When we listen to a live orchestra or band, we are listening to analog sound. The instruments make vibrations and those vibrations (a sound wave) are picked up by our ears. If that music is recorded and burned (transferred, written) to a CD, the sound has to be made into digital data. The same is true when sound is recorded to tape. With tape, the data is recorded magnetically, with a CD it is recorded by pits via a laser, thus the term "burned".

When we listen to a CD, the digital data has to be read and changed to analog again so that we can hear it. On a computer, that is the job of the sound card...it changes data to analog (sound), analog to data. That means when we want to record from an external source such as an LP, the sound is fed into the sound card which converts it to digital data which can then be saved in the form of a file. When we save that wave sound, the file normally has an extension of "WAV". The quality of that wave is a function of how frequently it was "sampled"; i.e., how frequently "snapshots" of the sound were taken and converted to digital data. For a much more elaborate and techie explanation, try here. For me, it is enough to know that in order to burn sound to a CD that will play in a standard CD player, the wave must have been sampled at 44,100Hz, 16 bit stereo.

If we want to rip (make a wave file) a CD, there is no need to convert from analog to digital because a CD already contains digital data. Consequently, the wave file is exactly what was on the CD plus a bit of information called a "header" that tells the system that this file is a wave file. Because the data is a duplicate, people generally prefer these digital "rips" to a file made from an analog source. However, because of digital limitations, I'm not sure that one couldn't get a better recording from an LP. I don't know, just wondering...

We can also rip a CD to MP3 files. A better term would be "rip and encode" because that is what is happening...the ripping program is extracting the digital wave data and then sending it to something else to encode the wave to MP3. That said, I think it better to do the two as separate steps. It is less taxing on the system and can help avoid glitches and noises in the resultant file. Doesn't really take any additional time either. Unfortunately, in our "I want it now" society, all too many are unwilling to take the time to learn exactly what it is they are doing and how to do it. The fact that you are reading this means you aren't one of "them"...a strong indication, at least...:)

Wave files are quite large - around 10.6 megabytes for each minute of music. A few years ago, a German company named Fraunhofer devised a way to compress those waves thereby making it feasible to store quantities of music files on our computers. Exactly how that compression is done I do not know but it is a "lossy" compression; i.e., some data present in the wave file is thrown away. The more that is thrown away, the smaller the resultant file - called MP3 - becomes. The degree of compression is determined by the bit rate...the lower the bit rate, the more compression, the more is thrown away. MP3 bit rates can be from as low as 32kbps to as high as 320kbps.

So if some of the data is thrown away when encoding a wave to MP3, how can MP3s sound good? I can't answer that but I can give you an analogy: movies are made up of a series of individual, still pictures but we don't see that...we see a smooth, continuous flow. We see that because we are seeing 24 of those still pictures per second...too fast for us to be able to resolve them as individual pictures. Could sound be done the same way? I see no reason it could not. Is it? I have no idea. Regardless of how the compression is actually accomplished, MP3s do sound good. How good depends upon the bitrate used when encoding them and the encoder used.

When an MP3 is played, the data in it must first be decoded to a wave format; that data is then sent to the soundcard which in turn converts it to analog sound. That means you never actually hear an MP3; think of them sort of like a bunch of shorthand instructions for making waves.

The same thing is true if you want to make an audio CD (one that will play on a standard stereo)...the MP3 must be decoded to wave. That process may be done without your knowledge by the program you use to write the CD but it is being done. Moreover, the wave from which the MP3 was made must have been sampled 44,100Hz, 16 bit stereo. If you have trouble burning audio CDs from some of your MP3s, check them to make sure the wave parameters are correct. If they are not, you can fix them by decoding to a wave file, then resampling that file with Windows Sound Recorder. No need to encode to MP3 again, just burn the wave. Also, simply decoding your MP3s to wave yourself and burning the waves will often fix burning problems...less work for the computer do do.

If you want to make an MP3 CD - one that will only play on a device capable of decoding the MP3 - all you need do is make a "data" CD by selecting that option in the program you are using to burn the CD. Nothing else is required; after all, you are merely transferring data from one place (the hard drive) to another (the CD).