Web Browser Grand Prix Setup
Test System Specs
64-bit Desktop Test System | |
---|---|
Operating System | Microsoft Windows 7 Ultimate (64-bit) |
Processor | Intel Core i5-750 @ 2.8 GHz (Quad Core) |
Motherboard | Gigabyte GA-P55A-UD7 (F7 BIOS) |
Memory | 8 GB Crucial DDR3 @ 1333 MT/s (2 x 4 GB) |
Graphics | AMD Radeon HD 4870 Reference Board 512 MB GDDR5 (PCEe 2.0) |
Storage | Seagate Barracuda 7200.12 500 GB SATA 3Gb/s, 7200 RPM |
Optical | Asus DRW-241ST/BLK/B/AS |
Power Supply | Corsair TX750W (750 W) |
Chassis | Zalman MS1000-HS2 |
CPU Cooler | Scythe Mugen 2 Revision B |
Starting with this article, we're adding a battery life benchmark to the WBGP. Obviously, the desktop test system doesn't work for this type of metric, so we pulled out the old first-gen Dell Inspiron Mini 10v for this new test.
32-bit Netbook Test System | |
---|---|
Operating System | Microsoft Windows 7 Ultimate (32-bit) |
Model | Dell Inspiron Mini 10v (first-generation) |
Processor | Intel Atom N270 @ 1.6 GHz |
Memory | 1 GB DDR2-533 |
Graphics | Intel Graphics Media Accelerator (GMA) 950 |
Storage | 120 GB HDD 2.5-inch 5400 RPM SATA 3Gb/s |
Both our page load time tests and the new Facebook JSGameBench are server/client benchmarks and require a local Web server. For this, we polished up one of our finest old beige boxen, installed Ubuntu 10.04 LTS Server, and connected it to our LAN.
Local Server Specs | |
---|---|
Operating System | Ubuntu 10.04 LTS Lucid Lynx Server Edition (32-bit) |
Processor | AMD Athlon @ 1150 MHZ |
Motherboard | Soyo Dragon Platinum |
Memory | 512 MB DDR |
Graphics | AMD Radeon 9550 256 MB DDR |
Storage | 40 GB Western Digital HDD WD400BB |
Optical | Samsung DVD-ROM SD-616T |
Additional Packages | Apache2, MySQL Client, MySQL Server, PHP5, PHP-GD, PHP5-MySQL, PHPMyAdmin, SSH, node.jes, NPM, socket.io |
Setup
Before testing, Windows 7 was fully updated to Service Pack 1 on 6/20/11 for the 64-bit desktop test system, and on 6/25/2011 for the 32-bit netbook test system. On both test systems, power management was set to never engage and all Action Center notifications were disabled. The table below lists all the additional software installed onto the test systems, including drivers, plug-ins, and the Web browsers:
Additional Software | Version |
---|---|
AMD Catalyst | 11.6 |
Adobe Flash | 10.3.181.26 |
Microsoft Silverlight | 4.0.60531.0 |
Oracle Java | 6.0.260 |
Apple Safari | 5.05 (7533.21.1) |
Google Chrome | 12.0.742.112 |
Microsoft Internet Explorer | 9.0.8112.16421 |
Mozilla Firefox | 5.0 |
Opera | 11.50 (build 1074) |
Methodology
The computer was restarted and allowed to warm-up before benchmarking the next browser. Other than the conformance benchmarks, all of our final scores are an average of several iterations. More iterations were run on tests that have short durations, lower scales, and/or higher variations. The table below lists all of the 37 benchmarks in the current WBGP test suite, along with the number of iterations performed for each test.
Test | Iterations |
---|---|
Performance Benchmarks (27) | |
Startup Time (1 Tab and 8 Tabs) | 5 |
Page Load Time (Google, Facebook, YouTube, Yahoo!, MSN) | 5 |
Peacekeeper | 3 |
Kraken v1.1 | 3 |
SunSpider v0.9.1 | 3 |
Dromaeo DOM | 3 |
Maze Solver | 3 |
Flash Benchmark 2008 v1.09.1 | 3 |
GUIMark2 Flash (Vector Charting, Bitmap Gaming, Text Columns) | 5 |
GUIMark Java 5 Swing | 5 |
Encog Silverlight Benchmark | 5 |
Facebook JSGameBench v4.1 | 3 |
GUIMark2 HTML5 (Vector Charting 1 pixel, Bitmap Gaming, Text Columns) | 5 |
Asteroids HTML5 Canvas And JavaScript 2D | 3 |
Psychedelic Browsing | 3 |
Hardware Acceleration Stress Test | 3 |
Khronos Particles | 5 |
WebGL Aquarium | 5 |
Efficiency Benchmarks (5) | |
Memory Usage (1 Tab, 40 Tabs) | 3 |
Memory Management (-39 Tabs 5 minutes, -39 Tabs 10 minutes) | 3 |
Battery Life | 3 |
Reliability Benchmarks (1) | |
Proper Page Loads | 3 |
Conformance Benchmarks (4) | |
Sputnik v2 | 1 |
Acid3 | 1 |
CSS3 Selectors Test | 1 |
HTML5Test.com | 1 |
Additional details about a specific benchmark can be found on the corresponding benchmark results page.