Indexing updates could be retrofitted to Windows XP, and Vista turned out to be more hungry for resources than previous Windows versions. Windows 7, however, has been out for a few weeks, and feedback has generally been great. We decided to revisit our first look at testing these two operating systems head-to-head and directly compare Windows Vista and Windows 7, to see if the reported benefits are really noticeable.

What is Performance at the OS Level?

Most people think of application performance when they talk about performance in general. However, the operating system plays a major role in the process chain that creates everyday computing experiences. The OS is what switches a processor between power-saving and fast operating states (or even intermediate active states that determine performance in Intel’s Turbo Boost mode via P-states). The OS, or to be more precise its dispatcher, distributes threads across available processor resources, and Windows 7 is more aggressive about using thread headroom for the sake of performance.

However, performance also has to be defined by the user experience during ordinary operating system actions, such as system bootup, standby, application launching, hibernation, or shutdown. If the OS were a bit quicker on applications but consumed several minutes during start-up or shutdown, you’d probably want to ditch it. Hence our second round of testing not only includes benchmarks and application tests but also a look at these everyday operations.

The 550+ Windows 7 early adopters who took our survey mostly echo the positive response that the upgrade has received from professional reviewers, pundits, and users of pre-release editions. A sizable majority say they’re extremely satisfied with the OS and rate it as a clear improvement on both the beloved Windows XP and the widely-panned Windows Vista. Crippling installation problems-the bane of every upgrader’s existence, and always a legitimate reason to postpone switching OSes-were rare.

Our full report follows. But first, some quick facts on the folks who took our survey:

• Prior to using Windows 7, forty-six percent of respondents ran Windows Vista, and thirty-two percent ran Windows XP. Seventeen percent ran Vista and XP about equally, five percent ran an OS other than Windows, and a whopping 0.7 percent used a version of Windows other than Vista or XP.
• Seventy-three percent upgraded an existing PC to the final version of Windows 7, and eight percent are running it on a PC that came with the OS pre-installed. Fourteen percent are still using a pre-release version, and six percent are running it on a Mac via Boot Camp or a virtualization program.
• Sixty-four percent of respondents rate themselves as expert Windows users; thirty-five percent say they’re intermediate ones. Less than one percent call themselves beginners.
• Sixty-one percent are using Windows 7 entirely or mostly for home/personal use. Twenty-five percent are using it about equally for home/personal and business use. Just fourteen percent are using it entirely or mostly for business use.
• Sixty-one percent are using a 64-bit edition of Windows 7, and thirty-one percent are running a 32-bit version.
• Eighty-two percent did a “clean” install of the OS from scratch; nineteen percent installed it over Windows Vista.
• Fifty-nine percent say they’ve used Windows 7 extensively, and thirty-six percent say they’ve done so a fair amount. Five percent say they’ve just used it a little so far.

It’s important to note that our goal wasn’t to survey a representative, projectable, normalized sampling of all Windows 7 users. The responses that follow are from members of the Technologizer community who chose to take our survey. Their opinions are their own-but we think they make for interesting reading even if they’re not the last word on how average users will react to Windows 7.

Getting Up and Running

As with all things, first impressions count with operating systems. And if you’ve bought an OS as an upgrade, the first impressions it makes come in the form of the installation process.

As I’ve written before, there’s no such thing as an operating system upgrade that doesn’t cause headaches for some percentage of the people who install it. In the case of our survey takers who installed Windows 7, however, the number who encountered major hassles was very small-presumably in part because Windows 7 is so similar to Vista under the surface. The fact that the vast majority of respondents performed clean installations rather than installing on top of Vista surely helped, too.

Eighty-four percent say the process went off without meaningful hiccups; thirteen percent say it went fairly well. A total of three percent reported major problems, two-thirds of which were resolvable. Pretty impressive-when PC World surveyed Windows XP users shortly after that OS shipped, half reported installation difficulties.

The fact that respondents’ upgrades tended to go smoothly doesn’t mean that they didn’t encounter any issues with their new operating system. Two problem areas stand out: More than forty percent had to try and resolve driver issues, and more than a third needed to deal with software incompatibilities. Other problems, however, were reported by a much smaller percentage of respondents. For instance, eleven percent reported crashes or blue screens of death. And only six percent said that Windows 7′s performance was poor, which is a relief given that the original version Windows Vista quickly developed a reputation as a poky resource hog.

Are the percentages of users who reported problems impressively low, or unsettlingly high? That’s subject to debate. But here’s something that isn’t: Both Windows XP and Windows Vista also suffer to some degree from all the gotchas that respondents said they encountered in Windows 7.

(In this infographic and those that follow, the scale of 0% to 100% represents the percentage of survey respondents who answered a particular question as indicated by the bars.)