Wireless Performance - Mixed Mode
Since Linksys has the 110 in a class by itself, i.e. not "Wireless-G" and not "Wireless-N", I thought I'd see how it handled a mix of "Range Plus" and straight 802.11g clients. Figure 9 shows the 110 set to its default Auto 20/40 mode and running downlink, with the 11g client starting first.
Figure 9: Mixed clients
Resulting performance is very similar to a mix of draft 11n and 11g clients on a draft 11n router. The 11g client's throughput gets cut in half and the WPC110's speed is reduced by about 33%. I saw basically the same results running uplink or when starting the WPC110 first.
I also set up my neighboring WLAN test, on the off chance that the combination of Ralink 1T2R radio in the WRT110 might be friendlier to nearby "legacy" 802.11b/g wireless LANs than draft 11n routers are (actually, aren't), even though the 110 defaults to Auto 20/40 MHz mode.
I made sure that the 110 was using Channels 1 and 5 (via spectrum analyzer) and set up the 11g "neighbor" WLAN, consisting of a Linksys WRT54G and notebook with an Atheros Super-G client on the extension Channel 5.
Figure 10: Neighboring WLAN test
Figure 10 shows the results running uplink with the 11g neighbor started first. Throughput is knocked down to around 5 Mbps when the WRT110 / WPC100 pair start running, but recovers back to 20 Mbps once they stop. I monitored the spectrum during the test and saw that the WRT110 / WPC100 pair used a 40 MHz channel while they were on the air.
I suppose Linksys' marketing folks have to earn their salaries. But I wish that they—and their colleagues at every other consumer networking company—would stop relabeling technology purely for product positioning and competitive differentiation.
Although it's questionable whether technology using a single transmit stream can qualify as "MIMO", or draft 802.11n for that matter, the WRT110 is really another cut at making an inexpensive draft 802.11n router. And since the "RangePlus" WPC100 card uses the same 3T3R draft 802.11n chipset that I've seen in many of the draft 11n routers and cards that have hit the SmallNetBuilder test bench, calling it something other than draft 11n is really getting too creative.
But let's say that despite its problems with WPA/TKIP and inability to use both 40 MHz bandwidth and a channel of your choice, you are trying to decide between the 110 and Linksys' other routers. How to choose? I went back through my reviews and pulled together Table 1 of Linksys' various flavors of single-band wireless routers.
|20 MHz B/W||40 MHz B/W|
|Range Plus||~ 40||~ 60||WRT110||$50-$94|
|Draft 11n - Two Antenna||~ 60||~ 77||WRT160N||$68-$102|
|Draft 11n - Three Antenna||~ 70||~ 94||WRT350N||$129-$180|
Table 1: Linksys Product Line Comparison
Keep in mind that price is constantly changing and that the WRT350N is being replaced by the WRT310N, which is closer to $100. The table shows that from a price and performance viewpoint, the WRT110 does fall between the WRT54G2 (plain 11g) and its entry-level two-antenna draft 11n WRT160N. (Ah, the magic of marketing!)
The chart also shows that for not a lot more money ($5 - $10 if you shop carefully), the RangePlus can double your wireless throughput, if you use what is really a 3T3R draft 11n wireless client. But if you're going to do that, why not spend $10 more and get a real two-antenna draft 11n router?
It seems to me that the "RangePlus" line—at least the router side—is a false economy. Saving $10 or so compared to a low-end draft 11n router is peanuts compared to what you need to spend to upgrade to the draft 11n clients that you'll need to get any benefit from the WRT110.
At least the good news is that the WPC100N is a real 3T3R Draft 11n notebook card (essentially, a clone of the D-Link DWA-652, which sells for about $10 more). So if you choose to buy the WPC100, you'll be buying something that will play nicely with draft 11n gear, and vice versa for draft 11n cards and the WRT110.
In the end, the WRT110 is an odd duck that sure quacks and walks like a low-end draft 11n router.