Filed under: Internet Tools, Beta Beat
Google Chrome developer preview (unofficially) released
Digg's Kevin Rose, perennial purveyor of information that just "fell off the back of the truck" shared a link early this morning to Google's new, supercharged, Webkit-based browser -- for Mac.
The new browser, Chrome, is clearly marked as a developer preview, and not meant for general browsing. In fact, as a good Mac citizen, it will refuse to set itself as your default browser.
It scores a 100 on the Acid3 test straight out of the box, but doesn't pass: it fails something called the linktest, which involves interacting A tags and IFRAMEs. A little research suggests that it could be a bug with Webkit. If that's true, then it's a bug that Safari 4 beta has fixed.
Chrome appears to have Flash (and other plugin) support disabled as well. JavaScript support, however, is fully functional. While Chrome performed much better than Firefox on this cursory test, it still didn't beat Safari 4 by a long shot. While only a beta, its performance is respectable and sure to improve.
Again, Google Chrome isn't for everyone, but if you're a web developer who needs to keep up with the bleeding edge of browser development, then this preview should be stable and reliable enough for you to test what you need to.
Read through for the full JavaScript test details.
Thanks to everyone who sent this in!

I ran this test on my iMac with a 2.8 GHz Intel Core 2 Duo processor. It was meant more to show the relative execution speed differences between the browsers rather than a demonstration of raw horsepower. Chrome executed JavaScript instructions in this test 2.7 times slower than Safari 4. Firefox executed the instructions 7.4 times slower than Safari 4.
Update: Commenter Manu points out that there are other, possibly better, tests out there that give different results. Thanks, Manu!

![TUAW [Cafepress]](http://www.blogsmithmedia.com/www.tuaw.com/media/tuaw-cafepress-promo.png)


Reader Comments (Page 1 of 2)
AriX said 11:30AM on 6-05-2009
Umm... It's not unofficial if it's on Google's Chrome download site ;)
Reply
Robert Palmer said 11:39AM on 6-05-2009
True, but it's not linked from http://www.google.com/chrome. It still says "sign up for more information" about the Mac version. If it were official, I think they'd link to it from there, don't you?
AriX said 11:43AM on 6-05-2009
I consider it official since it's on Google's site, even if it's hidden. When I read unofficial, I assumed it was someone compiling Chromium for Mac... The definition of official is "Of or relating to an office or a post of authority", and in this case Google is the authority. If it's coming from Google, it's not unofficial.
Andre said 11:37AM on 6-05-2009
Does it include the dreaded Google Updater? If so, I won't be trying it.
Reply
Robert Palmer said 11:40AM on 6-05-2009
It didn't appear to. The DMG that you download appears to be just the Chrome app and an alias to your Applications folder (for convenient dropping).
Manu said 11:47AM on 6-05-2009
Don't think it has the google updater but don't think it needs to. Chrome has an inbuilt update mechanism which automatically updates your browser and it seems the dev channel build has that enabled. http://twitter.com/mikepinkerton/status/2042621181 [Mike is the lead developer for chrome on mac]
Drunken Economist said 3:43PM on 6-05-2009
Yes, Unfortunately it DOES include the freaking GOOG Updater.
As I ran Chrome for the first time, Growl informed me that a 'Google Updater.dmg was being mounted and executed in the background.
At this point, GOOG = MSFT
Oh wait, even MSFT isn't that evil.
I am not using Chrome. EVER.
Andre said 4:13PM on 6-05-2009
OMG, you're right it does. I just looked in side the APP container, and it contains the "Keystone Registration" components.
It never asked me for my authentication, so I guess it couldn't install anything that will run at startup though.
Still, out of protest, I am deleting the app.
Chris Deke said 3:56AM on 6-06-2009
OMG How dare an application try to update itself! Such blasphemy!
I really don't understand why people are getting so bent out of shape about a freaking updater app. I find Firefox far more annoying because I actually have to think about the updates, I absolutely LOVE that Chrome "just works".
Andre said 8:47AM on 6-06-2009
It's not that it updates itself, it's that in all other (released) Google products, it installs itself as root, and runs in the background all the time. It checks for updates to all Google products, even when you are not using a Google product.
If you're OK with something running all the time on your computer (and you have no choice), then good for you. The rest of us don't want a company like Google thinking they're so important that their updater needs to become part of our OS.
Besides these concerns:
- you may have a limited bandwidth Internet connection. Google doesn't care. They'll download updates when they feel like it with no input from you.
- Maybe you don't want the latest version. Maybe your IT department likes to test applications in a specific environment before rolling out new versions. Maybe you like to wait yourself. Some people like to hear from others if an update breaks anything before they update. Maybe you like the old version and don't want to update.
- Maybe having something running 100% of the time (even when you aren't running one of their apps) is a security risk. A daemon running as root, that you can't turn off... what if it has a vulnerability? It could be exposing your system 100% of the time.
- Maybe you just don't want Google's software running even when you didn't call up one of their apps. Maybe you want all of your CPU and bandwidth for your own purposes.
- Or maybe you just don't care, so you will support this and ridicule anyone that is against it.
An application checking for updates is a good thing. User control is even better. Most good apps check when you launch them, not when you're in the middle of running some critical application of your choosing.
Google apps used to do this. A checkbox in its preferences to automatically check for updates. It works well with most apps, and it worked for Google apps too. You had the choice, and you also were given the choice about actually performing an update. Google no longer gives you these choices, and they demand that you run their updater all the time.
Oh, and it's a 3rd-party app: Keystone. Remember Sony's forced software from a 3rd party?
If you check around on this topic, you'll see that people really don't want this. Sure, there's a few people arguing it's a good thing, but they don't grasp the issues of having some utility running every time they start their computers up, and running all the time. These people probably have tons of things in their startup, and either don't care that their system is slow, or don't have a clue what to do about it. In the Windows world, these are the people with 30 tiny icons in their system tray, each hogging some of the CPU, and doing who-knows-what.
Oh, and when Google started this, they weren't even telling people. Now they tell you as part of the installation but they don't let you opt out. If you say no, you don't get the app.
If you're running any recent releases of Google apps, you already have this thing installed. Either you installed it prior to them telling you about it, or you blindly clicked 'yes' on their little pop-up.
If you installed this pre-release version of Chrome, you have traces of the updater. Look in your ~/Library/Google folder. Look at the contents of Chrome's app file. You'll find the Keystone components in there (somewhere in Frameworks). Look through some of the settings files for clues on what that app does - and what it will be doing once the released version of Chrome comes out.
I suggest that TUAW take the lead on this issue, and do an investigative story. Most people don't realize what Google is up to. TUAW needs to help tell Google "it's my computer not yours", and to go back to their original plan of not being evil.
There is a good thread on this issue on one of Google's own forums. People are complaining about it, and Google basically says "too bad, we know better". You will notice that people just don't want this thing running, and Google's responses are not around removing it, but finding a way for it to exist and make everyone happy. They just don't get it.
http://groups.google.com/group/google-earth-browser-plugin/browse_thread/thread/146882feca2788c3/30a5a5eb60b0c6b6
Chris Deke said 10:12PM on 6-06-2009
The reality is that chrome is NOT a finished product. It got a VERY rushed 1.0 release (and a similarly rushed OS X release) due to public pressure.
This is NOT the browser you should be using if you want to control your version and build webapps. Firefox is thataway -->
At the end of the day, the updater is part of the chrome experience - in my mind a great one, in google's mind, a vital one. If you don't like it, again, Firefox is thataway. As with all things, its google's prerogative to decide how THEIR application behaves.
You can always go and fork webkit yourself if that's not good enough for you.
Dan said 11:38AM on 6-05-2009
"Fell off the back of a truck info"?
It was on Chromium's blog yesterday -- "http://blog.chromium.org/2009/06/danger-mac-and-linux-builds-available.htm -- Rose didn't exactly find a scoop...
Reply
Manu said 11:39AM on 6-05-2009
The test that you linked to has wildly varying test results on the same browser. Sunspider is the most popular javascript test and on that crome edges out safari and on V8 chrome beats every other browser easily. For more you can see this post,
http://www.manu-j.com/blog/mac-os-x-browsers-javascript-performance-shootout/310/
where I have provided the graphs.
Reply
Chachi said 11:42AM on 6-05-2009
Does it support HTTP_AUTH? I don't see a login window when visiting HTTP_AUTH protected pages...
Reply
Libby said 12:12PM on 6-05-2009
WebKit based Stainless (http://stainlessapp.com) scores 100/100 on Acid3 and has no linktest problem, so it's not a WebKit thing.
Reply
SpinThis! said 12:53PM on 6-05-2009
> if you're a web developer who needs to keep up with the bleeding edge...
Bleeding edge indeed. Chrome seems interesting as browser but *who* exactly are these bleeding edge web developers you speak about? Is there even any reason for us developers to even *test* Chrome for any reason other than curiosity? It should render similar to Safari... and even if it doesn't, is it really worth fixing (beta) bugs for 0.1% of people who might *actually* use it? Too bad with all these new browsers coming, developers are still hacking around IE bugs.
Reply
Max said 1:11PM on 6-05-2009
Plus, how can you use this to "proof" pages if it doesn't even have Flash support?
As you said, to proof webkit on mac, use Safari. To check for "real" Chrome users, better off using Windows.
Alex McKee said 1:34PM on 6-05-2009
Smart Scroll doesn't appear to work as of yet. I am not amused.
Reply
Mario said 6:14PM on 6-05-2009
MG's tests show different results. Chrome appears to be faster than Safari :(
http://www.myappleguide.com/news/2505/google-chrome-snappiest-them-all
Reply
NightOne said 7:30PM on 6-05-2009
Do we really need another browser? :)
Reply