ASUS Lyra AC2200 Tri-Band Wireless Performance & AiMesh Tests

(Refer to initial post HERE )

Wireless performance and AiMesh tests (2.4GHz & 5GHz) based on my home network setup as follows :

Testing scenario and setup as follows :

  • ASUS Lyra MAP-AC2200 HW ver: A3
  • All 3 units upgraded and performed Factory Restore with the current latest official stock firmware version 3.0.0.4.384.46249 (Released on 4Dec2019)
  • Lyra setup as AP only, with my custom OpenWRT device as the main router gateway
  • TM UniFi Turbo-ed home package with 800/200Mbps download/upload speed
  • LAN iPerf3 test, WAN Ookla SpeedTest (Windows desktop version)
  • All laptop/PC with current latest Win10-64bit (Pro/Enterprice) update, all with SSD

Please refer to initial post HERE on how to setup the AiMesh for the 3 IPQ4019 nodes. 

I'm only interested in mesh device with proper dedicated backhaul, in this case the Lyra's 5G-2 is solely used for backhaul traffic only not to mix with any other serving radio.

Device wireless setup as follows :

2020-07-14 17.34.05-Lyra.png2020-07-14 17.34.34.png2020-07-14 17.35.12.png2020-07-14 17.35.31.png2020-07-14 17.35.45.png2020-07-14 17.37.16-5G2-Backhaul.png2020-07-14 17.39.33-1node.png2020-07-14 17.58.04-2nodes.png

Note : The above testing is solely based on my current environment/setup/configuration, it's for my own reference only and should not be used for any other comparison or benchmark purposes.


Lyra MAP-AC2200 Wireless Performance - Single device as AP only

Before I setup the 3 IPQ4019 radios as AiMesh, I test the single radio performance first

2.4GHz Wireless Performance :

2020-07-14 17.16.18-PCEAC88-2G.png2020-07-14 17.16.39.png2020-07-14 17.16.59.png2020-07-14 17.17.18.png2020-07-14 17.17.35.png2020-07-14 17.17.47.png2020-07-14 17.18.50.png2020-07-14 17.19.01.png2020-07-15 01.35.32-AX200-2G.png2020-07-15 01.35.47.png2020-07-15 01.36.13.png2020-07-15 01.36.37.png2020-07-15 01.36.55.png2020-07-15 01.37.24.png2020-07-15 01.38.19.png2020-07-15 01.38.38.png2020-07-15 01.38.59.png2020-07-15 01.39.21.png

Normal usual 2.4G performance, nothing to shout about, stable with no sudden up/down spike/surge

5G-1GHz Wireless Performance :

2020-07-14 17.21.06-PCEAC88-5G1.png2020-07-14 17.21.18.png2020-07-14 17.21.30.png2020-07-14 17.21.45.png2020-07-14 17.21.54.png2020-07-14 17.22.05.png2020-07-15 01.40.32-AX200-5G1.png2020-07-15 01.40.43.png2020-07-15 01.40.53.png2020-07-15 01.43.11.png2020-07-15 01.43.26.png2020-07-15 01.43.34.png2020-07-15 01.45.44.png2020-07-15 01.46.04.png2020-07-15 01.46.12.png

Quite good and stable 5G performance, especially with the legendary PCE-AC88 can hit around 5xx/6xxMbps download/upload speed.  Average performance with WiFi6 AX200 client radio, this is expected as the IPQ40xx is considered old and may not compatible with WiFi6 client radio.


Lyra MAP-AC2200 Wireless Performance - AiMesh

Basically I'm only interested to test the AiMesh on 5G-1 as my mesh intention is to increase/improve throughput, no point to test any mesh on 2.4G radio.  I turn on first node and test at ground floor kitchen back area first, after turn on the second node and test at upstair room.

AiMesh 5G-1GHz Wireless Performance - first node :

2020-07-15 02.08.24-5G1.png2020-07-15 02.08.43.png2020-07-15 02.09.01.png2020-07-15 02.09.20.png2020-07-15 02.09.38.png2020-07-15 02.10.14.png2020-07-15 02.11.17.png2020-07-15 02.11.28.png2020-07-15 02.12.26.png2020-07-15 02.12.38.png
 

AiMesh 5G-1GHz Wireless Performance - second node :

2020-07-15 02.23.22-5G1.png2020-07-15 02.23.39.png2020-07-15 02.24.18.png2020-07-15 02.24.31.png2020-07-15 02.24.50.png2020-07-15 02.26.23.png2020-07-15 02.26.33.png2020-07-15 02.27.21.png2020-07-15 02.27.35.png
 

With the same AX200 client radio on 5G1, compare to the previous single AP close range testing, now with the AiMesh performance throughput dropped by at least half at first area (wet kitchen back area), and dropped by around two-third at second area (upstair room's toilet).  This is what I meant earlier, even with dedicated backhaul traffic, more devices/radios may not neccessary means can gain lots of throughput, the burden/penalty is still affected by what/how the backhaul is running on.  Well, it's definitely better than before without any AiMesh for my 5G connection at this 2 weak/blind spot area smile, I guess I'll need to place/test the nodes on proper location hopefully can improve the throughput ... so I'll still keep this toy swap to use it whenever I want.


Next, OpenWRT-ed the Lyra --> the main reason I play/test this old toy laughing


The Lyra is known to operate very hot on stock firmware, but somehow the temperature is not that hot after flashing to OpenWRT, it's like normal usual temperature, strange though ...