Dan Lavry, design guru new white paper on the Optimal Sampling Rate for Audio!

[mlsstl]

Someone will no doubt pop up saying that 32/384 is the only way. I think that the whole concept of "good enough" upsets audiophiles.

There's a certain psychology that seems innate in a certain percent of the population. It seems their default mode is that if "X" is good, then "2X" must be twice as good, ad infinitum, on up the chain.

Unfortunately, this attitude leaves them susceptible to becoming focused on one small detail to the exclusion of other, larger issues. The sad part is these folks will never be happy.

The sampling rate issue is a great example. Here we have people obsessing about 24/192 and beyond as if it will save the world, when in reality, the real issue is that most recordings (especially pop/rock) fail to use even a small fraction of the resolution available in the ordinary CD format.

But, these people are driven to rearrange deck chairs on the Titanic and nothing is likely to change their thinking.

[michael123]

Good setup will excel with quality recordings.
Not every 192/24 is better than 96/24 but some do.
Maybe something is done to degrade the quality of the later, but it does not matter.. right?

Same Phil Leigh posted the difference between Beatles EMI recordings - plain CD and 24-bit USB version.
When he amplified the difference, Phil could actually distinguish the words of the song!

I think that this pretty much proves the point, isn't it?

Dan Lavry goes with his design as many others stick to theirs (tubes, discrete converters, upsampling, NOS, whatever)
So what?

[mlsstl]

Same Phil Leigh posted the difference between Beatles EMI recordings - plain CD and 24-bit USB version. When he amplified the difference, Phil could actually distinguish the words of the song!

Wonder if anyone's ever done this with the Kingsmen's "Louie Louie"? ;-)

[Wombat]

Good setup will excel with quality recordings.
Same Phil Leigh posted the difference between Beatles EMI recordings - plain CD and 24-bit USB version.
When he amplified the difference, Phil could actually distinguish the words of the song!

I think that this pretty much proves the point, isn't it?

It proves nothing. The cd version was treated different, no matter what you read on the net.

[michael123]

It proves nothing. The cd version was treated different, no matter what you read on the net.

treated differently or not - does not matter, that's the message I am trying to explain.
24-bit version has more meaningful data that its CD version, that's the fact.
Same goes with 192/24 vs 96/24, certain albums on HDTracks do sound different (better or worse is a matter of personal preference)

[darrenyeats]

treated differently or not - does not matter, that's the message I am trying to explain.
24-bit version has more meaningful data that its CD version, that's the fact.
Same goes with 192/24 vs 96/24, certain albums on HDTracks do sound different (better or worse is a matter of personal preference)

Michael, I don't understand what you are saying. I am sure you understand there is a lot that can happen to a recording before it is encoded in 16/44 or 24/96 or whatever? For example, those recent Beatles remasters were slightly compressed for the CD version compared to the late eighties CDs. I am not surprised there was some difference between the new CD versions and some other new version, if that other new version lacks the compression.

This is the type of thing this thread is discussing which makes it difficult to reach solid conclusions about the different resolutions.
Darren

[mlsstl]

treated differently or not - does not matter, that's the message I am trying to explain.
24-bit version has more meaningful data that its CD version, that's the fact.
Same goes with 192/24 vs 96/24, certain albums on HDTracks do sound different (better or worse is a matter of personal preference)

If I make a ballpark guess that I have "about" a hundred of something, writing down "100.000000000" on a piece of paper does not make the original estimate any more accurate.

The problem is, if it gets restated often enough and separated from the original observation, many people may come to believe the number truly has an accuracy of 9 digits to the right of the decimal.

The original Beatles songs were recorded on a 1960s pre-Dolby open reel. The S/N ratio would have been 60 or maybe 65 dB. The frequency range of the tape might have gone to 16K or 17K. Many of the microphones in those days had a similarly limited high frequency bandwidth. To pretend I need a sampling rate that approaches a cut-off frequency of 100KHz with a S/N ratio that is more than double the original recording is simply adding zeros to my number.

It is a very legitimate criticism that virtually no one in the recording industry can leave things alone. When old material is re-released, it is invariably tinkered with to "improve" it. Digital tools make that incredibly easy -- some would argue too easy. Sometimes the effort is modest; they just remove a little tape hiss. But often, the frequency balance is tweaked, volume boosted, dynamic range capped and so on in order to give a more "up-to-date" sound.

The problem is that you can't simply assign the differences between the CD version and the high-rez versions solely to the format difference. It is far more likely that any difference one hears is more due to the choices made in the underlying tweaking done than anything else.

[Wombat]

24-bit version has more meaningful data that its CD version, that's the fact.

For humans and audio there is not anything meaningtfull in the 8bits more. 16bits is more thenm enough. You can´t hear anything of these added 8bit if the 16bits above play music. Masking alone forbids this. So listening 24bit noisefloors can indeed be a different experience as listening 16bit noisefloors. You can´t fool the limits of hearing and audiophiles have the same evolution stand as non-audiophiles. Of course there is more information stored but it doesn´t matter.

[michael123]

For humans and audio there is not anything meaningtfull in the 8bits more. 16bits is more thenm enough. You can´t hear anything of these added 8bit if the 16bits above play music. Masking alone forbids this. So listening 24bit noisefloors can indeed be a different experience as listening 16bit noisefloors. You can´t fool the limits of hearing and audiophiles have the same evolution stand as non-audiophiles. Of course there is more information stored but it doesn´t matter.

Phil is techie.. he just took a difference between the tracks and amplified it to the desired level... until he began hearing speech
It is not a noise, I remember it was something around -60db.. you can search this forum..

You can blame recording engineers, marketing campaigns.
Some say that since 24-bit recordings are targeted for audiophiles, the recording process is if higher quality.

Some guys play here with numbers but refuse (are you afraid? ) to do something practically.
Buy some music, do a blind test!

[Wombat]

Phil is techie.. he just took a difference between the tracks and amplified it to the desired level... until he began hearing speech
It is not a noise, I remember it was something around -60db.. you can search this forum..

Yes, we had exactly the same discussion more then a year back with you as part and i explained how to dither the 24bit version down to 16bit. The result is different to the CD version in every regard. Since you didn´t seem to do these tests for yourself and still spread the same misinformation i wonder what its worth to post in places like that.