I finally nailed down the Revision 11 test process, which was used to test the Hydra, using version 126.96.36.199016 firmware for the tests. Since it's no fun looking at a single router's test results, I retested the ASUS GT-AXE11000 and GT-AX11000 using the Revision 11 benchmarks for your comparison pleasure.
And yes, the Charts are back, baby! But because of the way they work, I had to create a different set of Charts to use the new set of benchmarks. So all Wi-Fi 6 and 6E routers will be found in the Wi-Fi Router Charts, Finder and Ranker. You can get to them all using the Tools link in the site navigation bar.
Routing Benchmark Results
Every router tested before I paused reviews two years ago was fully capable of wire-speed gigabit throughput. So until manufacturers include > 1 GbE ports on both WAN and LAN, wired throughput testing is going to be pretty boring. 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. The Hydra, and the other routers, are equally capable on these benchmarks.
Routing throughput results
Updated 8 June 2021: Latency measurements converted to scores.
The new benchmarks still include latency, which is generally pretty low. I'm showing the average and 90th percentile values that come right from the betterspeedtest.sh script used for wired router performance benchmarking. I've converted latencies to "scores" because the Ranker always ranks values from high to low and latency is the inverse (lower is better). To convert from score to latency use latency = 1/(score/1000).
The Hydra and ASUS GT-AXE11000 pretty much tie for lowest average latency, with scores of 51 and 49.5, respectively. The 90th percentile benchmark shows the Hydra lower than the AXE11000 with a score of 39.2 vs. 47.8. (This benchmark means 90% of the measured values are equal to or less than the value shown.) Remember, this is the round-trip latency (measured by ping) from LAN Ethernet port to WAN Ethernet port, i.e. through the router. For reference, the worst case 90th percentile score of 32.7 for the GT-AX11000 equates to around 31 ms.
Routing latency scores
Wi-Fi Throughput - Average
There are just two Wi-Fi benchmarks in the Revision 11 process, throughput vs. attenuation (aka "Profile") and multiband throughput. I'm no longer trying to assess the absolute maximum throughput that a single radio can produce, since that requires a four-stream device (aka another router in wireless bridge mode) and that's not the way most of use use our routers. That number is great for manufacturer bragging rights, but of little relevance to the average user. But for those of you concerned with the top throughput you can get from a two-stream Wi-Fi 6 (or 6E) device, that I can show you!
But let's start with average throughput. The values shown in the next set of charts represent the average of all the values measured by the throughput vs. attenuation benchmark. The 2.4 GHz results show the Linksys at the bottom of the pack, by a lot with 87 Mbps average downlink and 98 Mbps average uplink.
2.4 GHz throughput - average
After digging into the results, it appears the reason for the Hydra's lower throughput was that both transmit and receive link rates were not the same and they varied. A sample of the STA stats produced by the octoScope STApals is shown below for 2.4 GHz downlink; Linksys on the left, ASUS on the right. Since test conditions were the same, I suspect this could be the difference between the Linksys' two stream radios vs. the both the ASUS' four streams.
Sample STA stats - Linksys Hydra Pro 6E (left), ASUS GT-AXE11000 (right)
5 GHz average throughput tells a similar story, with the Linksys again at the bottom of the pack producing 290 Mbps average downlink and 276 average uplink throughput. Digging into the stats reveals that the Linksys reached a maximum receive rate of 720 Mbps (MCS 7) vs. 1200 Mbps (MCS 11) for the ASUS. This, again, could be due to two streams vs. four.
5 GHz throughput - average
Finally, for 6 GHz average throughput results, the GT-AX11000 drops out, leaving only the two 6E routers. This time, with both routers matched with four-stream radios, it's more of a fair fight. Although the Hydra lags slightly behind the ASUS for downlink with 1.241 Gbps, it handily beats it on uplink at 1.107 Gbps.
6 GHz throughput - average
Wi-Fi Throughput - Maximum (2 stream)
By changing the Charts selector to Maximum as shown below, we can now compare maximum throughput measured during the throughput vs. range tests. Remember, this is for a two-stream client, so curb your enthusiasm!
Switching from Average to Maximum benchmarks
The difference in 2.4 GHz maximum results between the Hydra and ASUS routers tracks what we saw in the average results, as would be expected. Keep in mind maximum link rate is 286 Mbps. The Hydra clocks in at 129 Mbps downlink and 171 Mbps uplink.
2.4 GHz throughput - maximum
I have to say the 5 GHz maximum results surprised me. I haven't thought that two-stream STAs could produce > 1 Gbps throughput. But with a maximum 80 MHz channel bandwidth link rate of 1200 Mbps and a > 1 Gbps Ethernet port for the traffic generator connection, apparently it's possible! And it's also interesting that the GT-AX11000's Broadcom Wi-Fi 6 radios do better than the GT-AXE 11000's Broadcom 6E radios in making the most of airtime. Unfortunately, the Linksys once again suffers due to its 2 stream radios, producing only 563 Mbps downlink and 785 Mbps uplink.
5 GHz throughput - maximum
Finally, 6 GHz maximum results show more evenly-matched performance between the ASUS and Linksys on downlink and the Hydra racing way ahead of the GT-AXE11000 on uplink. Since these tests are run with 160 MHz channel bandwidth, maximum link rate is 2402 Mbps. It's a shame you'll only be able to achieve this throughput in actual use if you have a 2 Gbps or higher internet connection since the 5 GbE port is locked to the WAN side. The Linksys beat the ASUS with 1.936 Gbps downlink and 2.049 Gbps uplink. Remember this is with 160 MHz channel bandwidth, which is how most consumers will configure their 6E routers.