Mp3 vs. CD : Can you tell the difference?

[ed2288]

Can you tell the difference between mp3's and CD's? If you subtract the mp3 frequencies away from the original, what would be left and what would it sound like?

[rojash]

Generally speaking, yes I can tell the difference, and most people could if offered the chance to hear the original and the mp3 back-to-back.
Not sure that the second part of your question makes sense. An mp3 is made by analysing the orginal and compressing it, and in the process losing some of the infromation. This is very similar to converting a bitmap picture to a jpeg, and in this context I'm sure that you'll see that the question is meaningless.

[wideboy]

Most people (?? !!) can't tell the difference when listening on an mp3 player (unless they use a very high 'bit rate'), but play mp3s through a hi-fi and then you'll notice the compression and lack of frequencies. However, I'm quite happy with mp3s just now - I can always listen to the 'originals' if it bothers me (PS. - use 160 or 192 bit-rate minimum).

[ed2288]

Its difficult to explain, but if you "watch" the music on an oscilloscope, or graphic equalizer, they will look ever so slightly different. If you then put them in antiphase and play them at the same time, the mp3 will cancel out with the original and just leave what wasn't compressed in the first place. I was wondering what this sounded like, ie how much detail is actually lost in converting to mp3

[wideboy]

This is the theory taught to our Higher Computing Students (there used to be more detail including stereo to mono conversion):-

To achieve such high rates of compression, a lossy compression scheme is used. This
means that data is removed from the original sound in order to reduce the file size - this data can never be recovered.

So what data does the MP3 method throw away? This is all based on what humans
hear best, called Perceptual Audio Compression.

High or low pitched sounds which humans cannot hear are removed

When two similar sounds occur at the same time in a sound file and one is
significantly louder, then the quiet one is removed

The emphasis is placed on the frequencies of sound that humans hear best