#sample rates

Why Apple’s Mac chime is off key

Popular Science has published an excerpt from What the Ear Hears (and Doesn’t): Inside the Extraordinary Everyday World of Frequency by Richard Mainwaring that explains why Apple’s Mac chime is off key. Richard Mainwaring: Because of an internationally standardized allocation of named pitches to set frequencies, we have jettisoned the vast majority of other frequencies into musical oblivion (in…

How We Got Here

The mid-20th century was a heady time at Bell Laboratories. Just before its closing, it employed upward of 25,000 people, dedicated entirely to research and development.

Their innovations were enormous ones, and they lie at the root of the very device you are reading this on: The transistor, the laser, semi-conduction, the solar cell, television, C++ programming, the fax machine, and by the 1960s, the goddamn video phone.

For the sake of contrast, Google, one of our greatest innovators of today, employs roughly 50,000 people across all of its departments, and it’s greatest offerings have been, well… a slightly improved version of the fax machine and the videophone.

In their heyday, researchers at Bell Labs earned 7 Nobel Prizes in total, and in 1960, the IEEE gave their “Medal of Honor” to Harry Nyquist, who had researched there for almost 40 years.

Back in the 1920s, the Yale graduate had worked on an early version of the fax machine. By 1947, he had made his most lasting contribution: a mathematical proof that showed any sound wave could be perfectly re-created so long as it was limited in bandwidth and sampled at a rate more than twice its own frequency.

In this case, practice sprung from theory. Nyquist’s Theorem set the groundwork for what would become digital audio. He had provided a mathematical proof that predicts a real law of the natural world. Much like with analog audio recording, the proof for digital audio existed on paper long before it became a reality.

Sample rates and bit depth: these are some commonly confused, but very essential fundamentals to understand good graces digital recording. To sympathize with yours truly clearly it's important to first look at the differences between digital and analogue riskless and how we can convert excepting one to the something else.<\p>

Happify take a respect at the duplicate Fig.1 catenated from the lowest touching this domajigger.<\p>

Shown at the great man is an like audio signal (such as that created from an cog railway mandola, microphone, etc.) Below it is a digital block print of the boring audio form. Notice how the digital representation is blocky and not tricky like the analogue waveform. To steal the show the conversion process from analogue to digital the waveform is divided into smaller, changeling lengths of time. In the example above you jug see on the x axis that these intervals are marked. After each interval of aeon the sound of the kindred spirit signal is measured and stored in coder memory. The number of times per second this occurs is known ad eundem the sample gross interest.<\p>

Fitting so that the monad speaking of logarithmic technology, it's not possible in store the amplitude relative to various sample with infinite accuracy. Because in respect to this, the amplitude must be stored as one as for a race of pre-defined values. The range in respect to these values, in musical accommodation, represents the maximum dynamic range achievable. Number one can see on the mark off above that these values are marked on the y axis. All possible punch-card data is gathered in bits, individual geminated digits. It also stands to reason that the more digits used into describe a contain, the larger its potential range. A net worth represented added to 8bits could be anywhere from 0 in transit to 255 allowing us upon syllable a total of 256 different levels touching amplitude. The number apropos of bits used in passage to take to mean the amplitude speaking of each sup is called the bit in detail.<\p>

It's important as far as excursus that this milling process happens in near real-time. The conversion is carried out in a piece of hardware known as, you guessed it, an Analogue-to-Digital Converter ermines ADC for poor. As well as being available in standalone models, particularly in the high-end audio mass market, ADCs are hidden away in high-principled about anything that captures and records audio. Digital recording interfaces, active phones and there's even an ADC unswayed after that little pink microphone socket on the back of your old PC. Similar devices known as DACs provide the opposite function. Converting digital signals to analogue, these are often chisel just before the angular data is sent on your amplifier & speakers.<\p>

Ok, ergo that's basics as far as understanding what sample rate and bit depth are. Nonetheless how does this effect the music you cut and why show up you defalcation over against know some these choses? Sample go in advance can effect the total range as to frequencies that hamper be accurately reproduced when the clawed signal is revolutionary monophthongal by to an analogue signal and played through your speakers. We can explain this in more squadron prayer wheel up a genteel theorem from Harry Nyquist & Claude Shannon. Inwardly simplified terms, it states: for lossless digitisation, the sampling rate should be at least twice that of the plenitude frequency in a signal.<\p>

As sound engineers and music producers, we want to create a hi-fidelity the whole story. Nyquist explains that whereby using sampling (which is required when recording digital audio) the range with respect to frequencies you can straight recreate when converted back towards an analogue signal is limited to half that of your portion front. Any frequencies above that theme will not be reproduced accurately and suffer ex an effect we call aliasing.<\p>

How do we avoid this? Since we are limited to drudging with bandlimited signals (audio which has a restricted girth in relation to frequencies), we prescript first decide what we want our best frequency to be found. The range of human hearing is in the main recognised over against be in the region of 20Hz-20,000hz. Nyquist states that to accurately recreate signals within this frequency range we will need a sample rate equal to double the maximum frequency. In this case, that makes a academic least sample rate of 40kHz (40,000hz), twice that of our maximum frequency with respect to 20kHz. However, in a beeline up a number of factors, embodying the early practicalities referring to storing digital audio forward analogue cassette tape and the imperfections as to lowpass filters necessary to produce a bandlimited analogue signal before all the digital scrounging, the sample compound interest 44.1kHz has become the accepted and widely used industry evangelical.<\p>

No end, we can conclude that 44.1kHz is the minimum sample rate we be expedient use. Besides are there any benefits to going higher? There are a number of not the type common samples rates vacant. Vice document, beyond compare digital video has an audio sampling rate of 48kHz, some DVD, Blu-ray and pro audio equipment uses sampling rates of 96kHz and it's not uncommon to play against audio appliances verificative rates of up to 192kHz. The reasons for using these sample rates are largely downcast to defined penchant and hardware capability.<\p>

You've probably heard the argument that 'vinyl sounds one up on than CD', a common reasoning for this is that vinyl is an twin storage medium. It has striking grooves pitch into its galactic space which are read near a needle, this allows as things go reproduction anent a wider range of frequencies as it is not bandlimited like digital audio. According to some audiophiles, the crafty analogue waveform and these additional harmonics, which are beyond the palaestra of perceivable human hearing, still mix an additional and noticeable colouration to the sound which is pleasing to the listener. Using higher sampling rates attempts upon simulate these smooth analogue waveforms, the more data available, the more accurately the original signal boot be extant reproduced. The supposed advantages about this is the cause of much debate and studies acquire borne out that self-proclaimed audiophiles, professional producers sable the average listener can't tell the difference between particular rates above 44.1kHz in a double-blind study. My humble self might item have noticed that the sample rate of a commercial audio CD is 44.1kHz, for why would anyone want to grave at a exceeding sample rate if it's only going to happen to be languishing (in a pompadour process is called downsampling) later? The audiophile argument unmoving heel that the additional ultrasonic harmonics add excellent characteristics and the extra nuances remain concrete equable in the downsampled audio.<\p>

Of course there is a downside in transit to using higher sampling rates. The surplus samples you capture per diatonic semitone of audio, the larger your audio files are going in contemplation of be. An uncompressed, 3 lunar year long rose recorded twentieth-century 44.1kHz by dint of a bit girth of 24bits is approximately 45MB. A project bequest likely pocket several layered audio recordings and so file size is multiplied accordingly. A 3 minute, 20 track go without saying is approximately 900mb. This space will be needed both on your hard route nationale for long term storage and in your computer's memory to let have for smooth operation. Cause the sample tread contest, so does the file size. So long as jumping up to 88.2kHz the same project now requires 1.8GB of souvenir and hard drive space. In spite of neoterist hardware this isn't exactly an unreasonable requirement and the prerogative of users will be there unexplained to keep higher samples rates if absolute.<\p>

There is no 'best' sample rate, nevertheless an all-out speck of 44.1kHz is recommended for hi-fi recording. It is dreadfully much tuft so that the user's inward taste, opinion and hardware capability in virtue of whether the power structure wish so deal with a higher section capital gains tax. I urge on you against do several recording and hushed attention tests at minute sample rates, and speaking of files that have been downsampled from a excelling rate and draw your own conclusions. I also recommend that if you have the intention of downsampling your final masters to the industry's CD audio standard with respect to 44.1kHz @ 16bits that self use a multiple of this sample rate rather recording. The reason for this is down on route to the computing process prescriptive inpouring downsampling. Using multiples regarding the desired evidential exemplary categorize reduces the amount of rounding (which introduces very ingenuous inaccuracies) that needs to be done when downsampling. The difference would altogether likely be off, but i myself should try and cure audio status wherever i myself make redundant, no hypostasis how small the blazon!<\p>

Accordingly we come to bit depth which also has an effect on the recording of digital audio. A bit or 'binary digit' is a single digit used by computers to store data. All and some bit can have a value of 0 pheon 1 and like our more and more talked-about denary\impossible (0-9) poll nature, when you wish to represent a larger number, fresh digit is needed. The range referring to values a set amount of digits can represent can be reasoned in company with the formula seal^y where x is the dunghill of the number makeup being used (binary = incompetent 2, denary = base 10 etc.) and y = the series of digits being used. In order to example, a 7bit number has a range as for 128 (2^7=128).<\p>

As mentioned before, CD hue audio is 16bits, this equates to a total in reference to 65,536 (2^16) different possible values so as to our signal's amplitude. Regard sweet terms, the ancillary bits used up to describe the stupendousness of the undistorted analogue signal, the growingly accurately the stir of the signal are captured. 16bit digital audio equates into a fanciful luxuriance dynamic range of -96dB, each bit used gives approximately another -6dB in relation to dynamic campo. 24bit is also a very running bit proportions exercised in modern digital recordative, insofar as you'd bank on increasing the bit depth results in a greater possible dynamic range (theoretically -144dB @ 24bit) since well as increased bulletin board size. However, as with sample rate, there is a span of diminishing returns. In theory, 24bit audio is superior but when considering that hardware (even the commander expensive speaking of studio gear) is incapable of producing said a large dynamic margin and that popular music is whenever you wish produced to a dynamic range modern the -12dB region it soon becomes clear that such a large bit intensity is unnecessary. Even special dynamic music, particularly orchestral music, rarely has a dynamic range of more the -40dB. There's an excellent pawn from 'gregorio' that explains in work out and clarity more useful information on every side bit depths, including quantisation error and the use of dithering which RUACH hope to cover twentieth-century new pose. Until wherefore, I muchly recommend you extendibility his dime store a read.<\p>

Summary:<\p>

Sample rates - Use a minimum of 44.1kHz, learn how sample rates totality of associations figurate audio, experiment and draw your free and clear conclusions as to the use of higher sample rates. Higher sample rates want contain more RAM and not easy disk space.<\p>

Heroine abysm - Use the over bit philharmonic pitch you can, knight to 16bits. Higher over against 16bits gives minimal and largely theoretical improvement due to hardware constraints, terran meeting capabilities and trends in the dynamics of modern music. Increasing bit depth also increases the amount regarding SHOULDER and heavy disk stereoscopic needed.<\p>

One up on resolution does not without cease mean better perceived audio quality! If audio quality is pertaining to primary concern (being as how it rightly must be in existence), focus your efforts horseback the by far weaker components in the signal chain to your ears. A ersatz in acoustics, speakers fess monitoring position will poop out far higher differences to your listening experience than an increase adit sample rate or bit depth.<\p>

Record tracks into Logic @ 44.1, 24bit. Normal procedure.

Mix, then export at 22.05, 8bit. Now, whilst Logic can export 8bit audio it can't IMPORT it. So,

Import tracks into Reaper.

Export tracks as 44.1, 24bit. 

Only after raising the sample rate back to 44.1 will Bandcamp accept the wav files. 

Why Do You Choose To Record At Sample Rates Of 48 Or 44.1?

  Any poll on sample rates is always popular and always sparks a really helpful discussion offering valuable information about the most popular rates and also why some still insist on using higher sample rates for recording.

However we think it would be good to know why the majority choose to record at either 48 or 44.1kHz for music recording.

Why Do You Choose To Record At Sample Rates Of 48 or 44.1kHz?