dafiend wrote: > > Your claim that somebody with impaired hearing is better able to > identify artifacts in lossy audio is highly unorthodox. > Not at all. There have been scientific studies on this and I've tested it myself with earplugs. I'll see whether I can find any of it, it's a bit hard due to all the noise about mp3 supposedly impairing your hearing on Google. One non-scientific test is here (in German) but I've seen methodically better ones: http://www.heise.de/ct/artikel/Kreuzverhoertest-287592.html > > Common sense suggests just the opposite: the better your hearing, the > more likely to hear artifacts introduced by MP3. > That's only the common sense of people who don't understand how mp3 works. Common sense of people who don't understand a topic is often wrong.
> > Think of it this way: somebody on the side of the road is humming. > There's noise from cars driving by. The cars may mask the humming. But > the better your hearing, the more likely you can still hear the humming > against the car noise. > But it's nothing like that. What you describe is a low SNR which would make it hard to identify a certain sound. But in this case it has nothing to do with SNR, it's complete frequency ranges which completely get removed from the signal because the ear will suppress the information about them (they are still being heard by the ear itself, the information is just not being passed on to the brain). If that mechanism fails because your ear can't hear the frequency that's supposed to mask out the lower frequency you will hear the lower frequencies instead but they are missing in the mp3 file. mp3 works by compressing based on how a "normal ear" perceives sound, if your ears are not "normal" in some way, this works less well. This is also why to a majority of people (with normal or good listening capabilities) mp3s they identify usually sound "better" than the original: this is simply how the codec has been developed. In the lab they tested the responses of people when certain frequencies were removed from the spectrum of music and they picked the algorithms where the result sounded better or equal to the listeners. > > There's one more thing you got backwards. An encoder may start by > low-pass filtering the input signal. This would discard high frequency > content, not low frequency content. When you hearing deteriorates, what > typically happens is that your ability to hear high frequencies is > impaired the most. This again contradicts your argument that people with > poor hearing are more likely to successfully discriminate between lossy > and lossless. Again: this has nothing to do with normal filtering. Read up on how lossy compression works, you've probably got it all wrong. It's not a straightforward signal processing thing, it's actually modifying the signal but in a way that the human perception doesn't recognize. It's easy to recognize with a spectrometer, it's not just some gradual signal degradation or reduction in SNR or something like that, it really completely changes the signal. It's not like the discussion about high-sample-rate audio (where indeed the higher sample rates add nothing to the signal). --- learn more about iPeng, the iPhone and iPad remote for the Squeezebox and Logitech UE Smart Radio as well as iPeng Party, the free Party-App, at penguinlovesmusic.com *New: iPeng 8, the Universal App for iOS 7 and iOS 8* ------------------------------------------------------------------------ pippin's Profile: http://forums.slimdevices.com/member.php?userid=13777 View this thread: http://forums.slimdevices.com/showthread.php?t=102351 _______________________________________________ plugins mailing list [email protected] http://lists.slimdevices.com/mailman/listinfo/plugins
