[GRUMPYOLDGUY]

I still can't get over that the same people that were excited about 24 and even 32 bits are now so enthusiastic about 16 bit with just 96 dBV of resolution. Many of the same people want Amps with enormous amounts of power that far exceed their headphone's requirements and pine for vast amounts of headroom that can never be achieved. Yet they are thrilled to go to 16 bit R2R DACs.
When I've tried to explain this conundrum, it for the most part, if you can pardon the expression, it falls upon deaf ears.
I wonder how much dBSPL one could realize if they could fully harness 32 bits with Amps and speakers to go to the max? Could one level a city block with such a rig?

Why is 32 bits necessarily louder than 16 bits? Couldn't I keep the reference voltage the same and simply get better precision without increasing the full scale amplitude voltage? It would be like adding 16 extra fractional bits to the DAC word.

 

[cel4145]

I still can't get over that the same people that were excited about 24 and even 32 bits are now so enthusiastic about 16 bit with just 96 dBV of resolution. Many of the same people want Amps with enormous amounts of power that far exceed their headphone's requirements and pine for vast amounts of headroom that can never be achieved. Yet they are thrilled to go to 16 bit R2R DACs.
When I've tried to explain this conundrum, it for the most part, if you can pardon the expression, it falls upon deaf ears.
I wonder how much dBSPL one could realize if they could fully harness 32 bits with Amps and speakers to go to the max? Could one level a city block with such a rig?

The problem is that their epistemology is based on (1) a form of rationalism with the primary premise, "If I can hear better sound, then it must be true that it exists." Then (2) they apply their own form of skepticism based on the premise that science is often wrong, so it must be wrong when it doesn't support their anecdotal experience. So referencing science and engineering can never work to counter their anecdotal experience because it has already been negated by the 2nd premise. It can only work as a valid argument for those that haven't fully embraced that epistemological model because you still have a chance to adjust their frame for understanding audio.

 

[StanD]

Why is 32 bits necessarily louder than 16 bits? Couldn't I keep the reference voltage the same and simply get better precision without increasing the full scale amplitude voltage? It would be like adding 16 extra fractional bits to the DAC word.

Then much of the lower portion of the range will be below the noise floor and serve no purpose. Either end of the imaginary DR has issues, a conundrum.

 

[StanD]

The problem is that their epistemology is based on (1) a form of rationalism with the primary premise, "If I can hear better sound, then it must be true that it exists." Then (2) they apply their own form of skepticism based on the premise that science is often wrong, so it must be wrong when it doesn't support their anecdotal experience. So referencing science and engineering can never work to counter their anecdotal experience because it has already been negated by the 2nd premise. It can only work as a valid argument for those that haven't fully embraced that epistemological model because you still have a chance to adjust their frame for understanding audio.

I know what you mean. When I try to explain that science and engineering has brought them all of the nice goodies we are discussing, that point is either ignored or dismissed by some hand waving. Sometimes, someone gets the point.

 

[ExtremeGamerBR]

The problem is that their epistemology is based on (1) a form of rationalism with the primary premise, "If I can hear better sound, then it must be true that it exists." Then (2) they apply their own form of skepticism based on the premise that science is often wrong, so it must be wrong when it doesn't support their anecdotal experience. So referencing science and engineering can never work to counter their anecdotal experience because it has already been negated by the 2nd premise. It can only work as a valid argument for those that haven't fully embraced that epistemological model because you still have a chance to adjust their frame for understanding audio.

Yes, it really makes sense. 
 
I just want to give a copy of "The demon-haunted world" to everyone on this forum.

 

[cel4145]

Yes, it really makes sense. 

I just want to give a copy of "The demon-haunted world" to everyone on this forum.

Too bad Sagan isn't around so we could have him write a chapter on audiophiles. LOL

 

[StanD]

Too bad Sagan isn't around so we could have him write a chapter on audiophiles. LOL

He might be overwhelmed as there are Billions and Billions of Audio myths.

 

[cel4145]

I know what you mean. When I try to explain that science and engineering has brought them all of the nice goodies we are discussing, that point is either ignored or dismissed by some hand waving. Sometimes, someone gets the point.

You must be finding the rare few that are not totally entrenched in that epistemology. But that's the hard part here on Head-Fi: so many already are probably beyond convincing. It's been demonstrated that people resist major paradigm shifts, and that's the situation here. It would be hard for them to shake those beliefs.

The bigger implication is the sociological/cultural impact of Sagan's book ExtremeGamerBR referenced. People in our society (both audiophiles and not) use the same types of premises to reject accepted science, and logic based on science, to support their personal experiences, their political beliefs, or even just the fact they are feeling stubborn that day about considering something new.

 

[GRUMPYOLDGUY]

Then much of the lower portion of the range will be below the noise floor and serve no purpose. Either end of the imaginary DR has issues, a conundrum.

Looking at Chord Mojo as an example, the noise floor is about -150dBFS... That means one could get about 25 bits of resolution. Shy of 32, but this is a portable device. Chord DAVE noise floor is about -178dBFS, giving about 29 bits of resolution. I think the technology is getting close to allowing 32 bits of resolution. Or am I not understanding the data correctly?

 

[Joe Bloggs]

Then much of the lower portion of the range will be below the noise floor and serve no purpose. Either end of the imaginary DR has issues, a conundrum.

Looking at Chord Mojo as an example, the noise floor is about -150dBFS... That means one could get about 25 bits of resolution. Shy of 32, but this is a portable device. Chord DAVE noise floor is about -178dBFS, giving about 29 bits of resolution. I think the technology is getting close to allowing 32 bits of resolution. Or am I not understanding the data correctly?

Where are you getting this? Chord themselves are advertising no more than 125dB dynamic range for the Mojo and 127dB A-weighted for the Dave (A-weighted being the equivalent of PMPO for wattage :rolleyes: )

 

[GRUMPYOLDGUY]

Where are you getting this? Chord themselves are advertising no more than 125dB dynamic range for the Mojo and 127dB A-weighted for the Dave (A-weighted being the equivalent of PMPO for wattage :rolleyes: )

Some third party measurements, I'll find them again in the morning and post them.

 

[Joe Bloggs]

Where are you getting this? Chord themselves are advertising no more than 125dB dynamic range for the Mojo and 127dB A-weighted for the Dave (A-weighted being the equivalent of PMPO for wattage :rolleyes: )

Some third party measurements, I'll find them again in the morning and post them.

If they are spectral analyses (showing noise floor as a function of frequency) then they have nothing to do with the broadband noise floor we're talking about. Noise floor as a function of frequency depends on the size of the FFT window used; the larger the FFT window, the finer the frequency divisions, and the less energy contained in each division (hence lower dB levels)

 

[GRUMPYOLDGUY]

If they are spectral analyses (showing noise floor as a function of frequency) then they have nothing to do with the broadband noise floor we're talking about. Noise floor as a function of frequency depends on the size of the FFT window used; the larger the FFT window, the finer the frequency divisions, and the less energy contained in each division (hence lower dB levels)

Understood. Disregard my comment then.

 

[gregorio]

  I wonder how much dBSPL one could realize if they could fully harness 32 bits with Amps and speakers to go to the max? Could one level a city block with such a rig?

 
No, one would level the entire city! A full 32bits would in theory have a dynamic range of 192dB. In order to hear the least significant bit, the starting point of this dynamic range would have to be equivalent to or above the noise floor of the listening environment, let's say 50dBSPL, so peak output of the 32bit system would have to be somewhere around 242dBSPL. Off the top of my head, there's only a couple of things on earth capable of creating pressure waves of that magnitude; massive, explosive volcanic eruptions and thermonuclear warheads! Unfortunately, we've already demonstrated that the latter will level an entire city, rather than just one block.
 
G

 

[StanD]

   
No, one would level the entire city! A full 32bits would in theory have a dynamic range of 192dB. In order to hear the least significant bit, the starting point of this dynamic range would have to be equivalent to or above the noise floor of the listening environment, let's say 50dBSPL, so peak output of the 32bit system would have to be somewhere around 242dBSPL. Off the top of my head, there's only a couple of things on earth capable of creating pressure waves of that magnitude; massive, explosive volcanic eruptions and thermonuclear warheads! Unfortunately, we've already demonstrated that the latter will level an entire city, rather than just one block.
 
G

I didn't bother doing the full Monty of math because in the past working it for lesser amounts it was already way over the top and knowing the log nature it doesn't take much imagination to see the outcome. So we can market the system under a product name like, "Volcano Sound" or "Nuke Audio."
 
This is why I brought up this topic, it's another dysfunctional belief in the world of audio. So we can further think about, how much DR is really available in recorded material? What is the ambient noise in a live recording that limits the end product? Even studio recordings use microphones or pickups for instruments. Then there's the use of limiters and compressors. Thus most recordings have baked in limitations, probably for reasons both practical and many times due to the music's genre. So when the crowd went wild for Hirez 24 bit, I scratched my head. When some were proclaiming 32 bits I really shook my head. In the end, how much DR is really available in recorded material?
For those that are into Vinyl, how much can a stylus take and stay in the groove?