Audio - Arduino Sketches: Tools and Techniques for Programming Wizardry - page 293

Original
wave pattern

Wave
converted to
digital form

Output wave

Figure 18-1: Digital resolution

As you can see, the analog signal l ows through different values, but the
digital representation creates a “step” effect; the representation of the data is not
precise, and quality is lost. Thankfully, sound cards do not have values that go
from zero to 10; most are 16 bits for a total of 65,536 values. Previous generations
had 8-bit sampling for a total of 256 values, and 256 values are too low for an
accurate representation. However, the 16-bit value of 65,536 is considered to be
more than enough for most audiophiles. This is the quality found on CDs and
even some Blu-ray audio i les. However, the resolution is not the only factor to
take into consideration.
Sound waves are a mixture of different frequencies; the higher the frequency,
the higher the pitch. Humans can normally hear sounds from as low as 20 Hz
all the way to 20 kHz and above. To digitally sample frequencies as high as
20 kHz, the effective sampling rate (the speed at which the sound is sampled)
must be at least doubled or 40 kHz. For typical applications, a sampling rate
of 44.1 kHz is used. A microchip was already on the market that used this
frequency for sampling, designed by Sony Corporation. For professional
applications, sampling was done at rates as high as 48 kHz. 44.1 kHz and
48 kHz are common sample frequencies found on computers, as are multiples
of 44.1 kHz; 22.05 kHz, and 11.025 kHz. 8 kHz was used for a long time for
telephone systems, where audio quality was adequate to understand human
voice conversations. Professional sampling devices can sample at a high rate
for even more accurate results; DVD audio is sampled at 192 kHz, and other
devices can go as high as 2 MHz.