Fundamentals of data representation: Streaming audio

PAPER 2 - ⇑ Fundamentals of data representation ⇑

← Sound synthesis Streaming audio


Gone are the days when you had to buy music on a wax cylinder, LP, tape, CD or MiniDisc. Nowadays if you want to get some music you can download it using a program such as iTunes or stream it using a website such as MySpace, YouTube, BBC iPlayer or Spotify.

The big difference between downloading audio and streaming it is that when you download something you generally have to acquire the whole file before you can play it, with streaming audio you play the file whilst you are downloading it. There are two types of streaming technology that you need to know, multi-cast and uni-cast.

Unicast connections require multiple connections from the same streaming server even when it streams the same content.
Multicasting broadcasts the same copy of the audio file over the entire network to a group of clients
Exercise: Streaming audio

Why might you want to stream audio instead of downloading it?

Answer:

Streaming audio allows you to play sound as soon as you have buffered a few seconds, downloading might require you to download the entire file before you can play any of it

Name a benefit and a drawback of using multicast streaming:

Answer:

a server only has to distribute one version of a file to serve many people, saving processing power and bandwidth. on the other hand if someone wants to join a stream after it has started they might miss the beginning

Name a benefit and a drawback of using unicast streaming:

Answer:

Unicast streaming allows users to access an individual version of a file accessible as and when they want it. On the other hand it places a lot of stress on the server as 10 people watching a video will require ten times the bandwidth to deliver than through multicast