One of the most talked about aspects of the new version of Windows is it's apparent speedier operation compared to Windows Vista. Early reports, and our experience of the various beta and release candidate versions, certainly suggested it felt quicker, with faster boot up times and a generally more responsive interface. However, what we're going to find out today is if there truly is any fundamental difference.
Of course, many people have chosen not to bother with Windows Vista and have stuck with XP, which has proved to be a stable and speedy OS since it had its own teething problems when it launched way back in 2001. As such we'll be comparing all three OS's to see how things have changed over time. We've chosen not to use the 64-bit versions of the three operating systems as the vast majority of people will still be using 32-bit versions of XP and Vista. We appreciate the majority of PCs now shipping do include the 64-bit version of Vista and it will probably be the case for Windows 7 as well but we feel any performance conclusions drawn here will be just as applicable to the 64-bit versions. Also, with XP 64-bit having such a small user base and having limited driver support we were inclined to avoid testing on that anyway.
To keep our results as in-tune as possible with the hardware that we feel the majority of you will be using in your PCs at home, we've chosen to use a fairly modest test system in terms of cost, though the hardware itself is mostly quite new. The full system is detailed below.
- Intel Core i5 750 CPU
- Intel DP55KG Motherboard
- 2 x 1GB Qimonda IMSH1GU03A1F1C-10F PC3-8500 DDR3 RAM
- 150GB Western Digital Raptor Hard Drive
- AMD ATI Radeon HD 4770 Graphics Card
It's common, when benchmarking, to disable certain programs and services (things like Superfetch, and Windows Defender) on an operating system to ensure performance is consistent as possible when the test setup is used for a period of time. However, for this testing we wanted to have as pure a user experience as possible so for the most part we left the operating systems in the exact state they were when installed. The only thing we did was to disable User Access Control (UAC) in Vista and Windows 7 as some of the automated benchmarks have a bit of a hissy fit if they get interrupted by UAC warnings.
Being brand new, Windows 7 has no service packs or significant performance updates so we used the installation as it was when installed from the RTM disc. For Windows Vista we patched it two service pack 2 while XP was brought up to service pack 3. This most fairly reflects the true difference in performance and features you'll see moving from one operating system to another at this moment in time.
As well as looking at raw performance results like how long it takes to transcode a video clip or how many frames per second you can get in a game, we will also look at the subjective feel of the operating systems to verify whether all those first impressions were just exagerations or truly worth shouting about. So, without further ado, let's get testing.