Lets start the story of the chain with a very traditional thing. A thing called wiki-mickey- monkey. 

Sound is not wave, the recording from the voice looks like like it was a wave, but it is not a wave. Voice is born, when air pressure changes in short sequences. In order to give birth to a pressure, you need power and force. Singers for example are known to need big lungs, into where they gather air and create pressure. Pressure they release and modify with larynges, tongue and lips.

In the wave files, the wave looking curves are changes in air pressure. The pressure, which is stored to sound file is relative to previous pressure. The lowest pressure is relative to soundless air pressure. In the recorded curve the high frequencies are smaller peaks over the low frequency. The lower frequency sets initial levels for the higher frequencies.

Air acts like spring. After you compress the air, the air bounces back. And generates an opposite force, which is almost equal to the force with what you compressed the air. The bouncing effect of the air makes the recordings look like they were waves.

Wave research 2017 at Ekr833.Com site has more about sound pulses and their similarities with waves.

1. Microphone

The waterproof evidence from the thing, that sound is not wave, is needed right after you click record button. The sound is recorded with a gadget called microphone.

Microphone is actually a sensor, it registers the changes in air pressure. Inside the mike there is cone and magnetic heart. The heart is surrounded with dense copper-iron wiring. If you open the mike, and you are familiar with audio systems, you find there a very familiar looking system. The microphone is miniature from the speaker.

From the box of the high quality mike, you find a diagram or literal description from the areas from which the microphone captures sounds. The general use mikes captures sounds from all directions. Front, back and sides. The studio mikes captures sounds right in front, and from the very short distance.

The size of the mike makes it impossible for sound to be a wave. If sound was a wave, the size of the mikes which records low frequency basses would be enormous. If voice was a wave, small mikes could never register and record the low frequency voices. Sounds are not waves. The form of sound is pulse. The sound forwards directly from point A to point B. 

The low frequencies in the sound pulse are the strongest. The higher recorded frequencies are smaller changes in air pressure. Lowest recorded changes are relative to the soundless air pressure. The higher frequencies and pressures are relative to underlying lower frequency / pressure. Due this the recoding never shows how loud the music was played and how loud it will be played during the playback.

In this picture you see side crossing from the voiceless air.

In this picture sound pulses are on their way to microphone or to your ears. Yellow dots between the vertical red bars show the air compression ratios in the air. In real life compression ratios varies smoothly, without steps.

2 Recorder

If we follow the lead, the wire from the microphone leads us to the recording device. Computer, tape recorder, phone, etc. The recording device is very simple. When you press the record / call button the recording device starts to record and store voltages and voltage changes, it gets from the microphone.

Inside the reorder the signal from the microphone is directed to the small amplifier / adjuster unit.  With this unit you control the recording levels. Sliders for the recording levels controls this unit and it's amplifying levels.

When you are making digital recordings, the data you get from the microphone is always analog data. In digital recorder there is small A/D transformer. The complete name of the unit is Analog-To-Digital converter. A/D transformer splits and converts the analog data to digital format. Depending on the recording device, it can be before or after the amplifier. In sound cards made for computers it is usually before the amplifier. In the Audio systems, which were originally designed for analog recording, A/D transformer is usually placed after the amplifier.

3 Storage system

Continuing inside the recorder. From the recorder unit, the adjusted and converted signal is sent to storage. For the sounds and voices, five types of storages exists. Three of the types operate in real time, and can be used in recorders. In live broadcasts, like Concerts and Phone calls, the signal is not stored. The signal is sent directly to the playback system.

1. Magnetic tapes : Tapes, C-Cassettes and 8 Tracks.
2. Optical discs : Cd - Dvd - Bd 
3. Magnetic disks : Rotating and non-rotating disks
4. Records : Stone and vinyl records.
5. Tubes : Rolling metallic scrolls, Rolling soft scrolls.

Analog recordings are possible on the magnetic tapes only. Digital recordings are possible to magnetic tapes, optical discs and magnetic disks.

With the storage system you store the recording for the playback, which takes place later and usually with different system. This thing is the key thing in demands for the recording device. The recording speed must always match the playback speed.

For magnetic tapes, there are three predefined recording and playback speeds. The old reels used all three speeds,  4.75, 9.5, 19 centimeters per second. C-cassettes used 4.75 cm/s speeds. In some studio recorders it is / was possible to double the 19 cm/s speed to 38 cm / s.

The first sound recordings ever were made on metallic tubes, which had soft coating, usually made from wax. The tubes are used in mechanic instruments. Mostly in mechanic pianos. The music boxes and a-likes are also based on the very old tube technology. Before hard disks and diskettes came, the magnetic tapes were used to store digital data.

Digital formats

For digital recordings you have also recording and playback rates. The digital format has two definitions, like follows  :  

11 025 Hz - 8 bit - Phone quality
22 050 Hz 16bit - Web Radio quality
44 100 Hz 16 bit - Audio CD quality

There are very many others, too. These two parameters are stored into wave file headers, into predefined positions. Besides these parameters wave header has also one parameter for channel count. Mono (  channel count = 1 ) and Stereo ( channel count = 2 ), are the only widely supported counts for the channels.

When audio data is digitalized, the data is divided to slices. The slices are called samples. The slices have two definitions First number - Hz - tells how many data slices there are for one second recording / playback. Although the unit is the same, the sample count has nothing to do with the frequencies humans, monkeys and animals can hear. The sample count tells how accurate the digital sample data is along the time axis.

Although not directly connected with hearing, the higher the sound frequency goes, the more often you must take the samples. Small overlying changes go smaller and shorter in time all the time.

The bit count is another definition for the accuracy of digital data. The bit count tells how accurate the readings from the microphone are.

The bit count sets the maximum values for the stored digital data. For 8 bit sample rate the maximum value for the recorded signal is 255. For 16 bit samples the maximum value for the stored signal reading is 32767.

When storage system gets a reading from microphone, it checks the highest possible values for the microphone and stored data. Then it shifts the microphone data into the bit rate. After that it stores the shifted value. If you set recording level too high in the digital system, or you shout too load to your megaphone, all stored values are shifted to maximum values. If the microphone sends the maximum value to the recorder, the 8 bit sample will have a value of 255, and 16 bit will be 32 767.

When a request for the playback data comes to the digital storage system, bit rates are shifted to the range and values the playback device uses as input.

? If you want to use more accurate samples than 16 bit, and your system allows it, you should use byte aligned samples, like 24, 32, 48, 56, 64, 96 and 128 bits per sample. Bytes are the smallest programmable units in binary computers.

Analog vs. digital

When comparing digital and analog formats, the digital frequency is almost the same as the analog recording and playback speed. Sample rate is unique for digital formats. 

In tape world 19 cm/s speed is fast enough to store all frequencies human ears can hear. With 19cm/s you can record 20kHz sounds. From digital world the required full-coverage demands are not known. 

In theory with 44.1 kHz audio CD quality you can record 20 000 Hz frequencies, but with heavy demand. The high frequency sound must begin at the very same time than the sample is taken. Otherwise something is lost from the recording, it  is not perfect.

In analog world the relationship between the speed and frequencies ( you can record and store ) is not very clear. There are C-Cassette recorders, which operate with 4.75 speed and can record 20kHz voices. In order to achieve the high frequencies you need a special C-Cassettes which magnetizes the tapes faster and more accurate than common Reels and C-cassettes.

Records and CDs

Records and Audio CD's are made for mass production. They suit poorly for the real time recording. Usually the discs are made and based on master tapes or files. They are pressed / cast in the certain companies.

In the beginning the contents of the master recording is attached to mold / press plate. Then the image from the contents is pressed to soft vinyl or soft CD disc with one hammer-like strike. After that the CD is coated, and vinyl is bagged. Finally the CD / Vinyl is packed and foiled.

CD burner like record burners never existed. Records, stone and vinyl are / were always made from some other recording. 

CD burners and DVD players which would burn the live data / broadcasts directly to CD/DVD/BRD discs are very rare. They use buffers or record the AV-data to magnetic discs first.

Packed audio data

If storage system supports MP-3 and other compressed audio files, the storage system for recorder needs a unit called encoder. Encoder compresses the audio signals before storing the audio data.

When storage system can provide audio data  for playback, system needs a unit called decoder. Decoder unpacks the compressed audio data, before the storage system sends the data into playback device, usually amplifier.

Demand for the decoder and encoder comes from the nature of the sound recording system. The microphone and speakers are counterparts to each other. It is not possible to send packed audio data to the speakers. Speakers do not understand MP3 formats, they want the data which is similar to the data you get from the microphone.

Packing is always the last thing, recorder does before storing the audio data. Packing is also always the first thing playback device does. Due to the speaker-microphone demand, all audio systems are always made for the unpacked audio data.

Recording from line input

When you record sounds and voices from line input, integrated receiver / player, the recording process is similar to microphone recording.

There is small difference in the amplifier which is used. Amplifiers for line inputs are smaller ( in amplifying power )  than amplifiers which are made for microphones.

Recorders read the signals from the microphone. Storage units and controls are quite often packed to the recorder.

The best recorder

The old studio recorders, based on magnetic tapes are still the best recorders in the world. The weak point in the system is in the storage system.

The recordings which are made to magnetic tapes gets worse by the time. First the magnetic tape loses high frequencies, eventually the whole recording vanishes from the tape.

Another poor thing in the magnetic tapes is, that you cannot make copies from them. Each copy you make from the copy of the tape, is slightly poorer than the original source - tape. The magnetic tapes gather background voices from the recorder into them.

CD Discs

Philips did invent transistor, but it never invented method for storing the analog recordings : to the bottoms of the CD discs. Analog CD's belong to the devices and gadgets, which were developed for Deca Tech Computers. 

Side crossing from CD disc. There are only two heights [ on-off ] at the bottom of the disc. The laser beam, which reads the discs, calculates the distances from the light source to the bottom of the disc. The optical discs are based on the reflection, and reflection times. When beam reaches the bottom of the disc / track, the disc mirrors the beam back to the beam-reader unit.

Side crossing from AD disc. In AD discs, for Deca Tech computers, the bottom of the disc / track was analog. It had unlimited number of possible values. The accuracy of the reading was dependent on the accuracy of the beam and it's distance calculation routines. Due to the nature of the deca-tech discs, they were compatible with all AD drives / players. Deca Tech processors had a special multimedia mode, with what you could read analog data.

If Philips and others would have invented  AD's ( analog discs ) instead of CD's, you could have used fresh, brand new magnetic tapes in recording, and store / burn the recordings to the Analog Cd discs. With the system you could have made as many 1 to 1 backups and copies from the original master as you wish.

- - PCM - -

Philips and other sound experts have been very near to the correct answer in the past. They do call the digitalization of the sound waves PCM - Pulse Code Modulation.

But they think that sound forwards as vertical wave. Not as horizontal pulses as they do in real life.

When they start to talk about sounds they never talk about the pulses, all they talk about is waves.

- - Corrected Cones 2017 - -

The wave research 2017 suggests that cones angle has a connection to frequencies. Mike can capture and speaker can play.

The size / diameter cannot have a connection to frequencies. Small microphone can capture low basses. With headset you can hear low bass frequencies. 

Cone's diameter-angle system has a connection to maximum powers and separation capability. 

Wave research 2017 suggests that sound pulses for different frequencies and instruments travels in separate threads. So that required space for one thread grows with power and decreases with frequency.