What is Compression?
It’s hard to absorb how many ways data compression comes into your life today. For anyone involved in the media industry, the terms “compressed video” and “compressed audio” are thrown around like spaghetti. But what is compression actually? Why is it so pervasive? Let’s start with the basics.
First of all, I should point out that the idea of compression is by no means new – it’s been around since the earliest days of electronic communications. The reasons for it then are as valid as they are now. Today, we talk about digital compression; then it was analog compression. No matter: the needs and basic techniques remain the same.
Simply put, any kind of electronic communication has inherent costs. There are the costs in creating it in the first place – capturing sound waves via a microphone has a cost: the cost of the microphone, the costs of the wire. There’s the cost in storing it: tape decks cost money. Tape costs money. Hard drives cost money. There’s the cost of delivering it: loudspeakers cost money. Video displays cost money. Software costs money. Almost all of those costs are fixed. The cost of a specific microphone doesn’t change day by day (let’s ignore the principles of accountancy and depreciation here). In order to be able to record a video or audio signal, there is a fixed cost to the device that you simply have to pay (in the case of cable television, part of that cost is the wiring to and throughout your house – everything carries a price). The ability to transmit video or audio or text to your cell phone involves the cost of the transmitter, plus the license for the wireless frequency the data is transferred on.
Let’s be clear: every part of the path from source to consumption has a cost. And that’s why we care about compression.
For the sake of argument, let’s assume we need to transfer 1000 bytes of data from point A to point B, and we need that transfer to happen in 1 second or less. That means we have to engineer the path from point A to point B to have a minimum throughput of 1000 bytes per second. There’s a fixed cost to creating that path, and as described, it can only carry that one piece of data at a time. If we could find a way to squeeze that 1000 bytes down to 500 bytes, while – and here’s the major point – retaining its intelligibility at the other end, then we can actually transfer two pieces of data in that same period. The cost of transmitting each piece of data is halved. But that’s not all – if the piece of data now only takes up 500 hundred bytes, then we can store twice as many of them on a specific size of storage medium. The cost of storing each piece of data is halved.
That’s why we have compression: providers can deliver more data, more audio tracks, more movies per unit of cost.
But it’s not just movies and audio. You look at a graphic on a web page, the graphic is compressed to save storage and transmission costs (it costs money to move data over the internet, you know!!!). You read a .pdf document – the entire document is compressed for the same reason – there are other reasons, too, but for sure compression is involved.
I have to restate the main principle here though – you’re going to reduce the size of the data you’re transmitting, but you have to be sure its intelligible at the receiving end: the video still looks like video, the music still sounds like music. And you can still read the text on the scanned image. That’s where the technology comes in, and it turns out that in practice, most of it is at the transmitting end. We’ll talk more about that later.
Why should we care about compression? Because compression is everywhere!
Want to learn more about how we do compression and transcoding here at Telestream? Read more here!