|
For those of us who aren’t audio engineers, it’s curious how audio systems and processors perform analog-to-digital and digital-to-analog (AD/DA) conversions. And, although the average person would never need to know how to perform these types of conversions, since your equipment will generally do the basics for you, it’s always enlightening to learn about the inner workings of things. And, for those select individuals who would like to perfect their recordings, understanding the processes of AD/DA conversions will be invaluable to you. To start, when you record a sample with your sound card, it actually goes through many stages before it is saved as a sound file on your computer’s hard drive (or wherever you’re storing the files). First, the sound card utilizes a very accurate stopwatch, which is more commonly known as the sample rate. Every time the sample rate has completed a cycle, the sound card looks at the filtered input signal, and calculates how loud the incoming sound is at each precise moment in time. The sound card then feeds that number back to the computer, which stores it in memory—to state it all very simply. Before the sample is saved as a sound file, however, the sound card first transforms the sound coming in. In most, if not all, cases a sound card applies a built-in low pass filter—a digital filter which cuts the high frequencies, and lets the low frequencies pass through. The filter is known as a “brick-wall” filter because it cuts off so rapidly above its corner frequency. And, because it is not possible to filter a signal so steeply without imposing some sort of change as it passes through, these filters were fuel for early criticisms of digital audio recordings. Brick-wall filters tend to be complex, and can introduce all sorts of artifacts, like phase distortion and pre-ringing. Even though the brick-wall filter worked as a solution to a side-effect in digital recording called “foldover,” or “aliasing,” you can avoid using such a steep filter by oversampling, which we will discuss in more detail in part two of this article. Meanwhile, aliasing occurs when a system, moving at regular speed, is analyzing something that is moving faster than the system itself. It is like when you are watching the wheels spin on a passing car, and the wheels appear to be moving backwards. The maximum frequency that can be recorded, with a certain sample rate, is half the sample rate. To minimize aliasing, the sample rate should be much higher than the highest frequency present in the signal. You can accomplish this by either increasing the sampling rate, or limiting the signal bandwidth. Increasing the sampling rate will increase the amount of data which must be stored. Therefore, it is preferable to keep the sample rate as low as possible. On the other hand, limiting the bandwidth will decrease the high frequency content of the signal, which is also not very desirable. To read more, please see the second part of this article (coming soon). Additionally, if you are looking to effectively transfer vinyl records or cassette tapes to CD, and you want optimal quality without having to learn digital recording theory in the process, you may want to consult with a professional. For audio transfer services and support, I recommend a small shop in Brooklyn Park, MN that provides a full range of conversion services, called Two Squares
Two Squares, Inc
6272 Boone Ave N.
Brooklyn Park, MN 55428
763-400-4510
Related Articles -
analog to digital, tapes to CD, tapes to DVD, vinyl to CD, audio transfer,
|
