Weekend project: compare your browsers on Linux

With the release of Firefox 4.0 right corner, the big question for a large number of users is how fast is Firefox 4.0? How the new Moz compare with Google Chrome, Opera and the rest? If you're curious, take some time this weekend to perform your own performance tests and see for yourself. This consider a section of public participation: we are looking for your comments as well.

Whenever the released a new version of browser, you will see a rash of stories about browser benchmarks and how this or that browser are faster. The problem is that they focus generally on a test, and they are usually aimed at Windows - not users of Linux users.

Couple that with the fact that guides usually work with standard builds of browsers for Windows or maybe Mac OS X, on what systems are largely standard. Equipment varies, but the libraries system and such must be the same for all Windows 7 users, or all users of Mac OS X (modulo a few variations according to the question of whether reference systems have been updated correctly).

Linux users can get Firefox from their provider or from the official Firefox builds. They can use 32-bit version or 64-bit. They may have different versions of system libraries as they are on Fedora, Debian, Ubuntu, openSUSE, Debian, etc. In short, mileage may vary a bit depending on which build of Firefox that you run and what distribution you are running the. Not to mention the differences in hardware.

As the saying goes - if you want to do something right, do this yourself. Then let's look at the tools to test the performance of the browser in the privacy of your own home. In particular, we focus on three points of reference JavaScript, since that is where the real competition is today.

One of the oldest landmarks is the SunSpider JavaScript reference point. This test includes a variety of categories, some math, a chain of treatment, the creation of tagclouds, etc. Head just above the SunSpider page and click on "Start Sunspider 0.9.1 now" (or whatever the current version is when you check.)

When completed, Sunspider provide results something like this:

= Total results (means and 95% confidence intervals)-: 261 1ms ± 3.3%.

In fact, it will give much more than that - all the Sub-tests results will be display as well. You can watch and see not only what browser receives the best results, but the tests they do well or poorly in. You will also receive a URL, you can bookmark and revisit later to compare the results in a browser or in the same browser. Note that the lower numbers are better with SunSpider, this result (in Chrome) is very good.

SunSpider is the oldest of the benchmarks, but there have been a few questions about the tests that it uses and whether they are applicable to the actual speed.

People of chromium use the V8 benchmark Suite to address V8. As soon as you load the reference page, it starts running. With V8, you get a set of results, in a blue box on the right side as follows:

Note: 3748Richards: 6693DeltaBlue: 4390Crypto: 7368RayTrace: 2886EarleyBoyer: 3586RegExp: 1437Splay: 3226

Test results are broken down, and each test has a summary on the left side of the page. Save your results as V8 does not provide a linky practice to return to your results.

Finally, there is the most recent Mozilla reference point called Kraken. It is based on SunSpider, with improvements. The Kraken will take some time to download and run. As with SunSpider, lower numbers are better.

For best results, each testing bookmark, close all other Windows and apps and run each reference in its own window. You will likely skew the results by running the tests while making a compilation of the kernel while it analyzes a comparative browser and then then just fix running in a browser.

Run the tests several times each, as well. Does their execution once and then assuming that the number is the last word.

A thing are not benchmarks, tell you how well corresponds to a browser from Web pages and applications that you use on a daily basis. Perhaps Opera 11 blows the doors off a point of reference, but it sucks the exhaust pipe when it comes to Gmail. Perhaps Chrome comes last on a point of reference, but surprise arrives to do very well with Gmail.

Nothing compares to the real world test. If you have been unfortunate with Firefox performance prior to 4.0, give it another shot when 4.0 out instead of simply trust guides.

Don't forget that I said that this would be the participation of the audience?

Here's your weekend homework, if you're ready: close all your other applications and run the benchmarks in the latest version of Firefox 4.0, Chrome, Opera, etc. Provide your relevant system information and results of reference in the comments. What system information? Lucky that you requested!

Distribution and versionProcessorMemory

Thus, for example, I would post Linux Mint 10, Intel Core i7 Q720 1.60 GHz, 8 GB of RAM. Let us know if you use a 64-bit or 32-bit system. If you do not know what CPU you have, you can execute cat/proc/cpuinfo in terminal, or use one of the tools of GUI system information to know. If you have forgotten how much RAM you have, it is sufficient to run free-m for an index, or designate a GUI tool.

Sorry to give you assignments on weekend - but if even one of the twenty Linux.com readers runs tests and provides points of reference, we will have much to compare and a better image of the browser landscape. Happy benchmarking!

Comments (18)
You must be logged on to post a comment. Please register if you do not yet have an account.

View the original article here

This post was made using the Auto Blogging Software from WebMagnates.org This line will not appear when posts are made after activating the software to full version.