The 802.11n Reality Check
February 21st, 2007Understand that the new, higher-speed Wi-Fi protocol, known as 80211.n, hasn’t been approved yet. Instead, what you have now are “draft” standards, which simply means it’s a work in progress, something akin to beta software. That hasn’t stopped companies from selling computers and routers that support the new technology. Yet maybe it would be akin to selling a new DVD player before a new standard is final. Of course, some of you might think that’s already happening what with those duelling high-definition disc standards.
In releasing its new AirPort Extreme, Apple promised speeds that were up to five times greater, twice the range and several significant new features, such as the ability to set up a USB hard drive as a network storage device. Cool.
The published test results have shown that the new “n” standard can be temperamental, however. There’s no assurance you’ll get even twice the speed of the existing 802.11g standard, let alone the higher figure claimed by Apple. Transmissions are carried on either of two channels, 2.4GHz, same as the older standards, or 5GHz. In theory, the higher frequency ought to get you greater range, or maybe not. Turning on the most secure WPA2 security feature may cause a drop in performance — or maybe not.
As I said, performance is not entirely predictable. In addition, if you have other devices using the same frequency bands, such as a cordless phone, it may cause interference. Thus, another potential source of speed problems.
Your location can also cause reception difficulties. Thick walls, with lots of metal, for example, may serve as a huge impediment to even being able to network with a computer or router in another room. So you have to consider other options to expand the range, such as an external antenna, or additional base stations, which act in the fashion of a cell phone tower, where your Mac or PC communicates with the nearest signal source.
Then there’s the question of interoperability with other items that allegedly support the “n” protocol, particularly if they are made by different manufacturers. As most of you know, mixing and matching existing Wi-Fi hardware usually works, except when it doesn’t. The best chance for compatibility is when the chipsets are built by the same makers, but how do you know? It’s not as if the parts list and the source is listed in the manual.
The first generation of “n” devices were particularly impacted by such incompatibilities, not to mention the inability to achieve those soaring wireless speeds that were promised.
When all is said and done, the real question is whether it’s all really worth the bother right now. Regular 802.11g hardware at least adheres to a final standard, not one in flux. Even if there are only minor changes when “n” is complete, will the changes cause more troubles or eliminate them? In theory, existing hardware, such as the new AirPort Extreme, should be easily upgradeable via a firmware update, but will your hard-won installation have to be reconfigured as a result? Maybe, maybe not.
A lot of it depends, of course, on what kind of computer you have now. And, yes, I realize a lot of you exist in both platforms, and not all of you prefer Macs. If your Wi-Fi hardware is already compliant with the “n” draft standard then it’s understandable that you might want to stretch the limits or at least use your existing potential to the fullest.
In addition, if you need a new router, I suppose it makes sense to want to get the latest and greatest, even if your computer isn’t compliant just yet. It won’t make things run any faster right now, but I suppose the modified or improved antenna structures might get you better range. At least it won’t be any worse.
In saying all this, let me lay my cards on the table. I don’t have any hardware that’s compliant with the new standard, so I’m not about to rush into a store and buy an AirPort Extreme or comparable device. My present router, a Buffalo Technology Wireless G, offers decent range and performance. At worst, it has to be restarted from time to time when my MacBook Pro can’t connect, but that’s not an unusual symptom with such gear.
I suppose, however, that I shall one day have new Macs, with the faster Wi-Fi chips. The final standard may even be approved by then, and then I’ll reconsider my options. As for you, gentle reader, I really don’t think you need to rush into things just yet.
Print This Post
Four quick notes:
1) The higher frequency is absolutely not going to provide greater range. That’s just a physic things. Lower frequencies go further. Think of it this way, your energy only gets you so many squiggles. 5 GHz squiggles twice as often per unit distance than 2.4 GHz squiggles, so it only goes half as far physically per unit energy. Obviously, this is way oversimplified. 😉
2) Barring a huge surprise, 802.11n really is all but finished. What remains to be done at this point is mostly just completing the political side of the IEEE process. I kid you not, I once spent an entire day at an IEEE standards meeting where debate raged as to which revision of Roberts Rules of Order we were using. And, believe it or not, it mattered in the context of IEEE standards. Anyway, 97.99% of the 802.11n body approved the current draft as Draft 2.0. Now Draft 2.0 goes through another round of bureaucracy. End result, final approval sometime around May, 2009. With no substantive changes between now and then that would require new hardware (otherwise *all* the hardware vendors would be screwed and the hardware vendors pack the standards bodies, so that’s not going to happen).
3) As I understand it, Apple choose to only use double-wide channels in the 5 GHz band (I’ve not yet confirmed this, I’ve only read it in reviews). I view that as somewhat unfortunate since 2.4 GHz has much better range and substantially better ability to penetrate walls and such. Anyhow, it’s the double-wide channels that allow the outrageously high throughput numbers (hundreds of megabits). So if you stick with 2.4 GHz (and normal 20 MHz channels), you’re not going to be able to get superfast speeds. Alas, with 5 GHz channels, you’re not going to penetrate walls well. So you’re mostly limited to superfast speeds in the same room. Which is why I think this decision was unfortunate. The MIMO element of 802.11n will give you better range at 2.4 GHz…but no 540 megabit data rates.
4) The new base station should have had Gigabit Ethernet. That was just a stupid decision on Apple’s part. After a decade of accumulating Firewire hard drives, it’d also have been nice to be able to plug those into the base station, but Firewire ports are sometimes pricey, so I can forgive that. And even if there’s no real performance hit, there’s a perception hit by only have 100 megabit Ethernet ports. So I’m sticking with that was just stupid.
reinharden
I appreciate your response and your extensive knowledge of the situation.
Yes, I agree with you about Gigabit Ethernet. This is an unfortunate decision, though I suppose it would have meant a $229 product instead of a $179 product.
Peace,
Gene
From the review on Macintouch, Apple’s new base station has disappointing results. Although I would like to point out that other vendors who are touting the same technology are a little better, but no-one is able to hit the maximum bandwidth limit for more than just a few seconds. I personally think that the marketing of technology these days has really no governing body to verify these claims. Hard drive formatted capacity, battery life, internet speed claims and now a new wireless speed limit that will never be reached. This time however, I don’t think that the consumer will let it slide since they might encounter problems streaming HD to their Apple TVs.
Hammer
That are several tyro screwups in the Macintouch test. Not the least of which was presuming you achieve the fastest speeds by having wifi devices as close together as possible.
The sole time I tried to point this out in detail at his site — the Commenting software was so screwed up that proved impossible.
Not especially confidence inspiring. I think the test from the MacWorld labs are much more capable.
[Comment ID #5823 Will Be Quoted Here]
All right, Ed, how about tell us what you feel are the best ways to achieve the fastest speeds in such a test?
Clearly, I would think that real-world tests, mimicking what most users are apt to do, is the best way to see how such things perform. Too much artificiality never inspires confidence either.
Peace,
Gene