Acid 3 - The Web Browsers Test

[Smigit]

Does this test actually really even matter? Really? All my website load nicely in Firefox 2. Anyway, the IE Epic Fail's are funny. :laugh:

I honestly don't think it's THAT important really. I mean, it's good to have as much support as possible but the ACID tests are really subsets of what's out there standard wise. Certainly theres a certain benchmark that should be achieved, but whether you get 95 or 100% I don't think would matter. The fact that browsers that pass Acid 2 then go on and get 60's or 70's here just shoes it's rather dependent on what subset of features Acid supports with any given version. Had they chosen other tests the results could be fairly different so I wouldnt use it as the sole reference to who supports standards the best.

I'm sure any browser that gets even 15 - 20 in this test will work with 98% of sites with the possibility of slight alignment issues here and there.

Still, it's a decent benchmark of a browsers performance and should be a handy tool to find problems. Also a nice benchmark for browser developers aim for even if I think the test is a bit superficial. You definetly get brownie points for passing but.

edit: I suppose you could say Acid 3 is needed but since Acid 2 is about to become irrelevant with FFX3 and IE's new versions both said to pass it which I think leaves all the major players passing it now.

[megamanXplosion]

The Acid tests are important.

The reason 98% of websites may look and act perfectly fine while the Acid test completely breaks down is because web designers spend so much time discovering and employing CSS hacks, JavaScript hacks, client-side or server-side browser checking, etc. to get things to look and behave properly or they do without the functionality.

This means website designers spend a lot of time trying to get things to work when they should've been done with the task hours (or days) ago. This means the home-brew website will tend to be updated less frequently and the professional website designers will require more time to do their job which means higher prices for those buying their services.

Of course, all those hacks in CSS and JavaScript aren't free, bandwidth-wise, which means webpages tend to load slower or in the case of server-side hacking the server uses more CPU resources. Or this could mean website designers forego the use of certain functionality, which means websites aren't as attractive or interactive as they could otherwise be, which saves on bandwidth and CPU at the cost of a less satisfying website.

That's only looking at the web development side. The browser development side is interesting too. If websites were less hacky, and built properly, then the time needed for "regression testing" would be significantly reduced. The web standards are built to be as backwards compatible as possible so there should, ideally, be very little "regression testing" to do. Because websites are hacky and built improperly, it takes longer for web browser releases to be made.

That it takes longer for browser releases to be made means it takes longer for newer standards to come to fruition. Microsoft, Mozilla, Opera, etc. all come together to create these standards but they can't really push them out the door until their products have caught up with the existing standards, but they all have troubles catching up with existing standards because every little change in their rendering engine requires a lot of regression testing and involves irritating a lot of the people who build those hacky, improperly built websites.

In summary, if browsers followed the standards more closely, some websites would have more functionality and some websites would be faster with their current functionality, website designers could update their websites quicker, browsers could be updated quicker, standards could be pushed out the door quicker, and so on, Everyone wins when rendering behavior is standardized, which is why the Acid tests are important.

[nonick]

FF3 prebeta4 scores 66/100

[+Nik Louch Subscriber²]

In summary, if browsers followed the standards more closely, some websites would have more functionality and some websites would be faster with their current functionality, website designers could update their websites quicker, browsers could be updated quicker, standards could be pushed out the door quicker, and so on, Everyone wins when rendering behavior is standardized, which is why the Acid tests are important.

An Utopia we won't reach. You final shot is "standards could be pushed out the door quicker" - pushing us back to square 1.

I'd love the above to happen. Devs support standards, browsers supporting standards, standards supporting the devs.

[shakey_snake]

**** yeah, WebKit!

(What a coincidence... IE going standards and this in the same day...)

The real "coincidence" is that these things for whatever reason always seem to get published when Gecko is towards the end of it's development cycle. :wacko:

[Mathachew Veteran]

I'm pretty sure there's only one browser that passes the Acid 3 test ATM, Lynx with JS turned on

If you squint a little, you can see the 100/100

:laugh: that's great :D

[SnowRanger13]

Safari 3.0.4 on OS X 10.5.2 scores a 39.

Webkit (Pretty close to what Safari 3.1 will be) on OS X 10.5.2 scores a 87.

[seta-san]

and IE8 just got acid2 compliant too :p

[Fonze]

This is an over simplified explanation of why web standards are important.

Let's say you want to draw a square. You tell the browser to draw a square. Firefox interprets what you said and draws a square with slightly rounded edges, Opera draws a rectangle, and IE draws a triangle.

This sounds ridiculous, but when it comes down to it, this is basically what is happening. If you say draw a square, the browser should draw the same exact square as every other browser.

Web browsers adhering to web standards reduces the amount of work web developers have to do to make their pages look and work the way they intended because they don't have to worry about writing browser specific code.

[megamanXplosion]

An Utopia we won't reach. You final shot is "standards could be pushed out the door quicker" - pushing us back to square 1.

How does it push us back to square 1?

[GEIST]

Got stuck at 50/100 with Firefox 2.0.0.12 on Windows XP Pro 32bit

[cck_my]

on Vista

IE7 failed 12

Opera 9.50 build 9815 scores 65/100

Safari 3.0.4 scores 66/100

Firefox 3.0 beta4 scores a 67/100

[seta-san]

firefox 3.0 beta3 gets a 59

[Ayepecks]

Oh em gee, MOSIAC scored a ONE HUNDRED! Everyone go download it now, seriously!!!!

:shiftyninja:

[DreadBoat89]

Firefox 3 Pre-Beta 4 shows 67... Not great, but not bad.

Firefox 3 b5pre shows same result...

Webkit (Pretty close to what Safari 3.1 will be) on OS X 10.5.2 scores a 87.

same result on windows.

Edited March 4, 2008 by Dynames00

[Fourjays Veteran]

I'm not convinced that Acid tests are really useful in the real world. Don't get me wrong, I support standards, but it seems the Acid test purposely tests the areas that the browsers will interpret differently rather than the areas that are most commonly used in web design.

I'm currently working on a fairly complicated design, and so far it renders near perfectly in FF, Opera and IE5-7 with only 2 "* html" bits in my CSS. I personally think that if you are needing hundreds of hacks, you're doing something wrong (or plain lazy, which I've come to realise is very common amongst web designers/developers).

Even if standards were applied flawlessly across all browsers, some things (fonts) will never be the same across all browsers and all systems.

[Fonze]

Even if standards were applied flawlessly across all browsers, some things (fonts) will never be the same across all browsers and all systems.

Why not? If the web browser renders the font, why wouldn't it be the same? If the font isn't included in the browser, it could be downloaded like an image for a web page. There is no reason why fonts can't be universal across all browsers in the future.

[DreadBoat89]

Why not? If the web browser renders the font, why wouldn't it be the same? If the font isn't included in the browser, it could be downloaded like an image for a web page. There is no reason why fonts can't be universal across all browsers in the future.

or better yet, have an alternative font to fall back on, making sure it still looks proper

[jimbo11883]

67/100 on Minefield 3.0b5pre (Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9b5pre) Gecko/2008030405 Minefield/3.0b5pre)

[Menge]

Firefox 3.0b5pre scores a mighty 67/100 :p

[Fourjays Veteran]

Why not? If the web browser renders the font, why wouldn't it be the same? If the font isn't included in the browser, it could be downloaded like an image for a web page. There is no reason why fonts can't be universal across all browsers in the future.

If there is a way to automatically download a required font, I'd love to know about it!

or better yet, have an alternative font to fall back on, making sure it still looks proper

Extremely difficult in my experience. Tends to be a case of - if something similar design-wise can be found, it is slightly too big or too small (which can cause design problems).

It is more of a cross-OS issue than cross-browser. As long as you stick to Windows defaults, cross-browser doesn't make much difference in Windows. If you then go to Linux, it doesn't have the Windows defaults by default. Even with the MS Core Fonts package it is (in my experience) still different from how it is in Windows.

Unless everything is standardized, there is always going to be something slightly different that influences how a design looks across different systems.

[Steve Gosselin]

on Vista > Safari 3.1 scores 65/100

[megamanXplosion]

I'm not convinced that Acid tests are really useful in the real world. Don't get me wrong, I support standards, but it seems the Acid test purposely tests the areas that the browsers will interpret differently rather than the areas that are most commonly used in web design.

It does purposely test things that browsers render differently. The purpose of the tests is to encourage browser vendors to improve interoperability.

I'm currently working on a fairly complicated design, and so far it renders near perfectly in FF, Opera and IE5-7 with only 2 "* html" bits in my CSS. I personally think that if you are needing hundreds of hacks, you're doing something wrong (or plain lazy, which I've come to realise is very common amongst web designers/developers).

I agree that for most tasks you will not need many hacks, if any at all, but there are some tasks that require extensive hacking. Ideally, though, there should be as little hacking as possible. The reason hacks are used is because of interoperability problems, which is where the Acid tests come into the picture.

[torrentthief]

Hope firefox 3.0 Final gets 80%+ I doubt IE8 Final will get more than 60%

[Fonze]

If there is a way to automatically download a required font, I'd love to know about it!

Unfortunately I don't have much knowledge on how the browsers use fonts. I'm assuming that this is a harder feature to implement because browsers are probably relying on the OS to render the font. If this is true, then they would have to move to a model that does not rely on the OS, so the fonts could be universal across all browsers and could be downloaded on the fly.

Maybe someone that develops web browsers could respond to this one. Anyone from the Firefox or IE team out there? :D