Welcome to WebmasterWorld Guest from 54.160.131.144

Forum Moderators: incrediBILL

Message Too Old, No Replies

Acid3 Test Released

     

encyclo

11:18 am on Mar 4, 2008 (gmt 0)

WebmasterWorld Senior Member encyclo is a WebmasterWorld Top Contributor of All Time 10+ Year Member



Acid3: Putting Browser Makers on Notice, Again [webstandards.org]

The Web Standards Project (WaSP) today announced the release of Acid3, the latest in a line of tests designed to expose flaws in the implementation of mature Web standards in Web browsers. (...) 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.

Now that many browsers can pass the Acid2 test, Acid3 brings standards-testing to a new level. No browser passes Acid3 for the moment.

The test itself can be found here: [acid3.acidtests.org...]

httpwebwitch

7:57 pm on Mar 4, 2008 (gmt 0)

WebmasterWorld Administrator httpwebwitch is a WebmasterWorld Top Contributor of All Time 10+ Year Member



If no browser can pass it, how do they know the test itself is perfect? What did the acid brewers use to create the "reference rendering"?

And - if they have a 100% standards compliant rendering tool in their secret lab, can I buy a copy of it?

My bet is on Opera for the first fully compliant release. Not that it's a race or anything.

[edited by: encyclo at 8:49 pm (utc) on Mar. 4, 2008]

Lord Majestic

7:59 pm on Mar 4, 2008 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



If no browser can pass it, how do they know the test itself is perfect?

They probably just read spec and carefully create test case to see if a particular implementation works as it is described in the spec.

esllou

10:00 pm on Mar 4, 2008 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



seems to pass in IE4.

:-))

mexicanbanana

10:13 pm on Mar 4, 2008 (gmt 0)



>seems to pass in IE4.

Uh... it breaks horrible in MSIE 6. Latest stable Opera and Firefox also can't handle it even halfways.

SuzyUK

11:34 pm on Mar 4, 2008 (gmt 0)

WebmasterWorld Senior Member suzyuk is a WebmasterWorld Top Contributor of All Time 10+ Year Member



the "test" is designed to be a challenge to interpret all parts of the recommendations the same as.... ermm the test maker interprets them, it's so not important except as anything but a general guide (who wants to make a smiley face layout or a rainbow of boxes?)

Recommendations are just that, interpretations of them from users as well as UA's has always been different and if support for all these CSS properties and the new stuff in ACID1,2,3 had had a solid benchmark, there wouldn't be any need for the "award", the browser would simply work within recommended guidelines.

incrediBILL

11:50 pm on Mar 4, 2008 (gmt 0)

WebmasterWorld Administrator incredibill is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month



People people people...

It's called QA!

You have a spec for HTML, someone writes a test case to validate that spec for HTML in a browser, if your browser doesn't pass that test case your product is crap.

As a QA loving Silicon Valley engineer, I don't see the problem.

BillyS

12:18 am on Mar 5, 2008 (gmt 0)

WebmasterWorld Senior Member billys is a WebmasterWorld Top Contributor of All Time 10+ Year Member



Yikes! The browsers I use IE7, Opera and Firefox aren't even close, although Opera seems the closest.

mattglet

1:58 am on Mar 5, 2008 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



The newest snapshot of Opera 9.5 gets up to 65/100.

Receptional Andy

4:02 pm on Mar 5, 2008 (gmt 0)



It managed to crash Opera 9.25 ;)

balam

4:04 pm on Mar 5, 2008 (gmt 0)

10+ Year Member



I'd have some faith in the test if it could produce consistent results, but it doesn't. With IE6, I have gotten scores of 11, 12, 14 & 15. My testing methods were to close all browsers, open IE, visit site, allow downloading of fonts, allow running of ActiveX, watch test fail, close browser, rinse & repeat. (I also tested not allowing fonts & ActiveX; this had no perceivable effect on the final test results.)

Bottom line? The Acid Test fails the Balam Test by virtue of the fact that the Acid Test cannot produce consistent results when presented with a consistent test environment.

<added>
FWIW, Netscape 4.08 just explodes when tested... :)
</added>

mikedee

6:08 pm on Mar 5, 2008 (gmt 0)

5+ Year Member



Bottom line? The Acid Test fails the Balam Test by virtue of the fact that the Acid Test cannot produce consistent results when presented with a consistent test environment.

What makes you think IE is a consistent browser? Firefox 2.0 always produces 50% here.

penders

3:38 pm on Mar 6, 2008 (gmt 0)

WebmasterWorld Senior Member penders is a WebmasterWorld Top Contributor of All Time 5+ Year Member Top Contributors Of The Month



With IE6, I have gotten scores of 11, 12, 14 & 15.

With IE6 I get 12 (consistently).

Firefox 2.0 always produces 50% here.

With FF 2.0.0.12 I get 51 ! ;)

It managed to crash Opera 9.25 ;)

The first time I ran it on Opera 9.26 it got to 46. The second time it crashed!

mehh

6:01 pm on Mar 7, 2008 (gmt 0)

5+ Year Member



58 in a Opera 9.5 snapshot. Not sure which one

penders

10:17 pm on Mar 7, 2008 (gmt 0)

WebmasterWorld Senior Member penders is a WebmasterWorld Top Contributor of All Time 5+ Year Member Top Contributors Of The Month



Firefox 2.0 always produces 50% here.

With FF 2.0.0.12 I get 51 ! ;)

I get 50 in FF1.5 !

It managed to crash Opera 9.25 ;)

The first time I ran it on Opera 9.26 it got to 46. The second time it crashed!

A measly 2 in Opera 8 - but it didn't crash! ;)

youfoundjake

10:32 pm on Mar 7, 2008 (gmt 0)

WebmasterWorld Senior Member 5+ Year Member



IE8 gave me 17. Nice big letters, FAIL
heeh

dealguien

10:42 am on Mar 8, 2008 (gmt 0)

10+ Year Member



On Ubuntu:
- Firefox2 got 51 points.
- MSIE 6.0 got 11 points.
- Konqueror 3.5.8 crashed all 3 times I tried.
- Opera 9.25 got 46 points

Way to go. Interesting to look at the CSS code though...