Article: Why We Need Audiophiles
Reply #413 – 2009-04-26 14:56:10
My bottom line is that if I had gone to Atkinson prior to ripping my music and setting up my iPod then I would have gotten this line ( http://www.stereophile.com/features/308mp3cd/ ) "MP3s and their lossy-compressed ilk do not offer sufficient audio quality for serious music listening." *sigh* Time for science again. Mr. Atkinson, this is exactly what I was talking about. From the above URL regarding Figure 2 :However, a picket fence of very-low-level vertical lines can be seen. These represent spurious tones that result, I suspect, from mathematical limitations in the codec. No, they are not limitations in the codec. They are are a limitation of the particular decoder used . The spurious tones are the harmonic distortion you get when letting an MP3 decoder output to 16 bit by simply rounding the floating-point data. They would be completely gone when you decode to 16 bits with dither . LAME developers, maybe you should add a decoder option to dither before outputting to 16 bits so that the audiophile community has one less thing to worry about. IIRC, Winamp and foobar can dither MP3 decodes.Like the skirts that flank the 1kHz tone, these will not be audible. But they do reveal that the codec is working hard even with this most simple of signals. From my audio coding engineer's point of view, very tonal sounds belong to the most difficult sounds to code, especially for MP3. The skirts that flank the 1-kHz tone are due to the QMF in MP3. There is no QMF in AAC, hence there is much less smearing of the tones at low levels.Both MP3 and AAC introduce fairly large changes in the measured spectra, even at the highest rate of 320kbps. Regarding AAC at 320 kbps, I completely disagree with your conclusion. In the range below 17 kHz, the maximum distortion level is -110 dB. That's on the order of the distortion the best D/A-converter+amplifier+loudspeaker configuration can achieve.Given the bigger bit budget at 320kbps, the AAC codec produces a result that may well be indistinguishable from CD for some listeners some of the time with some music. I have never heard of a person being able to repeatedly distinguish a 320-kbps AAC encoded stationary signal from its original, especially if the signal is like the one shown in Figures 4 - 8. If there is, please let me know. He/she will surely be of help in improving the quality of today's audio coders. Chris P.S.: Mr. Atkinson, it's "Fraunhofer", not "Fraunhöfer"