Eric Hocking wrote:
> Bill, perhaps you could settle something I seem to remember regarding
> digital CDs. I seem to recall that *all* audio CD recordings clip
> upper and lower frequencies. This gives the impression of a 'crisp'
> sound, but, due to the digital recording process frequencies are
> invariably lost. As you say, the sound quality of the analogue
> recordings would then be superior.
> have I got this right?
Technically the frequencies are clipped, but for audio CD's this
is not an issue (frequencies are filtered above 20kHz, which is
beyond the range of human ears.) I'm not sure about low frequency
filtering, but this is normally limited by the performance of your
In computers, the quality is limited by the sampling rate (how many
measurements per second), the sampling accuracy (8 or 16 bits per
sample) and any compression algorithm that is applied. For audio
CD quality, you need to make a 16 bit samples at 44.1kHz and no
compression; in stereo, this requires 10 MBytes per minute of
My CD-ROM drive is on the fritz, so I haven't been able to see what
settings have been used by the Webster's Australian Birds CD. I'd
guess they are mono recordings at 22kHz; the highest frequency this
will capture is 11kHz (1/2 the sampling rate). The recordings probably
use 8 bit samples, but maybe 16 bit; they sound okay, but may be of
questionable value for analysis.
When you consider the video footage, pictures and software that also have
to fit on the CD-ROM (650MB of storage), compromises have to be
made somewhere. For comparison, the US "Stokes Field Guide to Bird
Songs (Eastern Region)" has digital recordings of 372 species, which
have been squeezed onto three audio CD's. Each CD uses the maximum
99 tracks; some tracks contain two species. This makes it very easy
to call up a particular bird (no pun intended.) The Western Region
set is in development. I'd love to see a set of CD's with digital
reference set like this of Australian birdsong.