how well the browser does on the Acid test, which is often just an indication of how much developers have worked on getting the browser to pass the Acid test.

Re:Firefox + WebKit?

[99BottlesOfBeerInMyF]

In that latest development version, Safari scores 90/100, but Firefox scores only 67/100. The Webkit developers are obviously deliberately fixing the bugs that Acid3 demonstrates.

I think they probably are, but I was too lazy to download a new nightly so I used one from a few days before the Acid3 test was published and it scores 87/100, which is still quite a bit better than the Firefox 3 beta nightly from the same day (67/100). That said, you're completely correct that the Acid tests are not a good general metric for adherence to standards.

I've heard several people ask about a Firefox fork using Webkit or both Webkit and Gecko and while the idea has merit, I doubt there is enough interest to sustain a project. Konquerer uses Webkit already on Linux and should be available for other platforms in the near future. It would be nice to have the ability to swap applications and rendering engines as one wished, but I don't see it happening anytime soon.

Re:Firefox + WebKit?

[BZ]

> so I used one from a few days before the Acid3 test was published

You do realize preliminary versions of this test have been available for a while now, right? For example, https://bugzilla.mozilla.org/show_bug.cgi?id=410460 was filed over two months ago.

If you want to see what things looked like before people started on fixing this stuff, you'll need to go back a few months.

Re:Firefox + WebKit?

[99BottlesOfBeerInMyF]

You do realize preliminary versions of this test have been available for a while now, right?

Yeah, but they did not finalize it until 3 days ago and they were dropping out tests that were not broken in either gecko or webkit. Admittedly both teams probably fixed a few that were initially broken in both.

Re:Firefox + WebKit?

[sapphire wyvern]

My comment wasn't just based on these Acid3 results, but also on WebKit's excellent Java/ECMAScript benchmarking results and other general positive press.

Well, I'm looking forward to Firefox 3 getting out of beta, anyway. A new version of Gecko will be nice.

Re:Firefox + WebKit?

[bunratty]

Both Firefox and Safari developers have obviously been fixing bugs that Acid3 demonstrates for the past few months. The Firefox tracking bug for Acid3 problems is over two months old, for example. If you want to see how well the browsers did before the test existed, use one from April 2007, when work on Acid3 started.

Re:Firefox + WebKit?

[bunratty]

Firefox and Opera render accurately, too. Firefox used to be a JavaScript slowpoke, but with the latest performance improvements, Firefox 3 is fast at JavaScript, too. I'm not trying to advocate one browser over another, just point out that there are at least three good, fast, popular browsers out there. It doesn't seem like one is generally better or worse than the others overall.

Re:Opera

[Anonymous Coward]

I love Opera (hey, I'm from Norway, I have no choice), but my Opera 9.5b (Ubuntu 7.10) segfaulted when I hovered the cursor over said x.

Re:Opera

[Lisandro]

Same here. I get a 60/100 on Opera 9.50b1 (i686, Linux 2.6.23).

Opera

[a_generic_name]

Internet explorer didn't even load the refrence page correctly. (Okay, maybe it did).

On a more serious note, Opera 9.21 got a 46, after crashing the first time at 36.

Re:Opera

[Dex1337]

The current Opera release 9.26 just scored 46/100 for me, and then crashed the browser...

67/100

[Dopeskills]

Firefox 3 nightly beta gets 67/100

100% Correct by Which Browser

[c0d3r]

If no browser renders the page 100% correct, how do we know if the test is even valid?

Re:100% Correct by Which Browser

[nneonneo]

All the tests conform to some part or another of some published and accepted specification. Thus, they are valid according to these specifications.

That said, they were specifically designed to fail browsers. In fact, during the last few weeks of finalizing the test, the makers specifically asked for code to fail WebKit, since it was doing quite well (and still is).

Re:100% Correct by Which Browser

[Arimus]

Very simple scenario to prove the test is valid (assuming noone made a mistake in interpreting the standards - lets leave that can of worms aside ;) )

1. I write a test which says I should run program x, x should display 'Hello World' in a dialogue box centred on my screen, it should auto close after ten seconds.

2. You write a program to display 'Hhello World' in a dialogue box centred on my screen, auto-closing after 10 seconds.

3. While coding you forget the auto-close code.

The test is still a valid test, the program is still valid, the failure is a valid fail. Everything is as it should be (other than I've now got an annoying dialogue box stuck on my screen - wait a tic that's just Vista)

Re:100% Correct by Which Browser

[c0d3r]

Often specs are not specific enough to make a test valid. Spec bugs also occur when multiple featues affect each other in adverse or undefined ways.

Re:How do the acid-test creators test the acid tes

[bunratty]

Ian Hickson actually did ask for additional tests that fail in Webkit or Firefox, so they did partly write the test against at least two particular browsers.

Lobo (100% Pure Java Browser) gets... zero

[mosburger]

Woohoo! 0%! And the page looks like complete crap, too!

Newest scores on Windows XP.

[nneonneo]

I am using Windows XP (SP2) Firefox 3.0a5 (2008030506): 67/100 (built today, March 5th) WebKit nightly (r30768): 88/100 (built today, March 5th) Opera 9.5 beta (b9815): 65/100 (built February 29th) IE7: 12/100 There you go. Information for the latest available builds, at time of writing, of the major Windows browsers.

Re:How do the acid-test creators test the acid tes

[VirusEqualsVeryYes]

Actually, I believe Firefox 1 was used, albeit with hacks in order to work around the known CSS bugs. Sorry I can't find any corroborating source; I don't remember where I read this (if I did at all), and Googling only turns up a recent post on Reddit that goes undisputed by "Hixie", who seems to be Ian Hickson.

For the case of Acid3, calculations were probably easier this time around, since Javascript is simpler to compute separately. I'm not intimately familiar with the tests, but it seems like their calculation would require far less coordination with a corrected rendering engine.

As a sister comment hinted, they don't know there are no bugs in the test. Development of Safari yielded a bug in Acid2 that was subsequently submitted and corrected (scroll to the bottom).

Re:Of Course IE will fail, ACID test is biased...

[99BottlesOfBeerInMyF]

Not quite. When none of the browsers are getting 100/100 and the only browser to get over a 60 is a safari beta, I think it's safe to say that it's a test designed so that every browser will fail.

...one minor correction for you. Firefox 3.0 beta on both OS X and Kubuntu gets 67.

Opera and Konqueror

[Apoorv]

Opera 9.25 scored a 46/100 for me, on PCLinuxOS. Konqueror 3.5.8 crashed all the 3 times I tried to open the test on it.The initial message was that JavaScript was unavailable, after which the score started increasing, and the browser crashed as soon as the count reached to 1/100.

Re:Geek version of a measuring contest?

[BZ]

None of the things this test tests are "International Standards" in the ISO sense.

win/ffx3b3

[airdrummer]

my score is 61... Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b3) Gecko/2008020514 Firefox/3.0b3

Test-driven development

[Nurgled]

This is an example of test-driven development. Rather than writing the code and then writing the test cases, you turn it on its head and write the test cases first. You produce the test cases from the requirements -- in this case, the specifications -- and then you're finished development when all of the tests pass.

If you run into a situation where you think the tests should pass and they don't, then there are two possibilities: there's a bug in your code, or there's a bug in the test. You then debug to figure out which is wrong and fix the error. The tests help you find bugs in your code, and your code helps find bugs in the tests.

Re:Of Course IE will fail, ACID test is biased...

[Glonk]

Not quite. When none of the browsers are getting 100/100 and the only browser to get over a 60 is a safari beta

The firefox nightlies (and upcoming beta 4) are getting 66/100.

Firefox 3 beta 5 pre

[sagematt]

Latest Minefield build scores 68/100.

http://img166.imageshack.us/my.php?image=minefieldb5pre200803060ey0.jpg

Acid3 tests Javascript and DOM

[alanhogan]

Acid3 does not mostly test CSS, but all kinds of things, with a particular interest in scripting and the DOM.