[asymcon]

I'm a big fan of 48/20, yes, that's 20bit at 48kHz.
16bit needs to be noise-shaped to actually reach ATH, which reduces dynamic range on high frequencies and makes audio post-processing more difficult and lossy.
24bit on the other hand is overkill and waste of space - no track or sample I have ever heard utilized full 24bit dynamic range.
 
20bit is perfect spot as the original signal is retained in its lossless form without need for noise-shaping and it's spot-on for dynamic range of most of those "hi-res" recordings.
When compressed through FLAC or WavPack 48/20 is usually 33% smaller than 48/24.

 

[JWolf]

Have any of you actually just listened for yourself? That's all that's need to be done. You don't have to read this article or that post. You don't need to know the science behind digital audio. You just need to listen and make your own decision as to what you prefer. HAVE YOU ACTUALLY LISTENED? All you seem to be doing is to try to discredit Hi-Res audio with meaningless articles/blog posts. Just listen.
 
Post yor results of listening. And no ABX please.IT's OK that you know what the source is you are listening to. So just listen. That's all.

 

[Joe Bloggs]

Have any of you actually just listened for yourself? That's all that's need to be done. You don't have to read this article or that post. You don't need to know the science behind digital audio. You just need to listen and make your own decision as to what you prefer. HAVE YOU ACTUALLY LISTENED? All you seem to be doing is to try to discredit Hi-Res audio with meaningless articles/blog posts. Just listen.

Post yor results of listening. And no ABX please.IT's OK that you know what the source is you are listening to. So just listen. That's all.

Why is sighted "testing" ok?

And why is ABX not ok?

Because only results from the flawed methodology of the former conforms with your expectations?

Is it not enough for us to concede that hi-res albums are worth buying in real life (for the mastering quality), even if you may safely downsample them to 16/44.1 for actual usage?

 

[Joe Bloggs]

In other news, any 44.1kHz material (including hi-res resampled to 44.1kHz) sounds miles better than any hi-res material to me on all my music systems (X7 in stock form excepted), because of real-time DSP correction processing for my loudspeakers and headphones that work only at 44.1kHz:

http://www.head-fi.org/t/782131/why-high-res-audio-is-bad-for-music-take-2

(I could set it all up to work at 48kHz as well. 192kHz, not so much. DSD? Forget it.)

 

[JWolf]

Why is sighted "testing" ok?

And why is ABX not ok?

Because only results from the flawed methodology of the former conforms with your expectations?

Is it not enough for us to concede that hi-res albums are worth buying in real life (for the mastering quality), even if you may safely downsample them to 16/44.1 for actual usage?

 
ABX puts too much pressure into the listening. Listening should just be stress free. Even though you can tell which recording is which, you can still make up your own mind.
 
If the Hi-Res sounds better than the CD version, converting to 16/44.1 will also sound better than the CD version. So go for it if that's what you want to do. The thing I don't get though is if we can make CD quality music sound better, why is it not better to start with?

 

[Joe Bloggs]

ABX puts too much pressure into the listening. Listening should just be stress free. Even though you can tell which recording is which, you can still make up your own mind.

If the Hi-Res sounds better than the CD version, converting to 16/44.1 will also sound better than the CD version. So go for it if that's what you want to do. The thing I don't get though is if we can make CD quality music sound better, why is it not better to start with?

There is no rule saying ABX must consist of 15 second clips switched between quickly. Go listen to a whole song, whole album, any way of listening you like. Reason fast switching is usually used is because discriminatory powers are found to be better that way, in contrast to the popular claim that one must settle down with a system for days, have a drink with it, sleep with it, etc. to tell the difference. If that's the way you roll, you can roll that way in ABX as well--just be prepared to find what little discriminatory power you actually had to start with disappear in thin air. Also, a statistically significant number of trials should be performed no matter how long each trial takes.

Reason why "it's not better to start with" is because I've made it sound better *for my system*. It is system-specific processing that must be performed by each listener individually for his / her system, or at best for the same model of headphones.

 

[asymcon]

Hi-res audio could make sense when you prefer specific downsampling algorithms for bit/sample rate reduction for final mixdown.
For example flat vs. noise-shaped 16bit. Music of the Spheres by Mike Oldfield for example has tons of dithering noise and some people might not like it, so they dither the 24bit master using TPDF.

 

[Cerastes]

  IT's OK that you know what the source is you are listening to.

Depends. If you are doing a comparison then it's definitely NOT OK, since we have this thing called confirmation bias.
 
 

   
ABX puts too much pressure into the listening. Listening should just be stress free.

That's really up to the listener and what kind of mentality / personality they have. If they cannot just listen to the music on any equipment  / situation that is not their own choosing, then maybe they shouldn't be talking about differences or stuff like this in the first place due to the inherit bias they have.
 
 

   
If the Hi-Res sounds better than the CD version, converting to 16/44.1 will also sound better than the CD version.

The only reason why it would sound better is that it has been mixed & mastered differently than it's CD counterpart.
 
 

   
The thing I don't get though is if we can make CD quality music sound better, why is it not better to start with?

I'm pretty sure we all are up for better music quality, however upping the settings that are beyond human hearing to begin with doesn't magically do that; that's only snake oil ... we can only achieve better quality music when audio / mastering engineers start doing their job properly.

 

[asymcon]

We can only achieve better quality music when audio / mastering engineers start doing their job properly.

+1
 
That's why as a composer, I mix/master my track on my own.

 

[JWolf]

  I'm pretty sure we all are up for better music quality, however upping the settings that are beyond human hearing to begin with doesn't magically do that; that's only snake oil ... we can only achieve better quality music when audio / mastering engineers start doing their job properly.

 
But, are you sure we are not getting any benefits in the audio range?

 

[RRod]

  Hi-res audio could make sense when you prefer specific downsampling algorithms for bit/sample rate reduction for final mixdown.
For example flat vs. noise-shaped 16bit. Music of the Spheres by Mike Oldfield for example has tons of dithering noise and some people might not like it, so they dither the 24bit master using TPDF.

 
People must listen to music much louder than I do and have much better tone thresholds to account for the torment shaped dither gives them.

 

[asymcon]

@JWolf - Does it matter? If it put smile on your face then that's what matter.
I couldn't tell difference between 48k and 96k...
BUT
I was able to detect sinewave tones in 38kHz range produced by CCFL 1.2kV drivers on multiple occasions. When measured with directional microphone, there were no other harmonics in the audible range, only that pulsating sine going from 36 to 40kHz.
Goes both ways I guess...

 

 

[asymcon]

@RRod - I personally don't mind those high noise levels, but it's possible in certain situations, especially with powerful noise-shaping in quiet portions, the high-frequency noise is disturbing.
Yes, most audiophiles listen close to 85dBA.

 

[Cerastes]

   
But, are you sure we are not getting any benefits in the audio range?

Depends how old you are and what your hearing capabilities are.
 
Cheapest way to test your hearing if it's "High-Res ready" is to buy a dog whistle... or just visit your local audiologist.
 
Personally as a 29yo male I can only hear up to 17 kHz now, so anything beyond CD specifications are wasted on me.

 

[RRod]

  @RRod - I personally don't mind those high noise levels, but it's possible in certain situations, especially with powerful noise-shaping in quiet portions, the high-frequency noise is disturbing.
Yes, most audiophiles listen close to 85dBA.

 
And RMS level of shaped dither should be in the -70dbFS range, so we're at 15dBA. So if we're willing to consider people listening in NR20ish rooms, then ok I guess.