All three systems were tested with the Version 2 Wi-Fi Mesh System test process. The V2 mesh process uses many of the tests from the Version 11 Wi-Fi Router test process, so we're now testing both throughput and latency in the mesh. Benchmarks also include a simple "capacity" test.
Firmware used was 184.108.40.206.386_43084, 220.127.116.11.386_43181 and 18.104.22.168.386_43981 for the AX92U, XT8 and ET8, respectively. Results are found in the Wi-Fi Mesh System Charts, Finder and Ranker. You can get to them all using the Tools link in the site navigation bar.
Routing Benchmark Results
Today's routers are all capable of wire-speed gigabit throughput. Any differences you see in results are more due to test process variation than product performance. Keep in mind the best you'd ever see in this benchmark would be ~ 944 Mbps, not 1000 Mbps, due to TCP/IP overhead. So there's not much to say about the three ASUS' results, or any of the others shown in the charts below.
Routing throughput results
Routing latency scores (higher is better) are shown below. Converting back to latency (latency = 1/(score/1000) (lower is better) yields about 20 ms for both Zens and 33 ms for the AX92U. Keep in mind that the benchmark really loads up the router, using 100 download and 100 upload connections running TCP/IP traffic full-blast, while running uplink ping. So I'd consider these numbers to be pretty worst case.
Routing latency score - 90th percentile
Although the Charts include average and maximum throughput in the ranking algorithm, along with range, all are pulled from the throughput vs. attenuation (aka RvR) benchmarks. So that's our focus for comparing basic radio performance. Keep in mind channel bandwidths used for the following benchmarks are 20 MHz @ 2.4 GHz , 80 MHz @ 5 GHz and 160 MHz @ 6 GHz.
2.4 GHz downlink shows the advantage Wi-Fi 6 provides for 2.4 GHz due to its ~ 2X higher link rate. It's easy to forget that Wi-Fi 5 products support 802.11ac only for 5 GHz; 2.4 GHz uses Wi-Fi 4 (802.11n). That means the AX92U's 2.4 GHz radio supports a maximum link rate of 144 Mbps @ 20 MHz channel width, which it achieved in this test run, while the AX radios in both Zens hit the maximum link rate of 286 Mbps at the start of the test.
2.4 GHz throughput vs. attenuation - downlink
2.4 GHz uplink shows the ET8 with an advantage over the XT8, which is odd, given they have the same radios. The AX92U again lags far behind.
2.4 GHz throughput vs. attenuation - uplink
5 GHz downlink again shows both Zens' AX radios providing a significant throughput advantage over the AX92U's AC radio. The AX92U hit its maximum 867 Mbps link rate at the start of the test, while the Zens achieved their 1201 Mbps maximum.
5 GHz throughput vs. attenuation - downlink
5 GHz uplink still shows a throughput advantage for the Zens over the AX92U, although not as much. Unlike downlink, the advantage doesn't last as signal levels drop.
5 GHz throughput vs. attenuation - uplink
Since routers and mesh systems are tracked in different charts, I manually pulled 6 GHz comparison data together for plotting. I'm comparing the only two Wi-Fi Mesh systems with 6E radios, the ET8 and Linksys Hydra Pro 6E. Note this comparison pits Linksys' Qualcomm-based 6E radio against ASUS' Broadcom 6E radio. This test also shows the necessity of a > 1 GbE Ethernet connection on a 6E capable router or AP. Settings used were channel 37 @ 160 MHz channel bandwidth.
The Linksys' 6 GHz downlink throughput is somewhat higher than the ASUS with high signal levels (low attenuation). But as signal levels drop, the ET8's throughput remains high while the Linksys' drops off. This indicates a likely effective range advantage for the ET8.
6 GHz throughput vs. attenuation - downlink
6 GHz uplink shows the Linksys with a significant throughput advantage at higher signal levels, but the advantage disappears as signal levels drop.