Terje Mathisen <terje.mathisen@tmsw.no> writes:

> George Neuner wrote:
>
>> On Wed, 4 Jan 2023 13:01:00 -0600, BGB <cr88192@gmail.com> wrote:
>>
>>> On 1/1/2023 4:18 PM, MitchAlsup wrote:
>>>
>>>> On Sunday, January 1, 2023 at 2:35:21 PM UTC-6, BGB wrote:
>>>>
>>>>> On 1/1/2023 11:54 AM, MitchAlsup wrote:
>>>>>
>>>>> During my childhood, we mostly had CDs (*), but by high-school CDs had
>>>>> been mostly abandoned in favor of MP3 files and MP3 players and similar.
>>>>
>>>> So, you abandoned good sound for longer sound. An interesting tradeoff.
>>>
>>> In general, an MP3 at 112 or 128 kbps is basically indistinguishable
>>> from the original in normal conditions.
>>
>> 128 kbps definitely is NOT "CD quality" ... although I will say Vorbis
>> and AAC both are way better than MP3. And VBR typically does much
>> better than CBR.
>>
>> With certain pop/rock I can still hear the difference between CD and
>> MP3 up to 192 kbps. For classical and even some country music, I can
>> hear differences up to 256.
>
> My ears are not particularly good, [...] but I do think that with
> variable bit rate 192 kbps MP3 is effectively "good enough".

I expect that your ears are good enough to hear the problems
with digitized audio, the shortcomings are most likely in
the reproduction equipment, especially speakers. Most audio
gear is not very good, and most speakers are terrible.

(Note: when I say "digitized audio", I mean undersampled digital
audio, which includes basically all the digitized audio that most
people hear.)

> BTW, that same cousin told me that "he could obviously hear the
> difference between any recording format and live music", [...]

I hypothesize that the difference is due to playback format
rather than recording format. In effect he is making an
apples and oranges comparison, and the apples are always
obviously apples. It's hard to do a good experiment in
psychoacoustics.

>> I have yet to hear ANY reductive coding approach the sound quality of
>> vinyl. To accurately reproduce the sound, you need 2..3 overtones and
>> 5..6 harmonics. CDs already clip way too much at the high end, and
>> further reduction just makes the sound more muddy.
>
> This is where my signal processing background steps in and says: Start
> with the 24 bits/sample, 48 KHz master, and compress that with aac/ogg
> down to something a bit less than CD bitrates, maybe ~50% which is
> what you typically get from the lossless CD compression
> formats.(FLAC?)

A 48 kHz sample rate is simply not fast enough. Ears are better
than that, not directly in the frequency domain, but in the phase
and information content domain.

> The result _has_ to be superior to both uncompressed CD and the best
> imaginable, half-speed mastered, vinyl LP.

Sampling at 48 kHz wrecks the signal in terms of what ears can
perceive. Even ordinary vinyl is better.