The comparison with the digital or film camera is not completely random: the sampling frequency of the audio signals, that is, the frequency of the samples per unit of time (usually given per second), is comparable to the frame rate per second from a film camera. The number of pixels in each individual image could be equated with the bit depth: HD movies “look better” than Super 8 movies. The higher the number of pixels on the sensor and the more often a photo is taken, more precisely, the “light to be recorded”, the landscape, can be digitally reproduced.
Bit depth
Fortunately for us, a certain Harry Nyquist inspired a certain Claude Shannon long ago to support him with a theorem (a theoretical statement or theorem) that stated that an audio signal at twice the frequency must be sampled uniformly to match. with the original signal. to be able to rebuild sufficiently. Limiting the bandwidth of audible frequencies practically frees us from our hearing, which is basically only capable of consciously perceiving frequencies between a maximum of 20 Hz and 20,000 Hz.
The expense of completely and exactly reconstructing the analog output signal is theoretically infinite, since digital signals are discontinuous by nature in any case, while analog signals are always continuous. Unfortunately, it is inevitable that digital information is only suitable for rough storage of analog signals. The starting signal is “rough”, good word, right? Nyquist’s theorem also applies to digital cameras: they also deal with frequencies, that is, those of light.
digital audio
For signals up to 20 kHz more or less relevant to humans, a sampling frequency of 40 kHz is sufficient according to the aforementioned theorem. The 44.1 kHz sample rate common for CD quality comes from the 1970s or Sony’s “pulse code modulation (PCM) process for storing digital signals on video tapes. Later, Sony developed the Red Book standard for audio CDs with Philips.
The frequency, which is slightly wider by an additional 4000 Hz than twice that audible to humans, has its origin in the simplest possible filters, which are intended to remove so-called aliasing effects from the audible range of the reconstructed analog signal. during digitization: the wider this “corridor”, the simpler the filter technology.
PCM pulse code modulation method
Exactly 44.1 kHz got out of this, because sample rate converters can be more easily designed (used for studio technology or data carrier transfer) if the sample rate is an integer multiple of the output frequency. The output frequency here was the 60 Hz network frequency used for video digitization with 525 lines to digitize the TV signal. Changing 60 Hz would have been very laborious, it stuck. It is not a coincidence that multiplying 525 by an integer factor results in a frequency greater than 44,000 Hz, which we want to achieve to keep filters for anti-aliasing simple: the next largest integer that is divisible by 525 is 44,100. The multiplication factor is 84, as a whole number is desired, which should not interest us otherwise.
