16-Bit or 24-Bit

[nidostar]

Back in January I asked a, perhaps, mischievous question whether we needed FLAC (https://librarymusicthemes.com/index.php?topic=5727.0). The responses were interesting and very helpful but also amusing when the discussion went down a rabbit hole discussing which operating system was better.

I have noticed a number of shares lately in FLAC at 48000Hz/24-bit. While I acknowledge there is a noticeable difference to the human ear when comparing lossless to lossy formats is there a noticeable difference between 48kHz/24-bit and 44.1kHz/16-bit? According to articles I have read on-line the difference is increased dynamic range however, the increase is outside the listening range of the human ear. So are we using the higher setting purely because it is an option in our software or because there is a belief that it results in a superior sound?

[stackjackson]

It's ok to 'hotlink' to internal LMT threads. I fixed it for you

[John_Fred]

i think, to a certain degree, it depends largely on what kind of equipment you play it on.  i.e. what the hardware has to work with, as it whether there's any improvement in noticeable quality.  For most people with standard equipment, (not high-end audiophile stuff), the difference is negligible at best.  But, it's always nice to have just in case you ever win the lottery, or get left a huge sum of cash left to you by your grand-aunt Maude. 

[Psyclon]

16-bit is absolutely fine (again) - just like lossy compression as stated in many ABX tests and the science and research behind it in general
For mixing, editing and such stuff, working with actual, multi-track raw material - a very sophisticated minority cares. And those are either Hollywood productions, production studios and the likes.

There is some talk that the 65,000 different values of 16-bit are not enough, because there might be roundings (up or down) happen, but that is absolutely inaudible. Then there is basically the theorycrafting on inaudible things such as "if you add this and that filter, it can distort all the way down to the source" and all that, but inaudible issues having minor inaudible alterations that are dithered away during finalization is nothing but ridiculous. We all listened to compact cassettes and 8-tracks, to FM radio (and recorded onto compact casettes) and many listened and still listen to vinyls. All of these are inferior to the CD. And now people claim you "need" MORE than CD standard? Noise floor is IIRC -96db, so you gotta pump up the volume so unrealistically high up.

More than 16-bit / more than 48kHz is like a car salesman claiming you need a car that goes 300 kph while the speed limit of the roads is 180 kph anyways. (Same with lossy compression btw, it goes into the same direction but I do not have a problem people using FLAC to avoid filter rules even though barely audible. That still makes sense, even to me However, I do have a "problem" with people that really go so deep down a rabbit hole that invalidates science just for the sake of big numbers.)

CD standard - 16 bits and 44kHz - is all you need. The engineers back in the day chose those for a reason. The inauble (!) differences are dithered away upon finalizing the audio information. As end-user [no remixing et cetera] this matters even less.

[stackjackson]

More than 16-bit / more than 48kHz is like a car salesman claiming you need a car that goes 300 kph while the speed limit of the roads is 180 kph anyways...

This is good

[kpmhill]

I believe it's still a matter of production vs. distribution. 24-bit is useful in an audio production pipeline, because audio gets sliced & diced during editing, and the extra headroom protects against potential processing artifacts.

But for distribution (listening) — essentially snake oil.