Forum

Scenario #1 - this shows me that you have a .11b client associated to a .11g AP where the client is configured to use long preambles.

Scenario #2 - this shows me that (provided both APs were on the same channel) the linksys is not acting according to the .11g standard because it's supposed to enable protection when it sees beacons from the cisco AP that only has .11b rates supported. Also, it shows me that the .11b client using long preambles is still associated.

Scenario #3 - the cisco's beacon shows me that it's acting according to the .11g amendment EVEN THOUGH you have the .11g rates disabled...which is wrong. What it should do is to remove these fields from the beacon altogether.

The linksys beacon shows me that it still thinks the .11b client is associated to it AND it isn't responding to the beacons of the cisco AP that should make it enable use of protection.

I completely agree with your assessment on throughput consequences of protection.

I disagree with the enterprise setting suggestion because we both know it's completely impossible to guage how many .11b clients will be where and how much traffic they (vs. .11g clients) will send at any given time and place. For that reason, the main focus should be interoperability and not throughput (which seems to be how the .11g TG designed the amendment). If you want throughput, go with .11a. :-)

Devinator