I understand your point, and I agree that regulatory compliance is important. However, I don’t think the discussion is only about whether v4.8.3 exceeded the limits.
From my own testing on an EU Flint 2, v4.8.3 consistently provides better real-world coverage than later firmware. I also tested v4.9.0, and while it includes many excellent new features, I still observed reduced Wi-Fi coverage compared to v4.8.3.
I’m not asking GL.iNet to ignore regulatory requirements. My suggestion is that they review whether the current implementation is unnecessarily conservative for some regulatory domains. If the selected country legally allows higher EIRP on certain channels (for example DFS channels where permitted), the firmware should make full use of those limits instead of applying a lower value across the board.
I appreciate the new features in v4.9, but I also hope GL.iNet can further optimize Wi-Fi performance while remaining fully compliant with local regulations.