Hello @lobsterboy
I'll try to answer as best I can.
1. When a wired point shows as "good" instead of "great" on a mesh test, it's often due to the Ethernet cable connected to that point only managing to negotiate and hold 100Mbps (using one pair of wires) instead of the full 1Gbps (all four pairs of wires). Occasionally, it can "flap" between 100Mbps and 1Gbps, leading to an effective throughput higher than 100Mbps, but less than 1Gbps. This is rare, but not impossible. I would check on the details of that point to ensure it shows as being connected "wired" rather than mesh, and then ensure the cable connecting it is working properly. One thing that can happen is using a Cat6 cable (which is stiffer than Cat5e) ends up bending enough to get to the jack on the Google WiFi unit that it results in poor connection of some pins. When this happens, the cable is fine – and testing it will show it's fine, including by connecting something else to it – but, it won't work reliably in the Google WiFi. If this is happening, and replacing the cable is not an option, you can pick up an inexpensive 5-port Ethernet switch, and connect it to the long cable run, then use a more flexible Cat5e cable from that switch to the Google WiFi unit.
2. The limit of 5 total units applies to mesh-connected systems, not wired. However, I would caution against installing a large number of wired points. It can work, but they still use a common 5GHz WiFi channel, so performance improvements when there are a lot of access points isn't necessarily better. If they are very close to each other, client devices may be less likely to hand off between them.
