Sam thanks for replying. I've had a look this and other associated articles and I can't locate KB000001616 KB 000001477 which are referred to in the guide you provided. I was unable to open the link that was in the attached reply via the E-Mail notification but have been able to open and view it logged in via the Hive Community.
With regards the original question even after reading several articles it's still not clear whether there is an optimum for Mesh - 802.11ac, 802.11n or simply 802.a? In theory 802.11ac should be better than 802.11n which should then be better than 802.11a. However from what I can gather the benefits achieved in ac where I believe achieved by the use of MIMO (Multiple Antennae). Correct me if I'm wrong but for a 5Ghz Mesh link we have only a single directional Antennae at both ends of the link so MIMO is not really going to work? There also seems to be recommendations to not bond channels but why? This seems to be one way that 802.11n and 802.11ac can be improved - bonding channels. I understand why we wouldn't want to do this if we were using the radio to connect clients or were we using 2.4Ghz, or were there other devices in the area using 5Ghz that might interfere with our radio. Looking at the output of the show acsp neighbour I'm not seeing this as an issue. Also are we sure we want to go with fixed channels? We've selected and used auto and it seems to work. I would have thought selecting auto would mean that the best channel choices are chosen? I do however see why we might want to fix the channels so as to ensure both ends are communicating on the same channel.
At the moment I've managed to get a Mesh connection up and working using bonded bandwidth of 80Mbps (4 x 20Mbps) and using 802.11ac. Personally I think it would be no better or worse were we to use 802.11n because it seems the only advantage we have for this type of connection is bonding even though everything I read seems to advise against this?
