Forums | MacLife

Yeah, that would suck if so, and if so, why bother with upgrading?

linky?

SamGuy wrote:

linky?

http://www.apple.com/airportextreme/performance.html

http://www.apple.com/airportextreme/performance.html wrote:

Based on a comparison with Apple's 802.11g products. Comparison assumes AirPort Extreme network with 802.11n-enabled computer. Speed and range will be less if an 802.11a/b/g product joins the network. Accessing the wireless network requires an AirPort- or AirPort Extreme-enabled computer or other Wi-Fi Certified 802.11a/b/g-enabled computer. Actual performance will vary based on range, connection rate, site conditions, size of network, and other factors. Range will vary with site conditions.

You could just run two Airports, one for b/g and the other for n.  BTW, I agree with you on the stupidity of Apple's new Airport.

This is also the case for the older AirPorts (or any other wireless router, for that matter): have a mixed environment of 11b and 11g devices and the network will slow down to 11b speeds.

There is an option to restrict AirPort Expresses to accept only 11g devices (cannot say if the AP Extreme Base Station can do the same). I imagine that the new AirPort Extreme will have something similar; perhaps only 11n, only 11n and 11a/g, or any 11a/b/g/n devices.

And the truth is it does it across the board. An 802.11b device drops the speed of an 802.11g network as well as a Draft-N Network. An 802.11g device drops the speed of a Draft-N Network. A similar effect happens with 10BaseT, 100BaseT, and/or 1000BaseT on the same router, IIRC. That's why I think ethernet switches exist.

Avari wrote:

Shadowless wrote:

Mr. T wrote:

You could just run two Airports, one for b/g and the other for n.  BTW, I agree with you on the stupidity of Apple's new Airport.

It's not just Apple's though. Every singe wireless router out there gets slower if they're being used by an older wireless tech. I would've thought someone would fix the problem by now, out of all those companies, including Apple.

Apple doesn't develop the network chips and would have to design something only macs could use to change that.

Right, but I'd suspect that some company came up with this N soon-to-be-standard, and I would've liked it if Apple tried a new standard that could always run at full speed.

But I don't even know if that's technically possible, so I guess what we have now is fine.

As long as it's better than 56k, I'll take it. (shudders at the memory)

I'm pretty sure it's interference. As each one comes out on the same bandwidth, they interfere with one-another. The simplest is the one that will win. The problem isn't in the hardware, it's in the radio-waves.

A hub is dumb, it can only move packets around. If one device can only operate at 10, then it slows them all down. It can't make the 10 work at 100. A router isn't as dumb, it can take packets and transfer between the different speeds. This isn't really my area, but that's how I understand it.

Wireless is essentially transmitting across the same wire. The faster protocols are more susceptible to interference. This includes other networking protocols on the same bandwidth. The simpler one will generally win as it is the most robust.

An example. If you were to transmit Morse code and voice over the same frequency which one would come through? The Morse as it is far simpler than the voice and the data behind it isn't lost where as someone trying to listen to the voice data would only get half of the conversation. I'm under the impression that this is the same for wireless protocols.