Browser Benchmark Test Results
This assignment brings us a request to select a series of seven benchmark tasks which may be important in selecting a browser. Each of these benchmark tasks were chosen as something many people require when selecting a browser. Admittedly, with the proliferation of Microsoft's Internet Explorer, the default browser for the unwitting remains Internet Explorer. And depending on the skill set of the novice, the variety of Internet Explorer is typically version 6.0 or if the user is adept at visiting the Microsoft download page, version 7 can be easily downloaded. However, for many of us, searching for something other than Microsoft products has many in a quandry about which browser is truly the best. Best is defined individually, so one would expect that performance and feel play a major role in one's selection of an Internet browser.
Below are the benchmark tasks and the results I have achieved in various browsers. I have not tested any of these browsers on a Macintosh. The computer itself of course plays a major role in speed and overall performance. To remove the added dilemma of varying Internet Service Providers speed, all files were placed and called from the local hard disk. The browsers I chose are listed below and this resulting report was called upon in Internet Explorer version 6 , Opera 9.02, Firefox 1.5.0.7, Mozilla 1.7.13 and Maxthon 1.5.7. The tasks I have chosen to benchmark for these browsers are Rendering CSS, Rendering Images, Rendering Tables, Acid 2, History, W3C DOM versus innerHTML and Java Scripting.
Fastest Browser Benchmarks - Windows XP |
|||||
(Time in Milliseconds) |
|||||
* |
IE6 |
FireFox 1.5.0.7 |
Mozilla 1.7.13 |
Opera 9.02 |
Maxthon |
CSS |
1457 | 2372 | 4457 | 1154 | 1473 |
Images |
2276 | 5523 | 4242 | 1996 | 2519 |
Tables |
1986 | 4898 | 4926 | 2703 | 2098 |
JavaScripts |
10672 | 9059 | 13020 | 4624 | 12556 |
Acid 2 * |
Did Not Pass | Did Not Pass | Did Not Pass | Pass | Did Not Pass |
History * |
32 | 38 | 40 | 8 | 32 |
W3C DOM * |
67 | 82 | 116 | 20 | 73 |
* Notes below in the section labels of each category.
How I performed the tests and the rules for each test are outlined below.
Rendering CSS. To test CSS rendering speed I used a CSS benchmarking tool designed by nontroppo. The test measures the time it takes the browser to render a page consisting of almost 2500 positioned DIVs. The page is stored locally, loaded once to pre-load it, then reloaded 5 times, and the average time taken for those three renderings. The page is the first page loaded after starting the browser (after logging out and in).
Rendering Tables. To test table rendering speed I load a local copy of the javascript libraries page from here. I ignored the first load and then go back and click the link again (ensuring the page and all associated files are cached). I then use a script to time how long it takes to render the page. I take the average of five loads.
Rendering Images. To test image rendering speed, I loaded a local copy of a page with six different images. I opened the page, closed it then reloaded the page. I used a script to time how long it takes to render the page. I averaged five loads.
Javascripts. To test how fast the scripts complete after opening the page, I used precreated java from a previous class, inserted into an HTML page. I loaded the page, then closed the page. I then used a script to time how long it took to load the scripts. I took an average of five tests for the results.
Acid 2. This test was designed so that it is valid only with the browser's default settings. Changing items such as font sizes, zoom levels, applying user stylesheets, etc can break teh display of the test. A browser is only considered passing or compliant if a smiley face is rendered in the browser, with the nose of the face turning blue when moused over. Current data indicates only five browsers have passed the test. To be certain, this test is not about a smiley face, but rather how your browser handles bad code. The face is simply an indication of how your browser deals with bad code. Opera 9.02 was the only browser in my test to have passed. Test your browser now.
History. Something which may seem unimportant to most people when considering browser speed is history. In reality, history is quite important to browsing speed. For this test, I used Google image search for silver bells. I clicked through the result pages from 1 through 25. At this point, I started timing the test, clicking back as quickly as possible to the first result page, then back again through to the end. Each page in between, was allowed to load in its entirety, in particular the images, before clicking forward to the next page. I used the keyboard shortcuts, as this is typically faster than a mouse click. This test is designed to test the efficiency of the browsers caching feature.
W3C DOM versus innerHTML. This test is intended to find out which method of generating large amounts of content is fastest in teh browsers. The results do differ slightly from browser to browser, but this is expected. Each of the scripts created a 50 x 50 table, all of whose cells are filled with one character, an asterisk. The time the browser needs is measured and compared to the time it needed to execute the other methods. Each test was performed five times, then used an index value to judge a relative speed of the various methods. An important note here is that innerHTML is faster in all browsers than the W3C DOM methods.
Summary and Conclusions
While there are obvious differences in speed and performance, there are several things to note. The least of which is the rendering engine used by individual browsers. Opera 9.02 uses the Presto rendering engine. It is also an important note Opera is the ONLY browser using this engine. IE 6 and Maxthon are using the Trident rendering engine and what makes Firefox and Mozilla seem faster you ask? Its the use of the Gecko rendering engine. The trident engine, although faster, is not optimally designed for rendering web pages. This is why FireFox and Mozilla both appear to be faster than IE 6. Microsoft did improve itself in IE 7, but it is still using the trident engine, and it is still not optimal for web browsing. Another benchmark which can be used and influences browser behavior is the SVG Rendering comparison. More importantly, your computer hardware and memory is important when trying to determine what works best for you. However, it is clear tha Opera has developed the fastest browser for Windows based machines. And while it's scripting is slowest scripting has an index of 450, it is still faster than Mozilla's fastest script It also offers a high level of standards support, security and features. Firefox, is only faster than IE when it comes to scripting, but for support and scripting and features, it does appear to be a better choice. The one drawback I have found, is that not all secure pages are being designed for multiple "platforms" of browsing. For instance, the HR management site my company uses for timesheets and benefits was designed solely for IE. Again, that blasted Microsoft. For all of the above results and the reasons to get away from browser insecurity, Opera is the choice for speed and results. Your time, my time is worth the effort of downloading Opera for free.
