Forking from the cables thread to a discussion of whether hi-res is audible.....


While on the subject of embarrassing testing...

Let me preface this by saying I have invested in hi-res audio tracks, both on my server and in my Qobuz subscription. I've always felt I could tell the difference, although there are duds in hi-res just like redbook. And some great redbook recordings.

Going through this test, particularly looking at the control groups, is certainly humbling.  I particularly like the "hardware reviewers" group.

http://archimago.blogspot.com/2014/06/24-bit-vs-16-bit-audio-test-part-ii.html
ahofer
Post removed 
I think so. One possibility I've considered is that the studios that put out hi-res recordings may, on average, be more conscientious about quality.

Also, higher bit-depth has significant advantages in the recording and mixing process.
Speaking only on PCM:

44.1kHz perfectly captures everything <21kHz

16-Bit undithered puts the noise down past -95dBFS, down past around -105dB or better with noise shaped dither (meaning a 24Bit master must exist).

No one whose a teenager or older can hear >19kHz. And no, that paper about hypersonics by Akira’s composer has never been validated. And unless using an MQA DAC with only 1 filter, almost no good DACs have issue with Nyquist filters. Here’s the filter performance of the ~$900 Outlaw RR2160 integrated amp; and here’s the performance of the $100 AudioQuest DragonFly Black DAC.

Keeping in mind that while human hearing does have a range better than ~96dB, you listen in a room, whose noise floor is very high relative to 16Bit, most residential living rooms only allow for about 12Bit.
But the ear’s cilia has now (just recently) been found to ’function’ (detect sound) at a motional level, of less than the width of an atom.

And that is one single aspect of a very complex system.

We can easily hear in a single tone, at an accuracy in timing, of arrival to the ears (each ear)..... of less than 1/100 000th of a second.

This also supposes that there is zero jitter in the model and results.

Where lots of jitter and distortion is involved.

Then add in VERY complex harmonics (which music is) and time to discern time to listen, time to think.... for a resolutional capacity for the human ear, that exceeds that hundred thousandth of a second of discernment ..likely by a factor of 10, even.

So no, it’s not all that simple.

Some aspects of some human created testing gear can be a one trick pony in some single ways, but overall, the human ear is the highest resolution device known to humanity for hearing discernment.