Dead Spot Terminator Performance
Testing and analysis by Tim Higgins
The DST is a powerline-connected access point, so end-to-end performance can be broken into two parts: the powerline "backhaul" and AP-to-STA (client) connection. I started by measuring powerling throughput with the DST in the same Locations A, C and E used for powerline adapter testing. You can learn all about the locations in How We Test Powerline Products.
An Acer Aspire S7 Running Windows 8 was plugged into an R7300DST LAN port to form one end of the IxChariot pair. My trusty Lenovo x220i running Windows 7 traveled among test locations as the other IxChariot endpoint and ran the console.
The DST adapter's WiFi settings were Channel 6 for 2.4 GHz and Up to 65 Mbps Wireless mode (20 MHz bandwidth). Channel 153 and Up to 433 Mbps mode (80 MHz bandwidth) was set for 5 GHz. The R7300DST router's WiFi settings were Channel 6 for 2.4 GHz and Up to 289 Mbps Wireless mode (20 MHz bandwidth). Channel 153 and Up to 1300 Mbps mode (80 MHz bandwidth) was set for 5 GHz.
The IxChariot plot below summarizes powerline downlink (router to DST) throughput for the three locations. As we typically see, throughput falls off quickly as distance increases between adapters.
Powerline throughput - downlink
This next plot shows powerline uplink, with a sharp decrease in Location A throughput. Thinking the problem might be signal overload from the two adapters plugged into the same outlet, I tried moving one adapter to an outlet on the other side of the wall dividing two rooms with no significant change in results. So there is something wonky about powerline uplink, at least by my measurements. Note these results represent maximum wireless throughput available from the access point at each of the three test locations.
Powerline throughput - uplink
Location E also represents a wireless dead zone in my home. I can get a barely usable 2.4 GHz there and no 5 GHz signal at all when a router under test is sitting on my test bench. However for these tests, the R7300DST had to be moved to a spot closer to my office door, so its AC line cord could be plugged into the powerline test Location A hallway outlet outside my office. This outlet avoids the powerline throughput-killing AFCI breaker that protects my office. I was suprised to find I could get usable 2.4 and 5 GHz throughput in Location E when connected directly to the R7300DST with this new router location. Just proves once again the value of good router location.
I then ran two sets of tests for each band, with the R7300DST directly connected to the test laptop (the Lenovo x220i with NETGEAR A6200 AC1200 class USB adapter) and with the test laptop connected to the DST adapter.
The chart below shows results with the DST in Location E. Location E direct is throughput with the test laptop associated directly with the R7300DST. Location E powerline is throughput via powerline connection between the R7300DST and DST adapter, with the test laptop connected via Gigabit Ethernet (and wireless shut off). Location E DST is throughput with the test laptop associated with the DST adapter. When the test laptop was connected to the DST, it was located within 6 ft.
The DST connection provides higher throughput than being connected directly back to the R7300DST in all four benchmarks. At 40 Mbps, 2.4 GHz downlink isn't being limited by the powerline connection, but it appears that uplink might be. For both 5 GHz up and downlink, the powerline and DST results are so close, it's a good bet the powerline connection is the limiting factor.
Wireless throughput - DST Location E
This is probably a good time to talk about what class the DST radios really are. NETGEAR specs them as AC650 (2.4 GHz: 300 Mbps + 5 GHz: 350 Mbps). The 2.4 GHz part is easy; it looks like a 2x2 configuration. To suss the 5 GHz radio, I had to look at the MCS table. The closest rate I found for 80 MHz bandwidth was 351 Mbps for a 1x1 configuration.
But the DST's Wireless Mode selections and actual link rates experienced during testing tell the real story. When parking the test laptop within 6 ft. of the DST, the best link rate I got for 2.4 GHz was 72 Mbps. That's the maximum link rate for a 1x1 11n device using Short Guard Interval with 20 MHz bandwidth. Best link for 5 GHz was 433 Mbps, which we all know as best case for a 1x1 11ac adapter, again using Short Guard Interval with 80 MHz bandwidth. My guess is the radios are 2x2 for receive only.
Since Location C powerline throughput was much better than Location E, I ran a second set of tests. The DST adapter was plugged in at Location C, but the test laptop was in Location E, since that's the dead spot I was trying to light up. The plot below more clearly shows the 2.4 GHz radio as the limiting factor in what you'll get from the DST. The plot also shows what the 5 GHz radio can do when the powerline connection isn't capping throughput.
Wireless throughput - DST Location C
By Tim Higgins
Despite their focus on selling us expensive big honkin' routers, consumer wireless makers know the days are numbered for this approach. Beamforming, active antennas and other Wi-Fi wizardry are no match for the physics of 5 GHz signal propagation. In the end, the only way to cover large areas with decent throughput in 5 GHz is to add more access points and make multi-AP networks easy to set up and manage.
There are actually already multi-AP networks in many homes today, built with wireless extenders. Today's dual-band 11ac extenders are a far cry from the bandwidth-starved single-band N150 extenders we had to settle for only a few years back. With 802.11ac radios, simultaneous extension of both bands, smart backhaul management and much easier setup, it's no wonder router makers are selling a pantload of them.
But using powerline to connect a wall-plugged AP certainly has its advantages, especially when airwaves are crowded and both bands are heavily used. Moving backhaul traffic off the air can lower the load and make Wi-Fi users happier, if powerline doesn't cause more problems than it solves.
And there's the rub. Lke all Ethernet alternatives, powerline networking has its own foibles, something our testing showed. It's sensitive to noise, especially from ubiquitous USB and other wall-wart power supplies. Some brands of AFCI breakers that have been required in U.S. homes for years now can seriously knock down throughput, too. There's also the matter of just finding a place to plug them directly into the wall, especially in areas they're most likely to be used, such as the family media room.
In the end, you'll need to experiment with DST placement, just as you do with wireless extenders. Unfortunately, the tri-color LED signal quality method HomePlug AV has adopted isn't much help. I've yet to see that light any color than green even though measured throughput changes more than 10X! But at least you don't need to mess with getting the DST and router to talk to each other; that was very reliable and quick in my testing.
All said, it's not likely the R7300DST and DST adapter bundle will make a huge dent in wireless extender sales or set the Wi-Fi world on fire. After all, it's available only from Best Buy (currently for $300) and it hasn't been widely promoted or even reviewed. But if you're not happy with the results you've gotten from trying wireless extenders, and you're looking to trade up to an AC router, you should give the NETGEAR R7300DST a shot.