[Audio Network Information] 1. The process of sampling rate conversion is simple <br> The digital audio heard by general music listeners is 44,100 samples per second (44.1kHz). If this is not your production sample rate, then it is necessary to convert the sample rate. There are two ways to convert the sample rate, but no matter which conversion method, as long as the common multiple between the sample source and the target sample rate is smaller, the result will be better.
"synchronous" sample rate conversion:
It establishes a temporary high sampling rate by using the least common multiple between the sampling source and the target sampling rate, and then down-converts (filters) the high sampling rate data to a low target sampling rate. Using a high sampling rate as a medium, it is possible to avoid uneven conversion or residual data.
If the original material is 96 kHz, to reduce it to 44.1 kHz, you must first increase the sample rate to 14.112 MHz. This process consists of three stages of digital filtering, each of which becomes a potential hazard for frequency response and finishing delay distortion.
If the original material is 88.2 kHz, or any multiple of 44.1 kHz, then only one stage of filtering is required to convert to 44.1 kHz. Regardless of the quality of the filter you are using, less filtering means less distortion.
Sample rate conversion for "out of sync":
It uses separate converters and clocks during the conversion process to reduce the sample rate. The simpler the mathematical relationship between the sample source and the target sample rate, the less likely the sound quality will be compromised, and the same is true here. For example, 2:1 is simpler than 160:147, and the conversion loss is small. The less calculation, the less error tolerance.
2, the upper limit of the performance of the digital audio converter is 20kHz
In Nika Aldrich's well-documented article "Digital Audio Explained: for the Audio Engineer", he reminded us, "We can't hear frequencies above 20 kHz. We can't hear the effect of audio above 20 kHz." What are the benefits of recording at a sampling rate higher than 44.1 kHz?
Although the determined sampling theorem already exists, in reality, the Nyquist frequency [Note 1] has to be built on the performance of anti-aliasing filters (A/D conversion) and reconstruction filters (D/A conversion). . Although there is a perfect filter in theory, the situation is different in reality.
In reality, our anti-aliasing and reconstruction filters are imperfect, which can lead to aliasing, transient loop phenomena and high frequency phase shifts. These problems begin to worsen as the audio band approaches this limit, the Nyquist frequency.
The Nyquist frequency of 88.2 kHz is 44.1 kHz. This places the worst anomaly part of the filter above the audible frequency band of the human ear. The influential frequency band only exists below 20 kHz. Digital audio sampled at 88.2 kHz produces less aliasing and phase distortion than 44.1 kHz.
By the way, the more efficient the filter, the more expensive it is. So, if you don't have a very advanced converter, then using a higher sampling rate will give you better results.
3, 176.4kHz, the amount of data is too large, the advantage is insufficient <br> If we have a good effect using 88.2kHz, then why not double, use 176.4kHz?
We used 88.2kHz on the audio, which has the advantage of avoiding the bad parts that are caused by imperfect performance of the converter. When we can't hear if there is a difference, it is relatively difficult to make the result better.
And you don't have to think about it, the file size of 176.4kHz will be doubled. If you often move files between devices, upload to cloud storage or manage a large number of automated backup systems, it's clear that this is not the case.
Off-topic <br> It should be noted that according to the above logic, if the version handed over to the consumer is 48 kHz, the audio production can be extended to 96 kHz.
Also note that there are a number of solutions available for recording at 44.1kHz. And what we are thinking about here is an engineering problem, so the focus is on what happens outside of these perfectly designed solutions.
I completed the article "The Information: A History, A Theory, A Flood" with James Gleick. I learned that the theoretical basis of the digital audio system was established between 1928 and 1949. Now, it has become a very interesting thing to argue about the imperfect results caused by these theoretical applications.
[Note 1] The Nyquist frequency (Nyquist frequency) is half the sampling frequency of a discrete signal system and is named after the Harry Nyquist or Nyquist-Shannon sampling theorem. The sampling theorem states that aliasing can be avoided as long as the Nyquist frequency of the discrete system is higher than the highest frequency or bandwidth of the sampled signal.
"synchronous" sample rate conversion:
It establishes a temporary high sampling rate by using the least common multiple between the sampling source and the target sampling rate, and then down-converts (filters) the high sampling rate data to a low target sampling rate. Using a high sampling rate as a medium, it is possible to avoid uneven conversion or residual data.
If the original material is 96 kHz, to reduce it to 44.1 kHz, you must first increase the sample rate to 14.112 MHz. This process consists of three stages of digital filtering, each of which becomes a potential hazard for frequency response and finishing delay distortion.
If the original material is 88.2 kHz, or any multiple of 44.1 kHz, then only one stage of filtering is required to convert to 44.1 kHz. Regardless of the quality of the filter you are using, less filtering means less distortion.
Sample rate conversion for "out of sync":
It uses separate converters and clocks during the conversion process to reduce the sample rate. The simpler the mathematical relationship between the sample source and the target sample rate, the less likely the sound quality will be compromised, and the same is true here. For example, 2:1 is simpler than 160:147, and the conversion loss is small. The less calculation, the less error tolerance.
2, the upper limit of the performance of the digital audio converter is 20kHz
In Nika Aldrich's well-documented article "Digital Audio Explained: for the Audio Engineer", he reminded us, "We can't hear frequencies above 20 kHz. We can't hear the effect of audio above 20 kHz." What are the benefits of recording at a sampling rate higher than 44.1 kHz?
Although the determined sampling theorem already exists, in reality, the Nyquist frequency [Note 1] has to be built on the performance of anti-aliasing filters (A/D conversion) and reconstruction filters (D/A conversion). . Although there is a perfect filter in theory, the situation is different in reality.
In reality, our anti-aliasing and reconstruction filters are imperfect, which can lead to aliasing, transient loop phenomena and high frequency phase shifts. These problems begin to worsen as the audio band approaches this limit, the Nyquist frequency.
The Nyquist frequency of 88.2 kHz is 44.1 kHz. This places the worst anomaly part of the filter above the audible frequency band of the human ear. The influential frequency band only exists below 20 kHz. Digital audio sampled at 88.2 kHz produces less aliasing and phase distortion than 44.1 kHz.
By the way, the more efficient the filter, the more expensive it is. So, if you don't have a very advanced converter, then using a higher sampling rate will give you better results.
3, 176.4kHz, the amount of data is too large, the advantage is insufficient <br> If we have a good effect using 88.2kHz, then why not double, use 176.4kHz?
We used 88.2kHz on the audio, which has the advantage of avoiding the bad parts that are caused by imperfect performance of the converter. When we can't hear if there is a difference, it is relatively difficult to make the result better.
And you don't have to think about it, the file size of 176.4kHz will be doubled. If you often move files between devices, upload to cloud storage or manage a large number of automated backup systems, it's clear that this is not the case.
Off-topic <br> It should be noted that according to the above logic, if the version handed over to the consumer is 48 kHz, the audio production can be extended to 96 kHz.
Also note that there are a number of solutions available for recording at 44.1kHz. And what we are thinking about here is an engineering problem, so the focus is on what happens outside of these perfectly designed solutions.
I completed the article "The Information: A History, A Theory, A Flood" with James Gleick. I learned that the theoretical basis of the digital audio system was established between 1928 and 1949. Now, it has become a very interesting thing to argue about the imperfect results caused by these theoretical applications.
[Note 1] The Nyquist frequency (Nyquist frequency) is half the sampling frequency of a discrete signal system and is named after the Harry Nyquist or Nyquist-Shannon sampling theorem. The sampling theorem states that aliasing can be avoided as long as the Nyquist frequency of the discrete system is higher than the highest frequency or bandwidth of the sampled signal.
Charging Cable Bracelet,Charging Cable For Iphone,Charging Cable,Micro Usb
Hebei Baisiwei Import&Export Trade Co., LTD. , https://www.baisiweicable.com