Your music source

To gain the most benefit from any music you enjoy, listen to the best-quality version you can get hold of. A lossless download — .wav, FLAC or ALAC — carries the complete recording. If you are streaming, choose the lossless or hi-res setting if your service offers one. A 40-minute recording in .wav takes up roughly 400 MB.

 Your delivery system

“Invest” in your hearing when selecting headphones. We recommend either noise-isolating in-ear headphones, as in the photo, or normal headphones that cover the entire ear, both types with a frequency range of 20Hz to 20,000Hz. Basic ear pods supplied with most listening devices increase distortion, have a limited frequency range, and let ambient noise through, which tends to make people turn the volume up. Listening loudly for long periods is associated with tension and with hearing damage over time, so it is worth choosing well and keeping the volume moderate. There are links at the bottom of the page if you would like to read more.

.wav and .mp3 — what is the difference?

.wav and .mp3 are both digital audio file formats. .mp3 gained popularity through file-sharing, internet radio and downloading tracks over the internet because of faster download speeds, but lossless formats such as .wav remain the standard for full-quality recordings.

An .mp3 track downloads faster than a .wav version of the same track because the file is smaller, roughly one-tenth of the size. The difference is achieved by discarding some of the data from the original recording. Once music has been compressed into an .mp3 there is no going back — you do not get the data back by converting it to .wav again.

.mp3 changed our listening and buying habits and offered a cheaper alternative to CDs, but the hidden cost is a loss of quality that is not easy to notice.

Why lossless matters

Lossless files give a good representation of the original music, and .mp3 files do not. With .mp3, data is removed according to what is known as psycho-acoustic modelling. The graphic shows the idea in a basic way. Whole frequency bands are taken out, chosen as sounds that are hard to hear or that the compression algorithm treats as less meaningful.

Psycho-acoustics is the study of how the brain processes sound. This processing happens in the background, and it is not something we have any conscious control over. Our audible hearing range extends from 20 to 20,000 hertz, and we can feel frequencies both higher and lower than that. The data lost to .mp3 compression falls inside our audible range, and our own view, after many years of working with sound, is that the more complete the recording the more there is to receive from it.

There is also some evidence that long-term habitual listening to compressed audio shifts what people come to prefer — see the article “Kids prefer poor quality MP3”.

Read more

What Data Compression Does To Your Music, Sound On Sound — what is discarded when audio is compressed, and why.

Perceptual Coding: How MP3 Compression Works, Sound On Sound — the psycho-acoustic modelling described above, in more detail.

Kids Prefer Poor Quality MP3, Audioholics — on Jonathan Berger’s finding that students came to prefer the compressed versions.

Noise-Induced Hearing Loss, US National Institute on Deafness and Other Communication Disorders — how loudness and length of exposure affect hearing.

Making Listening Safe, World Health Organization — safe listening levels for personal audio devices.