Skeptics with a K: Episode #141


Exploding viruses, bad luck, zapping blood and the Emperor’s New Music Player. Plus obese mice, broken phones, stool transplants and the importance of back-ups. Pouring over hard work, it’s Skeptics with a K.

Play
  1. #1 by Phil on February 22, 2015 - 17:38

    As always your shows are interesting and quite often informative.
    When talking about the new music player, were you confusing sampling rates with the frequency of the notes in the music? My impression was that you think they are interchangeable. We can hear frequencies in the 20hz to 20000hz range. If the note being recorded is outside that range it matters not if you are sampling at 44000hz or 190000hz it still won’t be heard

  2. #2 by Mike on February 22, 2015 - 17:50

    Hi Phil,

    As mentioned on the show, the underlying science here is the Nyquist-Shannon sampling theorem, which states “if a function contains no frequencies higher than B Hz, it is completely determined by giving its ordinates at a series of points spaced 1/(2B) seconds apart.”

    Or, to put it another way, to accurately reproduce frequencies in the human audible range (20Hz – 20kHz), you have to sample the waveform every 1/40000 seconds — aka 40kHz.

    You’re right when you say if the note recorded is outside the range 20Hz – 20kHz then we won’t hear it, no matter the sample rate. However, if your sample rate is < 40kHz, then you start to lose some of those audible frequencies -- because your sampled data is insufficiently granular to accurately reconstruct those frequencies on playback. Does that make sense? Mike

(will not be published)