We developed a new test methodology for our initial wireless extender roundups that positioned the test client in a true "dead zone". As noted earlier, "lighting up" dead zones are what wireless extenders are intended to solve, so that's the way we test them. Since this is an AC class extender, we used a revised test procedure, which uses a different client. We also changed locations back to the SNB labs, since we have a much quieter wireless environment here.
The wireless results in the Benchmark Summary represent "extended" wireless throughput, i.e. from base router, through the EX6200 and to the test client laptop. The wired results are what we measured with the test laptop's wireless disabled and connected to the EX6200 Ethernet port.
NETGEAR EX6200 Benchmark Summary
Following the 50% throughput reduction rule of thumb, we would expect the wireless results to be around half the wired throughput. This appears to be the case for 2.4 GHz, but not for 5 GHz. 5 GHz wireless downlink results are only 22% lower and uplink only 31% lower than the wired throughput available at the EX6200 Ethernet ports. Keep in mind that all these results were obtained with only "two bar" signal qualities and about half the maximum available link rates, as shown in the EX6200's status below.
NETGEAR EX6200 Connection Status for Tests
For a look at how steady throughput is, we'll turn to the IxChariot plots. Each is a composite of the test plots for the wired test (extender_eth) and wireless (extended).
2.4 GHz downlink shows a good amount of throughput variation, but the expected 50% reduction in average throughput between wired and (extended) wireless results. The extended signal's link rate at the test laptop was a steady 144 Mbps, reflecting the fact we test 2.4 GHz in 20 MHz bandwidth mode.
NETGEAR EX6200 throughput - 2.4 GHz down
2.4 GHz uplink throughput is a bit more steady, but also shows a 50% reduction in average throughput between wired and (extended) wireless results.
NETGEAR EX6200 throughput - 2.4 GHz up
5 GHz downlink shows quite a high amount of variation in the extended plot, which could indicate a lot of rate shifting taking place. I didn't see this reflected in the link rate shown by the Windows laptop, which remained quite steady at 585 Mbps.
NETGEAR EX6200 throughput - 5 GHz down
5 GHz uplink show is also interesting because it shows high variation in the wired extender throughput and much steadier wireless extended throughput. If the extender throughput were steadier, extended throughput reduction probably would have been closer to the expected 50%.