How simple is it to network your home via existing power lines?
I connected my PC's Ethernet port to a powerline adapter and plugged that adapter into a power strip that already fed several other devices. Across the room, where my home office network's 100Mbit/sec. router is located, I connected a port on that router to a second powerline adapter and plugged it into the nearest power outlet. I sat down in front of the PC, and I was online.
It's that simple.
Also, the connection felt as fast as the Ethernet cable I had been using a few moments earlier.
Subsequently, I found that I could move to any part of the house, plug in my laptop through the nearest power outlet using another adapter and be online. Unsightly Ethernet cables that had been strung through the halls could be removed.
I was testing examples of the latest generation of 200Mbit/sec. powerline adapters, using the HomePlug AV and Universal Powerline Association (UPA) specifications. (I didn't have access to any Panasonic High-Definition Powerline Communications units.)
HomePlug was represented by units from ActionTec Electronics, Cisco-Linksys and ZyXEL Communications, while UPA was represented by a pair of adapters from Netgear. Older, slower HomePlug 1.0 and HomePlug Turbo adapters (running at 14Mbit/sec. and 85Mbit/sec., respectively) can also be found on the market but weren't tested.
I found that the tested units were true plug-and-play devices. The only thing the average consumer would have to be mindful of is the need for at least two adapters: typically, one for each PC and one for a router. The only maintenance problem I encountered was that, after a thunderstorm, two HomePlug adapters had to be reset. (Three others weren't affected.)
An ActionTec powerline adapter, on the right, shares a power strip with several other devices amid the rat's nest of wires aside the author's desk. Using it eliminates an additional cable.
An ActionTec powerline adapter, on the right, shares a power strip with several other devices amid the rat's nest of wires aside the author's desk. Using such adapters typically lets you replace a longer, room-to-room cable with a shorter one from a device to an outlet.
There were no obvious differences between the performance of the HomePlug and the UPA units -- except when I ran a hair dryer (considered a prime source of electrical noise) from the same receptacle. The HomePlug units didn't appear to notice. Throughput on the UPA units slowed to a crawl. Of course, hair dryers usually stay in other rooms, and at no point did I notice interference from any other household appliances or gadgets.
What did happen on a day-to-day basis was automatic network configuration when more than two powerline adapters were used at a time. When three or more are in use, they form a sort of client/server network. The first adapter would be the server or master, and the others would function as clients. That meant the first one could be plugged into a port on a router, and the others could be plugged into various PCs, which then automatically networked back to the master and through it to the router and out to the Internet.
The HomePlug units networked together regardless of brand, and units from any of the three vendors would serve as the master, with units from the other vendors serving with it as clients. I wasn't able to try that with the UPA units (because I had only two) but was informed that they also automatically configure themselves into networks.
The downside is that I couldn't get two point-to-point connections working simultaneously, such as connecting from two ports on a router to two different PCs. The units appeared to interfere with each other and throughput slowed to a crawl.
Meanwhile, the UPA units and the HomePlug units interoperated only grudgingly. When UPA adapters were already operating and I plugged in HomePlug adapters, the UPA adapters continued functioning normally. The HomePlug devices, presumably treating the UPA signals as noise, slowed down. When HomePlug adapters were already operating and I plugged in UPA units, the UPA units didn't function at all, while response times for the HomePlug units were about four times slower.