Advantages/Disadvantages of of 802.11g and 802.11a

[Mltsao]

nt

 

[StraightPipe]

i'm no wireless expert (or even a user of wireless)

but I'm pretty sure G gets better range and transfer speeds

 

[cmetz]

802.11a is on a bigger and less used frequency band.

 

[JackMDS]

Both 802.11a and 802.11g yield a bandwidth of about 20Mb/sec.

802.11g is much more prevalent.

It is downward compatible with 801.11b and it has slight distance advantage as compare to 801.11a.

Using 5GHz band 801.11a is less exposed to signal interferences.

I would use 801.11a when the distance is short, no need for compatibility with 801.11b, and a need for security is very important (The Wireless Bandits usually do not move around with 801.11a cards.)

Otherwise 801.11a has no future.

 

[Carp1812]

Agreed. I purchased 802.11a because I live in a house with a LOT of electrical interference. We have 7 phone lines and upwards of 10 different cordless phones. There are 8 TVs, 5 microwaves, 15 computers, etc, etc. The downside to my 802.11a connection is that it doesn't permeate very well at all. If I'm more than 1 room away from my access point, my signal strength is almost nil. Fortunately for me, I rarely have a need for such distance. If you do need distance, I'd go with 802.11g. As others have said, it's far more popular and will likely last a lot longer than 802.11a.

 

[LordJezo]

My $.02..

Had a b network in my house for a few days.. and it stunk. Speeds were bad, connections got dropped, pings were high.

Went to g and glad I did. It is a bit more expensive, but, you can hardly even tell it is a wireless connection. I get full speeds all the time, very low ping times, and connections hardly ever drop. Range is improved and all the places that got a low signal on b get perfect connection on g.

 

[cmetz]

Clearly, 802.11g is getting the mass market acceptance. I expect g to win for end stations. This might actually work out for a oddly enough, giving it the niche for network infrastructure applications (e.g., building to building, or leaf AP to core AP) while g is used for end stations. I don't think the future for a is as bleak as JackMDS's "no future" but I agree that g appears solidly poised to win on volume.

 

[JackMDS]

Originally posted by: cmetz I don't think the future for a is as bleak as JackMDS's "no future" but I agree that g appears solidly poised to win on volume.

LOL, I used a little too strong language.

However, while the next generation of ?g? i.e. 802.11i is in the works, and the planning for the post ?i? started (The post ?i? probably 2-3 years away will be really nice.)

I am not aware of the next level of ?a? i.e. using the 5GHz band.

 

[Storm]

What g routers do you guys have?

 

[cmetz]

JackMDS, I could be wrong not being an RF person, but I think that modulation/encoding and frequency band mostly are independent variables, so whatever is developed for a 2.4GHz post-G could pretty quickly be turned into a 5.2GHz/5.8GHz solution also. I'd be shocked if there wasn't at least a niche market for that, simply because there are folks for whom cost and compatibility isn't an issue but they want it to be fast and good signal quality. Getting out of the congested 2.4GHz band would be worth it to them.

 

[JackMDS]

The market is dominated by what sells in the business and cooperate arena. (It is the simple force of numbers).

They want compatibility; a lot of them have 802.11b that are not going to be upgrade very soon, and a/b/g combos are very expensive.

I guess that folks for whom cost and compatibility isn't an issue, are not in numbers that justify further development

Since the price of 802.11b is very low I know that some businesses just buy more of the Cheappos 802.11b and make denser grids to compensate for the interference.

 

[cmetz]

JackMDS, I'm interested in your last comment because what you're seeing doesn't match what I'm seeing. I see a ton of SOHO traction - folks love 802.11b at home because it lets them share DSL/Cable when they don't have wiring, and these days it's reasonably cheap and easy to set up. But mid / large enterprises I see don't have much traction because 802.11's still got some pretty serious warts from an enterprise perspective (manageability, security, roaming, reliability etc.) IT managers who have 802.11b at home and know a good bit about it are frankly kinda scared of it in the office (those who don't know about the issues with wireless are reckless and often have to learn the hard way... kinda like those IT managers who never patch their Windows boxes). Hence the current wave of new product introductions for enterprise wireless platforms which try to extract a hefty price premium for addressing those issues. 802.11 certainly started out as a business solution (I used WaveLAN and then 802.11DS back in the day), but the volume and cost driver I see is SOHO, and only now is it getting driven back into business as folks go into work and wonder why they can't have the same mobile convenience as they have at home (that, and there's a big notebook / mobile push in business, and wireless networking is pretty much a must for that to be viable).

Compatibility is generally a good engineering and IT practice. I can't fault folks at all for wanting g for its compatibility with the existing base with b - it makes a lot of sense. However, a lot of folks are finding that the performance of a mixed network looks a lot more like b's performance than g's, so it may end up that compatibility is less useful in practice than in theory. Or it could end up that nobody cares, or that they find out after they've already bought into g

Development priorities certainly do follow market share. Right now, everybody's chasing g because g has so much more share already than a and g is expected to grow wildly while a is expected to grow slowly. If I were a vendor, though, I might weigh the counter argument that the a market has fewer players and less interest and could be a nice high margin niche, while the g market is going to be absolutely cut-throat.

I just don't see the picture for 802.11a as being quite as bleak as you do. It's not looking like a is going to make it in the mass market / end system market - g hasn't been out all that long and already it looks like g has just won there. But I think that a might still have a lot left in it, particularly for network infrastructure type solutions (that is, building to building or AP to AP) where compatibility is pretty much a non-issue. In that respect, perhaps 802.11a is better thought of as a replacement for many of the wacky proprietary wireless solutions more than as a replacement for 802.11b.