I love my network.
I did a quick series of Ookla Speedtests with my iPhone 6s and iPad Air 2.
I started in my living room, which is one room above and a couple interior walls away from my basement access point and the opposite end of the house from my upstairs access point. This is probably my worst coverage room in the house. Both devices were on the basement AP at 2.4GHz. They got about 50 Mbs download and 12 Mbs upload (all the tests got around 12 Mbs upload).
I walked upstairs near the other, and both devices flipped to the other AP at 5GHz. They got 90 Mbs download.
I walked to the other side of the upstairs, and they stayed on 5GHz and got about 80 Mbs.
I walked down to the basement by the main router (the original AP hosting the first test) and both devices flipped to the local AP at 5Ghz. They scored 85 Mbs on that one.
Then, I went back to the living room and they both stayed on the basement AP but flipped to 2.4Ghz like originally. I disabled 5GHz on the basement router, and re-ran a speed test there. I got about 38 Mbs.
So, not only did my devices dynamically flip between access points and bands, but they intelligently choose the best AP and band that gave the best throughput. I love science!
Does the fact that responses from lap-people that contradict an alleged (former) Apple WiFi Engineer make it dubious? The article is describing a standard WiFi network configuration.
Not only in those comments, but in lots of other Mac help forums and blogs. There is a lot of debate as to whether separating 2.4 and 5 with the Airport is the right way to go. I'm not saying he's wrong, but I'm not sure it works as well as he thinks in all situations.