Audio numbers (ex. 16/48)

[Filterlab]

Originally Posted by ra1024

If the human ear can't hear above 20khz, why do they sample above 48khz? 192khz just seems like insane overkill unless you have dogs in the audience.

There's been a massive amount of research done in this area, too much to type in full. The outline of it is that the human hearing may only be attributed to interpret frequencies up to around 20Khz, however the natural dynamic range of musical instruments and standing waves/environmental affects go way beyond that figure. Therefore capturing those frequencies adds to the overall effect of music.

Also, the more resample calculations, the less chance of something important being missed. There's also the halving effect - limitation of resampling cuts the response range of high frequencies in half; i.e. 44.1Khz (CD) tops out at 22.05Khz audible frequency etc etc etc.

Have a look at Townshend's page on supertweeters, there's loads of info on there. http://www.townshendaudio.net/index....d=27&Itemid=81