GL-MT6000 OpenWrt 25 Beta – Real-World Wi-Fi Performance Tests (USA vs Panama, TX Power, RSSI & Coverage)

Router
1

Hello everyone,

I would like to share some real-world testing results from the latest GL-MT6000 OpenWrt 25 Beta (mt6000-op-4.9.1-op25_beta1-1032-0707-1783421695).

The purpose of this topic is to provide practical measurements that may help both the community and the GL.iNet development team.

Test Environment

  • Router: GL-MT6000 (Flint 2)

  • Firmware: OpenWrt 25 Beta (Kernel 6.12.94)

  • Test tools:

    • Apple AirPort Utility (RSSI)

    • TP-Link RE715X (Backhaul quality)

    • LuCI

    • GL.iNet Admin Panel

    • Speedtest

Test 1 – Panama vs USA

I compared exactly the same configuration using only different country codes.

USA

LuCI reports:

  • 2.4 GHz = 30 dBm

  • 5 GHz = 30 dBm

GL.iNet Admin Panel:

  • Wi-Fi Coverage = Max

Panama

When the GL.iNet Admin Panel shows:

  • Wi-Fi Coverage = High

LuCI reports:

  • 2.4 GHz = 33 dBm

  • 5 GHz = 32 dBm

However, if I manually change Wi-Fi Coverage from High to Max, LuCI immediately changes both radios to:

  • 2.4 GHz = 30 dBm

  • 5 GHz = 30 dBm

This seems unexpected because selecting “Max” actually lowers the reported transmit power.

Test 2 – AirPort Utility RSSI

Measurements were taken at exactly the same location.

Panama (High):

2.4 GHz:

  • approximately -53 to -56 dBm

5 GHz:

  • approximately -65 to -66 dBm

USA:

2.4 GHz:

  • approximately -56 to -60 dBm

5 GHz:

  • approximately -66 to -68 dBm

Result:

Panama + High consistently produced approximately 2–3 dB stronger RSSI than USA.

Test 3 – Client RSSI reported by the router

I also noticed that the router reports client RSSI values that differ significantly from external measurements.

Example:

iPhone located approximately 3 meters from the router.

Router reports:

  • approximately -67 dBm

Apple AirPort Utility reports:

  • 2.4 GHz = approximately -40 dBm

  • 5 GHz = approximately -46 dBm

Despite this difference, Wi-Fi performance is excellent.

Test 4 – Performance

Speedtest results remained almost identical:

  • Download: approximately 250–300 Mbps

No noticeable throughput difference was observed between USA and Panama.

Current observations

  • Panama + High appears to provide the strongest real-world signal.

  • Changing Wi-Fi Coverage from High to Max reduces the reported TX power in LuCI.

  • Client RSSI reported by the router appears significantly different from measurements obtained with Apple AirPort Utility.

  • Wi-Fi performance remains stable and excellent.

Questions for GL.iNet

  1. Is the Wi-Fi Coverage (High/Max) behaviour intentional?

  2. Is the TX power synchronization between the GL.iNet Admin Panel and LuCI working as expected?

  3. Are the client RSSI values displayed by the router known to be inaccurate in the current OpenWrt 25 Beta?

  4. Is this behaviour related to the MediaTek SDK/OpenWrt 25 driver?

I hope these measurements are useful for both the community and the GL.iNet developers.

Thank you.

5

Hi,

Thank you for sharing these detailed test results.

We performed a local test on a GL-MT6000 running firmware v4.9.1-op25 with kernel 6.12.94, using both the US and Panama country codes.
2.4 GHz channel: 6 bandwidth: 40Mhz.
5GHz channel: 36 bandwidth: 80Mhz.
Our local results were as follows:
US + High: 21 dBm on 2.4 GHz and 21 dBm on 5 GHz.

image
image1438×1032 115 KB

image
image1428×914 101 KB

US + Max: 30 dBm on 2.4 GHz and 30 dBm on 5 GHz.
image
image719×511 26.1 KB

image
image722×467 24.1 KB

Panama + High: 21 dBm on 2.4 GHz and 21 dBm on 5 GHz.
image
image1439×917 105 KB

image
image1440×1048 119 KB

Panama + Max: 30 dBm on 2.4 GHz and 30 dBm on 5 GHz.
image
image1440×1071 124 KB

image
image1441×846 92.4 KB

For your quesssion:

  1. Based on our local results, High reporting a higher transmit power than Max is not the expected behavior. On our test unit, Max produced the same or a higher reported transmit power than High.
  2. In our test, changing the TX Power setting in the GL.iNet Admin Panel was reflected in LuCI after the wireless radios restarted.
    In your test, LuCI also changed immediately after switching from High to Max, which suggests that the setting itself was applied, but the resulting power value or power-level mapping differed from our local result.
  3. AirPort Utility records the network from the iPhone’s point of view and displays the RSSI received by the iPhone from the access point. The router client page reports the signal received by the router from the wireless client. These are measurements in opposite transmission directions, taken by different wireless radios and antennas, so the values are not expected to be identical.
  4. Based on the current information, we do not yet have enough evidence to attribute this behavior to the MediaTek SDK or the OpenWrt 25 wireless driver.

Could you please provide LuCI screenshots for both High and Max, showing the country code, actual operating channel, channel width, Maximum transmit power and Current power for both radios?

Thank you again for your detailed testing and support.