Like all other AC3200 routers except ASUS' RT-AC3200, the DIR-890L/R treats Smart Connect like a black box. The only thing you can do is disable and enable it. D-Link but identifies only wired and wireless connections.
For Smart Connect testing, the DIR-890L/R was located in the wireless testbed upper test chamber with the door open. The bridge mode R7000 was in the lower chamber with the door closed. This allowed using the testbed programmable attenuators to control the signal and therefore the link rate of the R7000. I set 20 dB of attenuation so that the R7000 throughput wouldn't dominate the other AC devices. All other devices were located within 6 feet of the DIR-890L/R, outside the test chamber and all received a nice, strong signal.
Since I couldn't see how wireless clients were distributed among the 890L/R's three radios, the only thing I can report is observed throughput. I used the same five devices used in previous AC3200 router tests. Table 3 shows the devices and the device descriptions shown in the Legend panel in the following IxChariot plots.
|Moto X smartphone||1x1 AC||MotoX|
|NETGEAR R7000 in client bridge mode||3x3 AC||R7000 bridge|
|Laptop with NETGEAR A6200 USB adapter||2x2 AC||NETGEAR A6200|
|iPad 2nd gen||1x1 N||iPad2|
|iPod Touch 5th gen||1x1 N||iPod 5th Gen|
Table 3: Smart Connect Test devices
I first disabled Smart Connect and connected all clients to the Secondary 5GHz radio (D-Link's terminology) set to Channel 153. The IxChariot plot below shows this test yielded 221 Mbps of total throughput from the five clients. Total uplink throughput on this run was 214 Mbps.
Total downlink throughput - All clients on 5GHz-2 radio
I then enabled Smart Connect with its default settings, changed the SSID (because Smart Connect re-used one of the separate SSIDs) then reassociated all clients. The R7000 bridge used as a 3x3 AC client has to be assigned to a band, so I chose 5 GHz. This test yielded a lower 171 Mbps total throughput running downlink, but higher 287 Mbps total throughput running uplink.
This higher result was not a good thing, however. The IxChariot plot shows throughput was high mainly because the 3x3 R7000 bridge hogged throughput. All the other clients except for the MotoX were choked down to very little bandwidth. I suspect the MotoX managed to find its way to its own radio.
Total uplink throughput - First Smart Connect run
I then power-cycled the router, let the clients reconnect and ran another test. This resulted in 322 Mbps total downlink throughput and 277 Mbps total uplink.
To see if this result was repeatable, I power cycled the router again and let the clients reconnect. This final run yielded total throughput of 146 Mbps downlink, 268 Mbps uplink. Checking the uplink plot showed the same result as in the first Smart Connect run shown in the plot above, i.e. the R7000 bridge with high throughput and other devices starved for bandwidth.
I pulled all the results into Table 4 along with the % change in throughput in each direction compared to the initial run with all clients connected to the Secondary 5GHz radio. In only one out of three test runs was there a gain in total throughput in both directions. And where total throughput was worse, it was in the more desirable downlink direction!
|Total Downlink (Mbps)||Total Uplink (Mbps)||% change downlink||% change uplink|
|Smart Connect Powercycle 1||322||277||46||29|
|Smart Connect Powercycle 2||146||268||-34||25|
Table 4: Smart Connect Test summary
All the IxChariot plots are in the gallery below, so that you can see how individual devices behaved. Remember, we have no way to know where devices connected in each test.
So how does the DIR-890L/R's Smart Connect compare to other AC3200 routers? Table 5 compares the best results from all four products. The good news is the DIR-890L/R produced the highest total throughput with Smart Connect enabled. The bad is that it could not produce these results consistently.
|Product||Without (Mbps)||With (Mbps)||% gain|
Table 5: Smart Connect Total Throughput gain comparison - downlink
D-Link takes the prize for most eye-catching router we've seen in awhile in any Wi-Fi class. But, like all other AC3200 routers to date, it will consistenly produce higher total wireless throughput only if you have enough dual-band devices and only if you manually distribute them among radios. Whether it's D-Link's implementation or Broadcom's XStream core technology, reliable and really intelligent client distribution among tri-band router radios remains an elusive goal.
Broadcom and its partners still have plenty of time to work out the kinks, however. 4x4 MU-MIMO enabled routers based on QCA's chipset should start appearing shortly. But it will be many months before we have enough MU-MIMO enabled devices in the wild for buyers of those science experiments to see if they get a return on their investment. Buyers of first-generation Quantenna based MU-MIMO "ready" router could be in even worse shape, since no manufacturer has yet announced plans to remove the "ready" disclaimer from those products.
My advice remains the same. If you want a reliable AC router for a reasonable price and have a lot of wireless devices, get an AC1900 class router. If you have only a few wireless devices and/or a smaller space to cover, an AC1200 class should do you just fine. AC3200 class routers in general are just not way to spend your Wi-Fi equipment budget.