Our test hardware underwent some significant changes since Web Browser Grand Prix: Chrome 25, Firefox 19, And IE10. We're now on an Ivy Bridge-based Core i5 rig, upgraded from the equivalent Sandy Bridge parts. Our cable modem and ISP speeds are also better than last time. For now, the Web server and router remain the same, though both should be new by the next time we do one of these stories.
Our benchmark suite is also changing. First, that 40-tab workload in our memory tests is now timed using a stopwatch, and contributing to our overall wait time score (start-up and page load times).
Moving on to JavaScript, after we retired SunSpider from the test suite, WebKit updated the long-believed-abandoned JS performance test. We'll be running the newly-released SunSpider v1.0 to see if this benchmark's issues with Internet Explorer have been resolved.
KaizouMark makes a return now that IE9 is no longer in the mix, finally providing us with a long-overdue CSS3 performance test.
The final massive change to the benchmark line-up is in Hardware Acceleration. Both native HTML5 HWA and WebGL swapped out an old test for a newer one. First, we say goodbye to Psychedelic Browsing, and hello to CanvasMark 2013. Next, Airtight Interactive's WebGL Demo gets benched in favor of LUIC Cubes, a more intense metric that should be useful longer.
Test Setup And Benchmark Suite
Test System Specs | |
|---|---|
| Operating System | ![]() Microsoft Windows 8 Enterprise (64-bit) ![]() |
| Processor | Intel Core i5-3570K @ 4.2 GHz (quad-core) |
| Motherboard | Gigabyte GA-Z77X-UD5H (rev 1.0, F14 BIOS) |
| Memory | 16 GB Crucial DDR3 @ 1600 MT/s (4 x 4 GB) |
| Graphics | Gigabyte GeForce GTX 660 Ti 1 GB GDDR5 (PCIe 2.0 x16) |
| Storage | Seagate Barracuda 7200.12 500 GB SATA II 3Gb/s, 7200 RPM, 16 MB Cache |
| Optical | Asus DRW-24B1ST/BLK/B/AS |
| Power Supply | Corsair TX750W (750 W max) |
| Case | Zalman MS-800 Plus |
| CPU Cooler | NZXT Kraken X60 (closed-loop liquid cooler) |
| Monitor | AOC E2752Vh 27-inch LED (1920x1080) |
| Keyboard | Logitech Wireless Keyboard K320 |
| Mouse | Logitech Wireless Trackball M570 |
Local Web Server Specs | |
| Operating System | Ubuntu 12.04 LTS Server Edition "Precise Pangolin" (32-bit) |
| Processor | Intel Pentium 4 @ 2.41 GHz |
| Motherboard | Biostar P4M80-M4 |
| Memory | 768 MB DDR @ 333 MT/s |
| Storage | Western Digital Caviar SE WD1600AAJD, 160 GB EIDE, 7200 RPM |
| Extra Packages | Apache2, MySQL Client, MySQL Server, PHP5, PHP-GD, PHP5-MySQL, PHPMyAdmin, SSH, Node.js, NPM |
Network Specs | |
| ISP Service | Cox Preferred (18 Mb/s down, 2 Mb/s up) |
| Modem | Cisco Telephony Modem DPQ3212 (DOCSIS 3) |
| Router | Linksys WRT54G2 V1 |
Benchmark Suite | |
| Startup Time | Cold Start Time (Google SERP, Cached) |
| Hot Start Time (Google SERP, Cached) | |
| Cold Start Time (Eight Tabs, Cached) | |
| Hot Start Time (Eight Tabs, Cached) | |
| Page Load Time | EEMBC BrowsingBench |
| 40-Tab Load Time (Top 40 Websites) | |
| JavaScript | RIABench JavaScript (Eight Tests) |
| Futuremark Peacekeeper v2.0 | |
| Rightware Browsermark v2.0 | |
| JSBench | |
| WebKit SunSpider v1.0 | |
| DOM | Mozilla Dromaeo DOM (Core) |
| HTML5 | Principled Technologies WebXPRT CP1 |
| Impact HTML5 Benchmark | |
| Hardware Acceleration | Facebook JSGameBench v0.4.1 |
| HTML5 HWA | WebVizBench |
| CanvasMark 2013 | |
| WebGL | LUIC Cubes |
| Scirra WebGL Performance Test | |
| Memory Efficiency | Memory Usage (Single Tab) |
| Memory Usage (40 Tabs) | |
| Memory Management (-39 Tabs) | |
| Reliability | Proper Page Loads |
| Security | Browserscope Security |
| Standards Conformance | HTML5Test.com |
| The CSS3 Test | |
| Ecmascript Language test262 | |
While applicable links are included in the table above, we also have a public delicious account dedicated to Web Browser Grand Prix benchmark links. Detailed methodologies are explained on the individual benchmark pages.
We've highlighted Opera Next in the charts to emphasize that it's still in development and not part of the usual WBGP line-up.
- Opera: Has The Fat Lady Sung?
- Chrome, Firefox, IE, Opera
- Test Setup And Benchmark Suite
- Wait Times: Start-Up
- Wait Times: Page Load
- JavaScript And DOM Performance
- HTML5 And CSS3 Performance
- Hardware Acceleration Performance
- Memory Efficiency
- Reliability And Security
- Standards Conformance
- The WBGP XVI Winner's Circle


While this is interesting, I still encounter built-in pages (such as on routers or other network devices) that will not render cleanly in Firefox, but are perfect in IE. More often than not though, pages that would be filled with nuisance ads and popups are cleaned up nicely by Firefox with AdBlock+ and NoScript.
I have both and start up times for IE are quick but page load times are horrendously slow, whereas FF has slow start up times but superfast page load times. It's possible that add-ons are contributing to that.
1) Pages load noticeably slower
2) Memory usage is indeed high (as seen in the benchmark above)
3) FF add-ons are much better than chrome extensions.
I never noticed any startup time difference for both FF and Chrome; it's possible they're both fast enough that it doesn't even matter at this point. I also like the FF toolbars better although that's really more of a personal preference. I've never tried maxthon though; heard it's pretty good.
I'll miss a hell of a lot of stuff when I move off Presto-based Opera.
Still, this test shows us once more, that no modern browser - I exclude Opera from this, since it isn't a maintained release anymore - must absolutely be replaced by the winner of such tests. If you don't mind performance weaknesses of the Internet Explorer in certain areas, or if your most-accessed websites don't require you to use a certain alternative, then even Microsoft's browser of choice can be okay for daily use (if only as an engine in products like Avant, Maxthon, etc).
The one thing I'm a bit curious about: why does Opera Next suddenly behave so differently from Chrome? Yes, there's a difference between Chrome 27 (WebKit) and Opera Next (Blink = Chrome 28), but if that's the only reason for the browser's weaker showing, then the future of Chrome doesn't look too good. What's your take on this?
I'm guessing it should have said i5-3570K.
Can I make a request for your next test? Try comparing SSDs and HDDs in some of the tests, especially cold boot. They are becoming more and more popular, and at least with some of these tests, I imagine they do have a significant impact on performance.
When the Chromium-based version becomes stable, it will be called Opera.
I love customizability and plug-in support of Firefox. I also love the separate address and search bars. If I type "IBM" into the address bar I want to go directly to the website, if I want to search then I'll use the search bar. Unfortunately, whenever I install Firefox, I have to re-enable the "go directly to the website" ability in the address bar because, starting with Firefox 4, some dork at Mozilla changed the functionality of the address bar.
Up/down keys are your friend.
One other pet peeve I have is that the autopredict in browsers have a habit of interpreting '192.168.1.1' as '192.168.1.104', if you visit the latter more frequently. Someone disable autopredict for IP addresses, and I will use your browser.