Windows 7 Performance Guide

Conclusion

With the launch of Windows 7, there are really 3 different tales to tell. There’s Windows 7, the OS that replaces Vista. There’s Windows 7, the OS that needs to do what Vista failed to do and kill XP. And there’s Windows 7, the OS that needs to put a stop to Apple’s continuing growth. We’re going to look at all 3.

7 vs. Vista

I’ve never been entirely convinced that Microsoft was looking to significantly move away from Vista with Windows 7, and the final release version has not changed that. Win7 certainly has a number of new features, some of them such as the interface overhaul are even big enough to be classified as “major”, but none of them are important enough to be significant.

The analogy I’d like to use here to call Windows 7 Vista’s XP, but even there the change from 2000 to XP was more significant. There are a number of edge cases where this isn’t the case, but overall, in the general case, Windows 7 just isn’t a significant change from Windows Vista. The inclusion of more audio/video codecs is the only improvement that I think most users are going to encounter and benefit from.

Now as for the edge cases:

HTPC: If you’re using Windows Media Center to drive an HTPC, the changes to WMC are significant enough to justify an upgrade, particularly if you’re a cable TV user.

Low-End Hardware: We’ve seen this one ourselves – Win7 does much better here on marginal hardware. The only catch is that on such hardware the computer probably isn’t worth much more than the upgrade copy of Windows. Certainly Win7 is a better fit, but so is completely replacing such hardware.

Laptops: Windows 7 has better battery life than Vista, resumes from hibernation sooner, and given the lower performance of laptops often benefits from the better performance of Win7 on such hardware. If you need to squeeze out every minute of battery life or every point of performance, then it’s upgrade time. In fact laptops users are certainly going to be the easiest group to sell upgrades to, since Win7 consistently does so well.

Ultimately the issue with upgrading anything else is the price of a Windows 7 upgrade. It’s $110. So was Vista. And Vista was a much more profound change than Win7 is, bringing UAC, DirectX 10, Aero, and the other big features that are still prominent with Win7 today. Win7 isn’t a big enough upgrade for most Vista users given the price. If Microsoft did something similar to Apple and charged $30 or so for it, then it would become cheap enough to justify an upgrade even with the minor differences, but unless you’re eligible for a student upgrade ($30), then that isn’t going to happen. The closest Microsoft has come to that is the $50 pre-order sale, and the ship already sailed on that months ago.

In some degree of fairness, Microsoft is between a rock and a hard place here. XP is also a valid OS to upgrade from, and $30 to go from XP to Win7 would be ridiculously cheap. Not that we’d complain, but it’s not realistic. Microsoft’s other option would be to have different upgrade editions for XP and Vista, and while this is more reasonable the confusion it would cause would probably not be worth it.

With all of that said, this applies just to upgrades. The bottom line is that unlike the XP to Vista transition, I can’t fathom any good reason why anyone using a computer bigger than a netbook would want to stick to Vista. Win7 runs well and we haven’t encountered any software or hardware compatibility problems. Meanwhile it doesn’t bring with it any pitfalls like Vista did, so using Win7 on a new computer as opposed to Vista is the closest thing to a no-brainer that we’re going to see today. If you wanted Vista, you’re going to want Windows 7 instead.

7 vs. XP

If Windows 7 wasn’t meant to light a fire under Vista users, then it’s XP users that are the target. Microsoft couldn’t get them to move to Vista, so this is their second and possibly last real chance to move them before they become their own permanent faction of die-hard users. And to be frank, if Win7 succeeds here, it’s not going to be because of technical measures.

The vast majority of big improvements in Windows came with Vista, not with Win7. Microsoft did fix some edge cases for Win7 such as marginal performance and laptop battery life, but the dissent over Vista went far beyond those edge cases. If your hardware didn’t work under Vista, it still won’t under Win7. If you didn’t like UAC, you still won’t like it under Win7. If you found XP to be snappier than Vista even on a fast computer, then you’ll still find XP to be snappier than Win7. At best, if your software didn’t work under Vista, it might work under Win7 if you can put up with the Windows XP Mode virtual machine.

So if Win7 succeeds where Vista failed, it’s going to be because of marketing and word of mouth. It will be Microsoft convincing users that Win7 is great before anyone can convince them otherwise, because if that negative mindset were to set in, it can’t be erased no matter how good any of the service packs are. Vista had its problems, but what kept it down since SP1 was word of mouth much more than it was technical issues for edge cases.

The one exception to this is netbook users, and as we didn’t get a chance to test any netbooks, we’re not going to make any judgments. If Microsoft has Win7 to the point where it performs comfortably on the average netbook, then they’re going to be in a much better position than if it crawls like Vista. In which case the bigger problem will be weaning OEMs off of cheap XP licenses.

On a proper system, Vista has always been the better choice for Windows. So our recommendation isn’t really changing here. If you’re not on Windows Vista or Windows 7, you should be. XP has been outdated for quite some time.

7 vs. Snow Leopard

We’ll have a proper review of Snow Leopard in the near future, but for now we’ll talk about Snow Leopard as compared to Windows 7.

Snow Leopard (10.6) was a minor release as compared to Leopard (10.5), much like Win7 is compared to Vista. Apple was able to go ahead and charge $30 for it for an upgrade from Leopard while charging $130 for an upgrade from Tiger (10.4), which means that if nothing else, Apple has been able to avoid Microsoft’s pricing problems.

When comparing Windows and Mac OS X, Apple’s strength has been integration and the GUI. Microsoft can’t do anything about the former, but they have about the latter. Apple still has the better GUI, but the advantage is not quite as great as it used to be.

Without getting in to hardware, Snow Leopard has been even more stagnant than Windows has. Win7 brings some definite advantages over Snow Leopard: per-application volume controls, a wide audio/video codec selection (Apple’s the odd man out here; even Linux has them beat), SuperFetch, TRIM support, and of course all the applications Windows can run. Meanwhile Snow Leopard has its GUI, along with Apple’s gesture system, Exposé, and Time Machine.

Ultimately it’s either that the two aren’t different enough, or that they’re so different that they’re hard to compare. In either case Mac OS X doesn’t have the obvious advantage that it once had against Vista. Windows 7 has brought Windows to the point where it’s going to be Mac OS X’s peer in most cases, and right now it looks like it’s going to skip the teething issues that Snow Leopard is going through. For the time being, it’s going to be hardware that’s the real differentiator.

7 vs. Linux?

This late October timeframe also aligns with the 6-month cycle of Ubuntu Linux, and is close to several other Linux distributions. We haven’t had to a chance to see Ubuntu 9.10 yet, but it’s something to keep in mind. Win7 erodes the Linux advantage against Windows in the performance cases where Vista suffered, and Win7 is going to widen the GUI disparity some, but otherwise Win7 is much of the same. This can only be an advantage for Linux vendors, who get another 2-5 years to chase a very similar target to the one they’ve already been chasing for the last 3.