Analog is continuous and digital is discrete and that implies a loss information. Theoretically at least, any digitalization in the signal chain involves a significant, if not perceptible, loss in signal information content.
Personally I'm borderline tone-deaf so I can't tell the difference, but I'm not going to call the people who say they can tell the difference liars.
>Analog is continuous and digital is discrete and that implies a loss information.
Analog loses even more information.
First, something being "continuous" doesn't make it an accurate representation of the original continuous signal, or a better than digital representation. Among many factors, how deep/heavy the head goes into the groove for one affects the fidelity of the signal and how much DR you can have (that's why heavier vinyl is for more expensive/premium releases).
There is other loss of information between the signal (e.g. voice) and its transfer to the vinul, more so than with digital.
Vinyl also has more compression and more equalization thrown in to cut frequencies (heavy bass can make record player heads skip, so it's cut at mastering).
Cassete tape also has more compression (loss of signal dynamic range), more noise (loss of signal), than CDs, plus wow, flutter, etc. It's not even comparable, despite being "analog".
Also, contrary to popular belief, and simplified "layman" posts, digital doesn't produce "square-ized" versions of waveforms due to quantization.
If you digitally sample a pure sine wave, it comes out as a pure sine way when you play it (you can check that on an oscilloscope).
What vinyl does offer better than CD/streaming is lack of convenience (I consider it a pro: it forces more focus and dedication when listening), better tactility, better collectible value, better sentimental value (patina, etc), and extremely better showcase for the cover art.
It is a misconception that digital is lossy but analog is not. In digital recording, the finite bit depth of the sampling process effectively adds a small amount of white noise to the signal (the difference between the sampled value and the true value).
In analog recordings there are also plenty of noise sources, for example the grain size of a magnetic tape or the surface roughness of a vinyl. The difference is that the sampling noise level of a CD is way lower than what is physically possible for any analog medium. This is why we hear "tape hiss" and "vinyl crackle" during quiet parts of the music, but we don't hear "CD hiss". On top of that, nonlinearities anywhere within the analog signal chain introduces distortions that don't exist for a digital signal. Overall, a signal reproduced from a CD is much closer to the true signal than any analog medium could ever achieve.
>but I'm not going to call the people who say they can tell the difference liars.
They're not liars, there are just 2 cases:
(a) They confuse harmonic and other types of distortion (e.g. from tube amps) or the extra compression applied to vynil as better than cleaner signal. Which, subjectively, might be.
(b) They are delluded (as opposed to liars), and wouldn't be able to perform their "tell the difference" in a blind A/B test.
On (a), the distortion arising from electromechanical systems tends to be mainly second harmonic, which people don't seem to mind. It gives a sine wave some "body" or "warmth".
Badly designed DACs tend to produce third harmonic distortion, which sounds "harsh" or "bright" (= like a cafe with lots of hard surfaces and people crashing their cutlery around).
There are very few badly designed DACs being manufactured any more, but the loudness wars have made music sound both bright and "lifeless" (lacking in dynamic range).
So yeah, (b). (B) has been found to be the case even among people who make audio equipment for a living. People hear what they expect or want to hear, just like with vision.
Edit: main reference: Floyd Toole, Sound Reproduction. Toole worked in an acoustics research lab for a Canadian government, and then ran Harmon Kardon's research lab for many years.
> Analog is continuous and digital is discrete and that implies a loss information.
No. Digital audio with 44.1 kHz sampling rate is capable of perfectly reproducing all sound waves in the human hearing range. If any information is lost during the conversion from analog to digital, it is due to the quality of the recording hardware.
Personally I'm borderline tone-deaf so I can't tell the difference, but I'm not going to call the people who say they can tell the difference liars.