Haha. Well done @Mloraditch . I run into MTU-related performance issues at least every 90 days helping customers. If you can - always buy Internet connections with a clean 1500 byte MTU to avoid this (such as a connection that uses DHCP or a static IP).
@JThomas9999 there is a pretty good chance there is something in the old 4000 config changing the MTU or MSS (if this is the issue).
You can test this yourself initially by lowering the MTU on both servers and re-testing. On each server do something like:
netsh interface ipv4 show subinterface
This will identify all the interfaces. Choose the one being used. It is often called "Ethernet".
Then, try lowering the MTU a lot on the interface on each server:
netsh interface ipv4 set subinterface “Ehernet” mtu=1300
If that fixes, try again with maybe 1400. If you find a size that fixes it, make the change permanent on the server with:
netsh interface ipv4 set subinterface “Ethernet” mtu=1300 store=persistent
If it doesn't fix it then rerun the command with an MTU of 1500 to roll it back.
I have also run into issues with ISPs where the upstream and downstream traffic have different travel times. You can mitigate this problem in Windows by enabling TCP timestamps.
netsh int tcp set global timestamps=enable
Sometimes in Windows you need to disable bandwidth throttling (yes, Windows has this enabled by default!!!). Do this in Powershell.
Set-SmbClientConfiguration -EnableBandwidthThrottling 0 -EnableLargeMtu 1
And if all of that fails, try the SMB tuning guide:
https://learn.microsoft.com/en-us/windows-server/administration/performance-tuning/role/file-server/...
You could also consider doing a packet capture in case something else is happening, like a lot of packet loss, out of order packets, etc.
