Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Tangentially related: I have a vague memory of reading somewhere that the PCM sampling frequency for CD Audio was decided between Sony and Phillips by way of a surf contest. Their respective frequencies of choice were such that they could fit some arbitrarily-long piece of classical music onto a single disc so they decided to battle it out on the waves (the conference where they were hashing this out was in Hawaii). Phillips supposedly won so we got 44.1 kHz.

I just did a cursory web search to find this anecdote and was unsuccessful. Did I make this up whole cloth or is it just buried someplace? Or was I bamboozled at a young age by some random forumite on a now-defunct site?

*EDIT: This comprehensive account [1] seems to confirm that the story is completely apocryphal.

[1] https://www.dutchaudioclassics.nl/The-six-meetings-Philips-S...



Sony won, not Philips. It seems that the rationale is like so: a higher rate would not be compatible with both NTSC and PAL VCRs, and a lower rate would shrink the transition band for the antialiasing filter (or alternatively, reduce usable audio bandwidth to an extent that encroaches on the generally accepted limit of human hearing). Although the latter hardly seems relevant when the alternatives being debated were so close (Philips' 44056 Hz, for example)!

https://en.wikipedia.org/wiki/44,100_Hz#Origin


> a lower rate would shrink the transition band for the antialiasing filter (or alternatively, reduce usable audio bandwidth to an extent that encroaches on the generally accepted limit of human hearing)

I've seen people smarter than me argue that the ideal sampling rate is actually somewhere around 64 kHz because it would allow for a gentler anti-aliasing filter with fewer phase artifacts.


Why couldn't they make use of a sampling rate with five samples per frame (which would exactly give 88.2kHz by the way)?


Because doubling the sample rate would half the playing time of a CD, or require twice the density, while not bringing any audible increase in quality for the human ear.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: