4 ms·
> They’re actually pretty great, lossless CDs aren't lossless unless the material was recorded at or below 16 bit/44.1kHz. You can argue whether the differenc
by objclxt 5y ago
> They’re actually pretty great, lossless
CDs aren't lossless unless the material was recorded at or below 16 bit/44.1kHz.
You can argue whether the difference is perceptible - some blind studies suggest not - but this fact is what led to a variety of semi-obscure audiophile disc formats that offered higher sampling rates (https://en.wikipedia.org/wiki/Super_Audio_CD https://en.wikipedia.org/wiki/Super_Audio_CD).
- tobyhinloopen 5y agoOk, when people talk about _lossless_ streaming, they usually want their stream to match CD quality. Also, I’m aware SACDs exists. I actually have an amp that can decode DSD encoded audio but I failed to hear a difference.
- dsr_ 5y agoThat's not a definition of lossless that anyone else uses. The definition of lossless is that either no compression algorithm was used or the compression algorithm is capable of precisely recreating the original bits in as bits out. If you feed a lossless mechanism an 8 bit 3.5KHz mono bitstream, it must reproduce the same 8 bit 3.5Khz mono bitstream at the other end. Quality out equals quality in. In your definition, no sound is ever lossless because no microphones can ever capture the complete audio experience available from a live performance - if you move your head a quarter of a wavelength, the sound will change. That's why we don't use that definition.
- numpad0 5y agoOT but since there are tons of professionals and audiophiles here - do anyone notice what can only be called CD artifact? There seems to be something "wrong" with CD that no other formats has, digital or analog.
- nitrogen 5y agoIf the original has a SNR less than 96dB (120dB effectively with shaped dithering), and no frequencies above 20kHz, then CD is lossless.
- threeseed 5y ago30% of people tested could hear 26Khz. 10% could hear 28Khz. https://asa.scitation.org/doi/10.1121/1.2761883 https://asa.scitation.org/doi/10.1121/1.2761883
- marcan_42 5y agoA pure tone. At 110 dB SPL. On one study. Yes, if you blast humans with enough energy in any form they will eventually notice. This is not a relevant result for music.
- threeseed 5y agoBy all means provide an alternate study that shows 20Khz is the limit for human hearing. Because from all accounts it is a myth.
- Dylan16807 5y agoIf they can only hear it way above the threshold of physical damage, that's a good anecdote but it's definitely not relevant to music playback. Sensitivity to audio falls off a cliff as you approach 20KHz.
- simondotau 5y agoAnd that assumes you have young ears. If you're older, chances are it's nowhere near 20kHz.
- marcan_42 5y agoAt normal listening levels, yup. I'm 30 and I can certainly hear the hideous 19 kHz anti-rat blasters they have all over shops in Tokyo (and I think they almost certainly cause ear damage and should be illegal - IIRC they are marketed as putting out 138dB SPL!) but I am rather unlikely to hear the difference between music low-passed at 19kHz and 20kHz.