Is anyone aware of a method to compare the fidelity of encoded music files vs. "original" source material (cd, for example)?

I'm just wondering how much is actually lost in the encoding process - I know that bitrate and output type determine the encoded files' quality.

For example, is there a utility that will compare a track from a CD with an MP3 of the same track?

This is just for curiousity's sake, not for debating what type of encoding is best or for any other purpose.

Thanks,

BowRat

just look at the waveform of one against the other in adobe audition

the differences are clear as day

a good portion of freq. range is just plain dumped, which is where the majority of the space is saved

I don't think that is particularly useful.

MP3 and other lossy encoders based on theories on psycho acoustic models attempt to create an encoded copy that will decode to something that is identical or very close to the original signal according to the human ear - not according to a synthetic analysis.

Two signals that look very much alike according to a waveform comparison could sound perceptibly different according to the human ear, while two signals that look different may be impossible to tell apart by the human ear.

The only accurate way of comparing the original signal with the lossy encoded/decoded signal is by using double-blind listening tests on high-end listening equipment.

I wouldn't have said it if i hadn't seen it with my own eyes.

Compression visibly deteriorates the source.

Here's a CD



Here's a 320kbps mp3



Here's a 128kbps mp3




96kbps mp3



They don't call it loss for nothing.

And JPG is a lossy image compression that is VASTLY inferior to PNG (or GIF) where crispness of text and graphics need to be preserved - half a flaming megabyte each... I thought they were slow to load!

Frequency plots are good at showing lowpass and artefacts in the frequency domain, but tell you little more about the actual fidelity.

Direct waveform comparison also tells you next to nothing, as it does not jusge the significance of what has been lost or added.

In general, most lossy compressions are juggling:
1. Lowpass
2. Loss of lower level sounds, hopefully below ATH and masking level
3. Occurence of artefacts, such as pre-echo, distortions

A/B testing is the only reliable way, and with A/B, there is often another parameter that would be taken, an assessment of the quality / irritation factor of the version cleary though to be inferior. Like looking at a load of TVs set up in a store, the pictures are often quite different (an A/B fail), some are clearly poor (noise etc.) but you maight choose one that is actually not the best fidelity to the original.
I use that example, as most people are better at judging picture differences than audio.

The foundation of lossy visual/audio is based on human perceptions and this in itself varies between individual humans. The graphs are no use for showing if a certain human will be able to differentiate between source & lossy copy, just look at the debate about MPEG-2 reduction methods for an illustration of this fact or 128kb MP3 audio compared to higher bitrate MP3 or other lossy audio formats.

Heyman, the whole point of perceptual audio coding is to remove what you can't hear, not what you can't see.

But that said, the graphs you showed even look pretty bad. What MP3 encoder did you use?