[71 dB]

If redbook is overkill, then what sampling-rate/bit-depth would be the sweet spot?

The overkill offered by redbook doesn't harm much, but 13 bits of dynamic range is enough.

 

[Slaphead]

If redbook is overkill, then what sampling-rate/bit-depth would be the sweet spot?

Redbook is probably the sweet spot as it encompasses slightly more than we can hear, but as @71 dB says 13-14 bits of dynamic range would be more than adequate.

As for the 44.1 KHz sampling rate it's pretty much bang on for human hearing, although some would say that 48Khz allows for a less aggressive AA filter.

 

[bigshot]

Not that there's really all that much to be heard above 14 or 15kHz...

 

[71 dB]

Redbook is probably the sweet spot as it encompasses slightly more than we can hear, but as @71 dB says 13-14 bits of dynamic range would be more than adequate.

As for the 44.1 KHz sampling rate it's pretty much bang on for human hearing, although some would say that 48Khz allows for a less aggressive AA filter.

Yes, little higher sampling rate would give the benefit of easier anti-alias/reconstruction filtering, but it is a very very small benefit.
13 bit / 54 kHz would have almost the same bitrate as 16 bit / 44.1 kHz.

 

[ADUHF]

This topic is badly in need of an update imo. Because alot of people are still thinking in terms of technology that is now close to 30 years old, which has largely fallen into disuse.

While it's probably true that most humans don't have "bat ears" that can hear above the 22.05 kHz Nyquist frequency of 44.1 kHz content, alot of content now uses the 48 kHz video standard for audio instead. Including YouTube clips encoded with the Opus codec (which is what most browsers now use).

I sold off the majority of my CD collection ages ago. So the only time I really think about audio bit depths and sample rates is when using my PC for viewing/listening to online content. And since I don't have the ability to always use the native rates of the content, and am often manipulating the amplitude in the digital domain via EQs and what have you, I usually keep the bit depths and sample rates on my PC at the highest setting my audio devices will allow, which is currently 24-bit / 48 kHz. Because this seems to work best for me.

 

[ADUHF]

Removed by ADUHF.

 

[sander99]

This topic is badly in need of an update imo. Because alot of people are still thinking in terms of technology that is now close to 30 years old, which has largely fallen into disuse.

While it's probably true that most humans don't have "bat ears" that can hear above the 22.05 kHz Nyquist frequency of 44.1 kHz content, alot of content now uses the 48 kHz video standard for audio instead. Including YouTube clips encoded with the Opus codec (which is what most browsers now use).

I sold off the majority of my CD collection ages ago. So the only time I really think about audio bit depths and sample rates is when using my PC for viewing/listening to online content. And since I don't have the ability to always use the native rates of the content, and am often manipulating the amplitude of the content in the digital domain via EQs and what have you, I usually keep the bit depths and sample rates on my PC at the highest setting my audio devices will allow, which is currently 24-bit / 48 kHz.

This topic is about what is enough for distribution of music. Actually this thread is specifically about bit depth and not sampling rate, but never mind that.

I don't think anyone here wants to argue against using 24 bits in your computer/equipment for digital volume control and other digital processing.
48 kHz indeed has its place for practical reasons in relation to video. The 44.1 vs 48 kHz difference is not so important in the discussions about sampling rates, it is more about 44.1 vs 88.2/176.4/... and 48 vs 96/192/.... Again mainly for distribution: I don't think anyone here wants to argue against oversampling in your DAC (or delta sigma converters).

 

[ADUHF]

This topic is about what is enough for distribution of music. Actually this thread is specifically about bit depth and not sampling rate, but never mind that.

I don't think anyone here wants to argue against using 24 bits in your computer/equipment for digital volume control and other digital processing.
48 kHz indeed has its place for practical reasons in relation to video. The 44.1 vs 48 kHz difference is not so important in the discussions about sampling rates, it is more about 44.1 vs 88.2/176.4/... and 48 vs 96/192/.... Again mainly for distribution: I don't think anyone here wants to argue against oversampling in your DAC (or delta sigma converters).

Thank you for the reply, sander99. Fwiw, I wouldn't disagree with any of this. But I fear some of these subtleties may be lost on some of the newbies to this topic, like myself.

And I think alot of folks are now getting most of their music online (rather than via physical media), where a good deal of the content seems to be shifting over to the 48 kHz video standard.

 

[bigshot]

Sanders99, my understanding is that 48 divides more evenly. (48, 24, 12, 6, 3) That is the same reason film runs at 24. (24, 12, 6, 3) It isn't for higher frequencies as much as it is number crunching and dividing for technical reasons. Is that correct?

 

[sander99]

Sanders99, my understanding is that 48 divides more evenly. (48, 24, 12, 6, 3) That is the same reason film runs at 24. (24, 12, 6, 3) It isn't for higher frequencies as much as it is number crunching and dividing for technical reasons. Is that correct?

Actually I don't know for sure, but probably yes. Not much of an answer, I know, but I didn't want to ignore your question.

 

[castleofargh]

https://www.tvtechnology.com/opinions/digital-audio-sample-rates-the-48-khz-question

Once again, audio choices were based on sight instead of sound. ^_^'

 

[Roland P]

Many in the U.S. television industry liked 60 kHz as a standard sample rate because it was free of leap frames and split frequencies, and it synchronized readily with all timing signals used in 60 Hz and 50 Hz television systems, 24 Hz film and the 13.5 MHz component digital video sample rate. The professional audio industry, however, considered it wastefully high, and there was a quantity of 48 kHz software extant in Europe.

60kHz was considered wastefully high .

Quite amazing that this stuff has been sorted out in the late 1970's already.

 

[71 dB]

60kHz was considered wastefully high .

Quite amazing that this stuff has been sorted out in the late 1970's already.

(1) At that time "hi-rez" was extremely demanding if not impossible technologically and also REALLY expensive.
(2) There was no "hi-rez" snake oil bs to confuse people => opinions based on science.

No wonder they had rational views of the sensible sample rates.

 

[bigshot]

Too much is never enough!

 

[Brahmsian]

Sadly, people believe what they want to believe. And many of them cannot trust in science when it shows them something that seems counter-intuitive, or that seems to defy their personal observations. For example, even now, in the 21st Century, there are people who have trouble accepting that the earth is round.

Even scientists can have trouble accepting science and math. My older brother is a pyschiatrist and geneticist, and I've tried to explain to him many times why buying Hi-Res digital files is a waste of money, especially for his 60-year old ears. But he just can't wrap his head around it. I've challenged him many times to do ABX testing of high-res AAC files versus FLACs, 16bit or 24bit, and he won't do it. Why? Because he's afraid that he won't be able to tell the difference, and then he will feel remorse for all the money he has wasted, and he will feel humiliation. He does not want to feel those unpleasant things, so it is easier for his ego to go on buying the high-res files.

I still remember the first time I did an ABX test and discovered I could not tell the difference between 320kb mp3 and FLAC. It was hard to believe. The FLAC's are so much BIGGER! But, I've done probably a hundred ABX tests since then, and only a couple of times could tell FLAC from high bitrate lossy files.

What‘s so sad about it if he’s enjoying his music? I buy pure DSD 256 files. They sound fantastic. It’s just my imagination? Fine. If my imagination is enhancing my enjoyment, it’s enhancing my enjoyment. I don’t bother A/Bing because, first, I couldn‘t care less; I want to listen to music not waste my time comparing files. And second my memory of the sonics vanishes almost instantly so that I probably wouldn’t get it right even if I could hear a difference. Or 320kb mp3 might sound indistinguishable in the moment but be more fatiguing in the long run. Or I might find it fatiguing just because I have a bias against it. At the end of the day, labels like Channel and Reference Recordings are making superb records in DSD and DXD. Whether it’s the DSD/DXD or just better engineering, these are some seriously spectacular classical music recordings and anybody who cares about good sound should support them, hi res or not. RR’s Shostakovich 5th with Honeck conducting is truly reference-class, a contender for THE best recording ever.