Why do we use 44,1 kHz frequency on CDs while we’re only being able to hear up to 20 kHz?
In: 38
Edit: This was supposed to be a reply to another comment asking specifically why 44.1 rather than 40kHz, but I accidentally hit the button for a top-level reply.
44.1 kHz largely comes from how digital music was stored and exchanged *before* it ended up on the CD.
Compact discs came out in the early 80s, when computers were still rather primitive. Usable digital audio workstations (DAWs) wouldn’t really be a thing for another decade or so, and even then they were expensive dedicated hardware devices rather than programs that ran on normal computers. CD burners wouldn’t show up until closer to the end of the 80s, and it would be another few years before they were really practical. The only way to actually make a CD was the way they were mass-produced—making physical masters and stamping out copies.
So for the first decade and a half of the CD’s existence, you weren’t exactly recording your album on your MacBook and uploading it to Bandcamp to get your CD made. Nor could you burn it to a CD-R yourself. How, then, did studios send their digital recordings to the manufacturer to get turned into CDs? If you were to use floppy disks (used by the computers of the day) you’d have to mail over a thousand disks in for just one CD.
It turns out that there was a fairly common technology around at the time that was pretty good at storing large amounts of data on relatively cheap media: the VCR. To store digital audio on a video tape, a device called a *PCM adapter* was used. This converted the bits of data representing the digital audio to black-and-white dots in a video signal, which could then be plugged into a normal VCR and recorded onto a video tape that you could mail to the manufacturer, who could then read this special video signal off of the tape and turn it into a CD.
In this process, it’s useful to structure the black and white dots in the video signal so that there are nice round numbers of samples in each frame of video and so that no samples need to be split across lines. To make things more complicated, VCRs in North America and Japan used NTSC, which is 30 frames per second, while most of the rest of the world uses PAL at 25 frames per second. It would be annoying if CDs played at different speeds depending on which part of the world they were made in, so the chosen sample rate needed to fit evenly into both NTSC and PAL video frames in some reasonable way.
And that’s how we ended up with 44.1 kHz. It’s above the 40kHz Nyquist frequency to cover the whole range of human hearing, but divides evenly across video frames and lines in both PAL and NTSC. In fact, 44100 is the product of the first several prime numbers (it’s 2x2x3x3x5x5x7x7) meaning that you can divide it by just about any small number without getting a fraction, which was also useful for converting to lower sample rates before we had fancy digital signal processing (DSP) chips.
A lot of this is simplified, for further reading see Wikipedia: [PCM Adaptor](https://en.wikipedia.org/wiki/PCM_adaptor) and [44.1 kHz](https://en.wikipedia.org/wiki/44,100_Hz)
Latest Answers