Favour wired over wireless
July 10, 2011 in hardware by Dan Gravell
The rise of wireless (WiFi) networking now sees it as the de facto choice for home networking. However, there are certain applications relating to digital music, namely synchronisation and re-organisation, that really require a wired network. A useful and happy medium that combines speed with ease of use is powerline networking.
The reasons for wireless' popularity are obvious. The promise of no unsightly wires to look at or install means that wireless networking appears easy. The time from hooking up your new wireless router to browsing the Web is very short. Improvements in the technology over time now mean streaming standard-definition music and video is possible.
That's not to say wireless networking is without its drawbacks. Reliability and bandwidth can vary depending on signal strength of the various computers on your network. The signal strength in turn can be affected by the building in which you house your network and the presence of other wireless networks, which can interfere with yours.
So much for wireless's general drawbacks. But what about music networking over wireless? The good news is that, in general, wireless is fine for streaming music to play. In fact, it's ideal; it's easy to find a wireless audio streamer to plug into your speakers and begin streaming audio throughout the home with little effort. Things get more difficult when we begin to discuss high definition audio, such as that from DVD-A discs or higher sample rate audio, but for most applications playing music over wireless is just fine.
The reason playing music is easy for wireless networks is because it's a form of streaming. Streaming only needs access to the music you wish to playback at any moment in time. So, for one second of music, you need to be able to transmit one second of audio, however many bytes that translates to, over the wireless network within that one second and to be able to keep doing it as the audio plays back. Simple. (Music players typically work with an additional buffer to smooth the increases and decreases in bandwidth that just happen, but the same idea applies).
It's when we need access to the entirety of a given track of piece of audio at once that things begin to get difficult. This implies the entire bulk of a track being transmitted across the network. So instead of the requirement for accessing one second of music within one second, we need to access the four-odd minutes of music in a typical popular music track, and do it as fast as possible. Fans of Yes will be left twiddling their thumbs.
What type of applications require whole-track access? Two that spring to mind are synchronisation and re-organisation.
The burden of synchronisation
Synchronisation is the movement of audio between two devices and is most commonly done to equip mobile music players with music. Typically, although not always, there's a 'master' source of music. This might be the music on your computer or your NAS. Then, there's a 'slave', which receives the music and is often your mobile music player, be it an iPod, iPhone, Android based phone or alternatives.
Obviously, for the mobile music player to receive all the music that is transferred to it then it must receive the entire music file. There are two possible ways wireless can involved here. First, a wireless network could be used to transmit the files from the source to the mobile player. Second, the source may be connected by a wire to the mobile player, but the music files may be accessed by the source over wireless.
Either way, if the music is being transmitted over wireless it will take much longer to synchronise your mobile device.
Getting organised is slower over wireless
The other application was music re-organisation. Re-organisation typically means tagging your music files.
Tagging requires that the entire music file is accessible to the computer performing the tagging. Therefore, the whole file must be read in across the network. Then, after the files have been changed, the file must be written back, the whole file again, back across the network.
This forward-and-back communication is costly, and is effectively the equivalent of synchronising music twice, once reading it in across the network, and a second time writing it back. (As an aside, this is why bliss is written as a server side application - by installing it on your home server you save a lot of bandwidth and re-organisation is much faster).
Electric avenue
Where possible, if you have access to a wired Ethernet connection you should use it in preference to wireless networks. And, I suppose this goes without saying, make it as fast as possible; gigabit networking is now affordable for the home and should be enabled on all new networking purchases. Failing that, 10/100MBps networking is still preferable to wireless.
For homes without this luxury, which is probably roughly 99% of all homes, an alternative is powerline networking. Powerline networking adapts your electric circuits into a computer network. Aside from the cable from device to wall, this offers the same advantages of wireless networking, but the performance is better, typically falling between wireless and wired networking. These networks tend to be less susceptible to drop-outs and interference too.
Like wireless, powerline networking has been advancing in speed and reliability over the past few years. Also like wireless networking, their stated performance figures are wildly optimistic, but they still perform better than wireless.
I find the happy medium is to keep wireless links between music players and my music, while I network (using powerline networking) my computers together. This is because my computers are generally responsible for synchronisation and re-organisation, while my music players are responsible for... well, playing music.
Thanks to the improbably-named UggBoy♥UggGirl [ PHOTO // WORLD // TRAVEL ] and Elsie esq. for the images above.