COMMENTARY--Intel should never have added Wi-Fi to its Centrino chipset. Now is not the time, and 802.11b is not the protocol.
Which is why I'm hard-pressed to recommend Centrino-based machines unless you plan to replace the built-in wireless with an 802.11g PC Card when the time comes.
I bring this up now because some of you may have read our recent coverage of the new chipset and/or seen reviews of some of the new notebooks (such as the Toshiba Satellite Pro M10-S405 or the Sony Electronics Vaio PCG-Z1) that use it and, as a result, may be considering a Centrino purchase. My recommendation: Beware!
If this makes me (as some of you have accused) "anti-Intel," so be it. It wouldn't be the first time that charge has been leveled at me. My best friend Rick, to whom I hardly ever speak, works at Intel. The reason we don't talk as much as we used to has nothing to do with his employment--at least that's what he says.
Rick has complained--though not terribly--that I'm always trashing the company he loves. Actually, I love Intel too, but I hold them to a very high standard--a standard to which (I believe) Intel wants to be held.
So when Intel decides to include wireless capabilities in its new chipset for mobile computers, I want to see Intel do it right. And with Centrino I don't think it has.
Why? Because wireless standards aren't fully in place. We're at the beginning of a transition from 802.11b, which runs at 11Mbps (if you're very, very lucky), to 802.11g, which theoretically runs at 54Mbps but remains compatible with 802.11b. The 802.11a standard, which is as speedy as 'g' but not downwardly compatible, seems to have missed its window of opportunity.
By building 802.11b into all Centrino machines, Intel seems to have also built in obsolescence. Sure, the Centrinos will be compatible with 802.11g--but they'll be slow, second-class machines.
Now, there's more to Centrino than built-in wireless. It also has improved power management and some other valuable features for road warriors. Nothing incredibly new, but still welcome. I'm sure Intel is hoping Centrino will kick-start notebook sales, something I'd welcome. And I admit that built-in wireless is very attractive, a key feature in getting customers to buy in this economy.
Does the extra speed you get with 802.11g really matter? In many cases, no. Consider the case of a home network: Your Net connection speed is limited by your DSL or cable modem, which typically runs at less than 2Mbps, often a lot less. Since 802.11b runs between 2 and 11Mbps, depending on conditions and locations, your wireless net won't constrain your Internet connection speed.
In a business environment, you may have access to a much faster Internet connection. But the amount of network bandwidth you can commandeer as a single user will be limited by a variety of factors; the speed you experience will depend more on your company's network infrastructure than on your wireless chipset.
Still, suppose you're doing video conferencing on your corporate network or someday expect to stream video around your home. Or maybe you just move lots of really big files. In each of these cases, your notebook's wireless hardware should be a significant gating factor. (I say "should" because your experience will vary based on other network factors.)
As wireless entertainment and video become commonplace over the next few years, the need for wireless speed will only accelerate. The 802.11b that may be acceptable today will be too slow someday soon. Given that 802.11g is sliding down in price, I expect that it's what we'll all be buying in six months.
To be fair, I think Microsoft was remiss for bringing out a line of 802.11b wireless hardware instead of using its stature (and buying power) to push 802.11g adoption. One response to such criticism has been that the final 802.11g standard doesn't yet exist. True enough. But when the standard does get ratified, it should be easy to make the necessary software upgrades. The same situation existed in the early days of 802.11b and didn't slow the adoption of that standard.
So just as Microsoft should have chosen 802.11g, I believe Intel should have built 802.11g into Centrino (the way Apple built it into its latest PowerBooks) or simply waited until the market adopted 802.11g. The 'g' standard provides enough speed for the foreseeable future and assures that slow wireless won't hold users back.
It's not like the wireless industry--or users--really need Intel or Microsoft. Heck, Intel has already been forced out of the wireless peripheral business. And Microsoft's involvement raises issues regarding unfair competition.
So do I hate Intel and by extension Microsoft? Certainly not. Intel, along with Microsoft, Apple, and perhaps a very few others are what I call "destiny companies," because what they do will shape our collective technological destiny in very important ways. As a result, they should live up to the very highest standards.
In the case of Intel's Centrino and, more generally, Intel and Microsoft's insistence on pushing 802.11b when 802.11g is right around the corner, I don't think either company has.
What do you think? Would you buy a laptop with built-in wireless? Are you using 802.11b now? Let us know at edit@zdnet.com.au.
