3 ms·

However, for consumers, Apple has set the bar for audio quality at compressed AAC, which is arguably discernibly below CD quality. I'd be very happy if Apple we

However, for consumers, Apple has set the bar for audio quality at compressed AAC, which is arguably discernibly below CD quality. I'd be very happy if Apple went to CD quality, which very few people would need to downsample. People with the (perceived?) need for higher resolution audio than CD seem like the sort of people who would probably want to hand-select their own DAC anyway, which should obviate downsampling their higher sample rate tracks.

klodolph 10y ago
You say it's arguably discernibly below CD quality. Can you make that argument? Last time I checked, most audio compression formats could achieve transparency at reasonable bit rates with all or nearly all source material. For example, AAC encoded with iTunes generally achieves near transparency at 128 kbit/s, and complete transparency for by 192 kbit/s. Music you buy from iTunes is 256 kbit/s which, honestly, is overkill. Hydrogen Audio has done some of the more accessible tests, but there are other tests in more controlled conditions which you might prefer.
- peatmoss 10y ago
  I've never seen a robust study, but I'd love to! I've seen a lot of anecdata and quasi-scientific experimentation, but most of those seem to be fatally flawed or have a pretty obvious bias at the offset. Hence, my couched language :-)
  - klodolph 10y ago
    Here: http://wiki.hydrogenaud.io/index.php?title=Hydrogenaudio_Listening_Tests http://wiki.hydrogenaud.io/index.php?title=Hydrogenaudio_Lis... Read the methodology and the statistical analyses, they are reasonably thorough and rigorous (amateur, but definitely not "quasi-scientific"). Here's another: http://soundexpert.org/encoders-192-kbps http://soundexpert.org/encoders-192-kbps And another: http://soundexpert.org/encoders-256-kbps http://soundexpert.org/encoders-256-kbps You can see that the differences at 192 Kbit/s AAC are generally beyond the capabilities of human perception, and at 256 Kbit/s, AAC has a very healthy margin. So the argument that AAC is perceptually inferior to CD quality has little basis according to the evidence.

klodolph 10y ago

You say it's arguably discernibly below CD quality. Can you make that argument? Last time I checked, most audio compression formats could achieve transparency at reasonable bit rates with all or nearly all source material. For example, AAC encoded with iTunes generally achieves near transparency at 128 kbit/s, and complete transparency for by 192 kbit/s. Music you buy from iTunes is 256 kbit/s which, honestly, is overkill. Hydrogen Audio has done some of the more accessible tests, but there are other tests in more controlled conditions which you might prefer.

peatmoss 10y ago
I've never seen a robust study, but I'd love to! I've seen a lot of anecdata and quasi-scientific experimentation, but most of those seem to be fatally flawed or have a pretty obvious bias at the offset. Hence, my couched language :-)
- klodolph 10y ago
  Here: http://wiki.hydrogenaud.io/index.php?title=Hydrogenaudio_Listening_Tests http://wiki.hydrogenaud.io/index.php?title=Hydrogenaudio_Lis... Read the methodology and the statistical analyses, they are reasonably thorough and rigorous (amateur, but definitely not "quasi-scientific"). Here's another: http://soundexpert.org/encoders-192-kbps http://soundexpert.org/encoders-192-kbps And another: http://soundexpert.org/encoders-256-kbps http://soundexpert.org/encoders-256-kbps You can see that the differences at 192 Kbit/s AAC are generally beyond the capabilities of human perception, and at 256 Kbit/s, AAC has a very healthy margin. So the argument that AAC is perceptually inferior to CD quality has little basis according to the evidence.