Performance Testing

By Andrew Garcia  |  Posted 2009-08-17 Print this article Print

One of the biggest knocks against Windows Vista was that it was a resource hog. Although this abated somewhat as Vista evolved through Service Packs 1 and 2, the perception was common that the operating system used too much RAM by default and took too long to startup and (especially) to shut down. 

To see whether Microsoft gave Windows 7 a speed bump, I ran a series of benchmarks against the 32-bit and 64-bit iterations of both Windows 7 and Vista, using both the Dell laptop and Phenom II-based systems described earlier in this review.

With each installation of Windows 7 and Windows Vista, I relied almost exclusively on drivers included with the operating system install disk or via Microsoft Updates. The sole exception: I obtained drivers from Dell and for the video adapters on both machines. Tests were otherwise conducted on base installations, as no additional software (save the benchmark suite and its dependencies) or operating system patches were installed. 

Benchmarks were performed using the latest version of FutureMark's PCMark Vantage Professional suite, which performs a variety of tests on a system's ability to record, play back and edit various types of audio and video media, as well as its ability to process large amounts of text and to quickly render Websites--among many other modern usage scenarios.

I used the 32-bit version of PCMark Vantage to test the 32-bit Windows 7 and Vista SP2 iterations and the 64-bit version of the test suite to evaluate the 64-bit systems.

The PCMark score is an aggregate number, reflecting performance on a subset of the tests that make up each of the comprehensive test suites, including Memories, TV and Movies, Gaming, Music, Communications, Productivity, and HDD hard drive tests.

The breakdown of the complete results can be found in this slideshow.

Of the aggregate PCMark scores, Windows 7 scored better across the board. In 32-bit tests, Windows 7 showed a negligible 2.6 percent improvement over Vista SP2 on the laptop (Vista 3,603, Win7 3,698), but a whopping 15.1 percent improvement on the quad-core desktop (Vista 6,096, Win7 7,018). While further testing is necessary to bear this out, the disparity suggests that Windows 7 more efficiently utilizes 32-bit systems at the RAM ceiling for the architecture.

In the 64-bit tests, Windows 7 showed an 13.6 percent increase over Vista on the laptop (Vista 3,679, Win 7 4,183) and an 8.7 percent improvement on the desktop (Vista 6703, Win 7 7284). 

Senior Analyst Andrew Garcia can be reached at




Andrew cut his teeth as a systems administrator at the University of California, learning the ins and outs of server migration, Windows desktop management, Unix and Novell administration. After a tour of duty as a team leader for PC Magazine's Labs, Andrew turned to system integration - providing network, server, and desktop consulting services for small businesses throughout the Bay Area. With eWEEK Labs since 2003, Andrew concentrates on wireless networking technologies while moonlighting with Microsoft Windows, mobile devices and management, and unified communications. He produces product reviews, technology analysis and opinion pieces for, eWEEK magazine, and the Labs' Release Notes blog. Follow Andrew on Twitter at andrewrgarcia, or reach him by email at

Submit a Comment

Loading Comments...
Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel