If you’re reading this article, then I assume you have a reasonable understanding of how CDs work and how they store music. However, here’s a quick recap:
Rather than storing music by analogue means (constantly recording the audio), CDs hold digital audio. Anything digital is comprised of data, which is stored in binary (so discrete specific values; either 1 or 0). What this means is that, when recording audio, instead of storing an exact representation of the music, CDs sample the sound at a (very) specific moment in time – they do this 44.1 thousand times per second. The samples of audio are then played back, and because there’s so many of them per second, there’s no discernible difference between the recreated audio and its analogue counterpart. However, this method of storage leads to rather large files.
What the MP3 system does is to compress this audio in a “lossy” way. This means that the encoder actually forgets about some parts of the audio in favour of more relevant parts. The parts it forgets will be different for every track. For example, an acoustic guitar cover will contain a much different audio spectrum than a heavy metal track.
So how does the encoder choose which parts to keep? Well, basically by following a set of rules called an algorithm. It processes the audio at the millisecond level, and continually “listens” to each part and decides, amongst other things:
- Are some parts hard to hear by humans (i.e. close to or over 20 KHz, which is at the upper limit of human hearing. Or are there parts close to or below 20 Hz, which is the lower limit). If so, they can be stripped out.
- Some sounds are easier to hear than others, so if there’s an easier to hear sound going on at the same time as a less easy-to-hear sound, the less-easy-to hear one will be removed.
- If certain parts of the audio are much louder than others, and playing at the same time, the quieter parts will be stripped out because they’d be hard to hear anyway.
The MP3 encoder goes through the entire section of audio, stripping out the parts that aren’t strictly necessary. The lower the bitrate, the more information will be stripped out.
What does this mean in terms of playing back the audio?
It really depends on what you’re listening through. The lower the quality of audio playback device, the less a low bitrate will matter. For example, if you’re listening to an mp3 file through some cheap ear phones or portable mp3 speakers, you probably wouldn’t be able to tell the difference between an 128 kbps file, a 320 kbps file, or a pure lossless audio file.
The quality level in lower bitrate files is much more noticeable when they’re being played through higher quality speakers. So if you have made the investment into high quality speakers, it’s usually advisable to encode your mp3s at as high a bitrate as possible.
Related posts:
Comments on this entry are closed.