Hi @DanMesh,
Welcome! Indeed I have many posts on this (most in 2020). So you're saying two total EAP245V3's? One ethernet connected and one MESH?
That is a little different from my MESH setup, I have two Ethernet-connected EAP245V's and two MESH-connected EAP225-Outdoors (one per Root Node).
MESH is implemented using the 5GHz Wifi radio only. What I have observed for my setup, as a very rough rule of thumb, is there is roughly a 50% reduction in real throughput for each wireless hop based on the air link rate of said hop.
So, for example:
1) 2x2 Wifi5 client (iPhone XSM) connected to EAP245V3 directly. Air link rate 866 MBit/sec. Actual throughput ~350-500 Mbit/sec
2) 2x2 Wifi5 client (EAP225-Outdoor) connected to EAP245V3 directly -- ethernet connected speed test PC connected to EAP225-OD for testing. Actual throughput ~350-500 Mbit/sec
3) 2x2 Wifi5 client (iPhone XSM) connected to EAP225-OD via Wifi, which is in turn, MESHED to EAP245V3. (aka 2 hops). Actual throughput ~110-170 Mbit/sec
Or in other words, if all air links were negotiated at 866 Mbit/sec, two hops would give you a maximum theoretical throughput of, at most, 1/4 of the link rate:
866 * 0.5 * 0.5 = 216 Mbit/sec (I've seen up to ~170 Mbit/sec max with mine).
Now your case is somewhat different from mine. Because EAP245V3 to EAP245V3 via MESH is a 4x4 scenario -- Aka AC1750. I'm not sure I've ever seen anyone bench test this. Theoretically, it should be better than what I am seeing.
Your MESH backhaul real throughput should be roughly 867 Mbit/sec (assuming a pristine airlink of ~1734). So devices connected via Ethernet to the second AP should see speeds up to this speed. Wireless clients connected to the second AP should see about half that -- or roughly 433 Mbit/sec.
Good luck! Keep us posted,
Jonathan