Yet Another Google Chrome Post

Doubtless, there are already lots of blog entries about Google's entry into the browser wars with Google Chrome.  There are probably a large number of these entries that are focused on performance already out there, and countless more tests will be done using much more scientifically methods than I've used here.  Am I breaking new ground?  Probably not.  I'm merely lending my small voice to the crowds that are already disecting Google's actions and motives around this recent move.

I, like many others, rushed out and downloaded Chrome as soon as the site was up.  I've been an avid user of Firefox since 2.0, and I used Firefox 3.0 to download the installer for Windows.  The download was fast, because it seems that like many other application installers out there, Google opted for an installer that is small and downloads the real packages to be installed from the web after you launch it.

After waiting a few moments to get the rest of the data needed for the install, I was prompted that Chrome could import all my Firefox settings, but that I needed to close Firefox for it to do so.

After the install completed, I fired up Chrome to take it for a test drive.  It looked very much the same as the screen shots shown on the Chrome site, with a minimalist UI and browser page tabs on the top of the window.  The first thing I noticed when using Chrome was the speed.  It seemed quite fast to render pages compared to IE or Firefox.  More on that later.

There are many nice features for developers with a JavaScript console that reminds one of a popular Firefox plugin for examining HTML and JavaScript on a web page.  Also, Chome takes the approach of making the major engines of the browser run in separate processes.  Each tab and browser plugin get's its own separate process that won't impact the stability of the other tabs.  You can see this by looking at the Windows Task Manager and noting the multiple instances of chrome.exe with different PIDs.

I ran some basic tests on performance to compare just how my perception of Chrome's speed matched with reality.  I tried to make the tests as clean as possible by disabling all unecessary programs and services before running the benchmarks.  All the tests were performed on a Dell Latitude D830 with an Intel Core 2 Duo @ 2.20 GHz and 4 G of RAM on Windows Vista SP1.

JavaScript
Google Chrome sports a clean implementation of JavaScript in an engine called V8.  I ran Chrome through WebKit's SunSpider JavaScript benchmark and compared it to the results of IE 7, and Firefox 3.0.1.

BrowserTotal Time
IE 730368.0ms +/- 8.5%
Firefox 3.0.15995.2ms +/- 10.5%
Chrome 0.23943.0ms +/- 7.7%

Chrome was overall about 7.7x as fast as IE, and about 1.52x as fast as FireFox in the test. Interestingly, Chrome performed slower in both the Date and Regexp tests than IE and Firefox by about 1.5x. Both of these areas could be important to performance in dynamic, client-side sorting of data tables.

Render Speed

Using a page load timer, I found at http://cache.lifehacker.com/assets/resources/stopwatch.php, I tested loading the http://www.webstandards.org/action/acid2/guide/ page twice to make sure as much network latency was removed from the test as possible.  I refreshed the page 5 times to get an average for each browser.

BrowserTotal Time
IE 7577 ms
Firefox 3.0.1414 ms
Chrome 0.2506 ms

The times were so close it's hard to call, but it seems Firefox had an advantage in this particular test.

Comments

Post a Comment

Popular posts from this blog

Using Snapshot Isolation with SQL Server and Hibernate

In the course of working on some deadlock issues, I found that MS SQL Server exhibits some unexpected (to me, at least) locking behaviors that can affect performance and cause deadlocks. In the end, I found that SQL Server provides an isolation level called Snapshot isolation which removes the need for locks on rows and removes lock contention. To use this isolation level, you need to execute the following SQL on the database that you want to use the isolation level with (replacing MyDatabase with your database name, of course): ALTER DATABASE MyDatabase SET ALLOW_SNAPSHOT_ISOLATION ON ALTER DATABASE MyDatabase SET READ_COMMITTED_SNAPSHOT ON Next, you need to modify your hibernate.cfg.xml file to add a property to tell Hibernate to use the Snapshot Isolation level: <!-- The 4096 isolation level is the setting to use with the jTDS or Microsoft JDBC drivers --> < property name =”hibernate.connection.isolation”>4096</property> Snapshot isolation is not a panacea...
Read more

FreeNAS, CIFS, and user login problems

Image
I have been in the process of converting from the dead-end Windows Home Server product to a FreeNAS server for my home NAS.  I was trying to replicate the same sort of set up that WHS had for user directories.  In WHS, there was a share called "Users" and under that share there were various user directories corresponding to each user account.  I thought it would be pretty simple to replicate that same behavior to allow users to simply attach to \\nas-server-name\Users in Windows, and then navigate to their own user folder.  What follows is how I achieved this, as well as a problem I fought with for a while that prevented certain users from connecting via CIFS. First, I started off the process by installing FreeNAS to a USB stick, added a few disks in the server, plugged in the USB stick and booted up.  I performed basic config for the networking, and admin account passwords, and turned on the CIFS service. Next, I created a some user accounts fo...
Read more

Official Pandora Desktop and Hotkey Support?

EDIT: There is a slight problem occasionally with this script not being able to send keys to the Pandora Desktop Air app. I've got an updated version of the script that you can use instead of this one. Pandora is a different sort of online music service that uses the work of the Music Genome Project to automatically selects music similar to an artist or particular song, and then streams it over the Internet for your listening pleasure. You can't select individual songs, but you can skip ones you don't like and make as many music channels as you like to suit your particular tastes. Pandora provides a free version of the product that is supported by visual advertising, and a very inexpensive ($36 USD per year as of the creation of this entry) subscription supported version of the product that provides higher quality streams and no advertising. Traditionally, Pandora has been a Flash-based application that required you to have your bulky browser open to listen on the deskto...
Read more