Is the dynamic range of CD sufficient?

[amirm]

!. Try this one again. Talk real slow.

How clean should your glass be that you drink from? There is a spectrum from completely sterilized to dirty. Right? The advice people give is to go beyond the hazy line that may make you sick. Same here. When we build a bridge, we don't build to exact capacity but add a safety margin. That safety margin allows us to objectively declare that we can handle the load. Same here. Using analysis of the audibility of the system distortion, we can determine the fidelity line we need to exceed for this assurance. See this article I wrote on this point and #2 below: http://www.madronadigital.com/Library/AudibilityofSmallDistortions.html

2. We don't control it, no, but it's not hard to make a case that it's not music if it's not audible. Why does a spectrum outside of our range of hearing have to be transparent?

I am talking about audible range. Music is great in masking a lot of distortion. But since we can't control what is music, there can be situations where distortion becomes audible. Think of a music that is faintly recorded. Should I be subjected to distortion when I turn it up? If said music is at -30 db, we have a lot of our dynamic range of 16 bits.

4. It's not about economics. We have no good reason to believe there is an advantage in reproducing supersonic content. We do have good reason to believe there is a disadvantage in reproducing it; IMD.

Tim

We do have reason that it has some value. It may not have a ton of value but it has some.

 

[Julf]

When we build a bridge, we don't build to exact capacity but add a safety margin. That safety margin allows us to objectively declare that we can handle the load.

 

[DonH50]

^^^ If I recall my lone civil engineering course, there's a little more to it than that, but that is very funny!

How much margin is required is a source of endless debate, in virtually any field of endeavor.

 

[Phelonious Ponk]

How clean should your glass be that you drink from? There is a spectrum from completely sterilized to dirty. Right? The advice people give is to go beyond the hazy line that may make you sick. Same here. When we build a bridge, we don't build to exact capacity but add a safety margin. That safety margin allows us to objectively declare that we can handle the load. Same here. Using analysis of the audibility of the system distortion, we can determine the fidelity line we need to exceed for this assurance. See this article I wrote on this point and #2 below: http://www.madronadigital.com/Library/AudibilityofSmallDistortions.html


I am talking about audible range. Music is great in masking a lot of distortion. But since we can't control what is music, there can be situations where distortion becomes audible. Think of a music that is faintly recorded. Should I be subjected to distortion when I turn it up? If said music is at -30 db, we have a lot of our dynamic range of 16 bits.


We do have reason that it has some value. It may not have a ton of value but it has some.

Gotcha. How to we weigh that little bit of value against the IMD in the audible range caused by super sonic content of hi-res? Seriously, I doubt that IMD is audible most of the time. But I doubt that the little bit of value is either. You seem to think we have nothing to lose by going to higher resolution. Do you not believe it creates IMD in the audible range, or am I missing your point? are you talking about something else altogether?

Tim

 

[Gregadd]

Indeed performance is rarely linear over the performina
nce range. That makes it often advantageous to set the upper limit beyond the useful range.

 

[microstrip]

Is there any software freely available that can transcode HiRez 24/192 in 44.1/16 and 44.1/24 and is considered transparent, according to the 16 bit supporters? It could be an excellent way of people testing their beliefs.

 

[rbbert]

Is there any software freely available that can transcode HiRez 24/192 in 44.1/16 and 44.1/24 and is considered transparent, according to the 16 bit supporters? It could be an excellent way of people testing their beliefs.

Izotope's mBit+ and SRC are generally considered excellent although not free. Depending on the original source, each successive drop in resolution is usually audible to me, and I may not even be one of the more sensitive listeners here.

 

[rbbert]

My point. We nave nearly 100 years of audiology in which the same result has been repeated over and over again -- feed them a signal above 20khz and humans do not respond at all. It may as well not be there.

Tim

Pure B.S. That's like saying 20/20 vision is perfect and no one can see better than that. When I spent some time in the late '70's helping out doing physical exams on prospective Navy SEALs, most had vision better than 20/20 (some 20/10), and most heard (through headphones, so not even counting non-auditory sensing) well above 20 kHz, one I remember to nearly 24 kHz. I'm sure members of all the Armed Forces elite units have similar abilities.

 

[Phelonious Ponk]

Pure B.S. That's like saying 20/20 vision is perfect and no one can see better than that. When I spent some time in the late '70's helping out doing physical exams on prospective Navy SEALs, most had vision better than 20/20 (some 20/10), and most heard (through headphones, so not even counting non-auditory sensing) well above 20 kHz, one I remember to nearly 24 kHz. I'm sure members of all the Armed Forces elite units have similar abilities.

Slight overstatement might be more accurate than pure BS, 20 to 20k is the average, found over years of testing. Slightly higher frequencies have been recorded for young people. 24khz would be very rare, but you remember what you remember. If you're over 40, though, 20khz is a memory as well. The 22khz limit of CD is more than sufficient for the middle aged audiophiles who swear they can hear "hi res."

Tim

 

[Phelonious Ponk]

Is there any software freely available that can transcode HiRez 24/192 in 44.1/16 and 44.1/24 and is considered transparent, according to the 16 bit supporters? It could be an excellent way of people testing their beliefs.

Isn't this what they did in Meyer and Moran? Convert high res to 44.1/16 and compare it blind to the unconverted files?

Tim

 

[esldude]

And I don't know if you ever looked at thresholds for those few young people who hear to 24 or 25 khz. They are over 90 db. Meaning you can't hear it all until it reaches that level. I believe the very top couple khz for those people the threshold was closer to 100 db than 90 db.

There actually aren't any of the hair-like cells in the ear that respond directly to that. Such activity tops out around 15 -16 khz. But like many natural filters, the stopband isn't sharp so those respond though weakly to higher frequencies.

Further I seem to recall in a couple papers about the subject that out of a few hundred only about 2% or maybe 2.5% could hear at all above 20 khz. I think it was less an a percent that heard 24 khz. And those were just test tones. Could those people hear the difference in something playing music that topped out at 20 khz vs 25 khz. I think for most music the answer would be no. As far as I know there hasn't been formal testing of it however.

 

[Phelonious Ponk]

And I don't know if you ever looked at thresholds for those few young people who hear to 24 or 25 khz. They are over 90 db. Meaning you can't hear it all until it reaches that level. I believe the very top couple khz for those people the threshold was closer to 100 db than 90 db.

There actually aren't any of the hair-like cells in the ear that respond directly to that. Such activity tops out around 15 -16 khz. But like many natural filters, the stopband isn't sharp so those respond though weakly to higher frequencies.

Further I seem to recall in a couple papers about the subject that out of a few hundred only about 2% or maybe 2.5% could hear at all above 20 khz. I think it was less an a percent that heard 24 khz. And those were just test tones. Could those people hear the difference in something playing music that topped out at 20 khz vs 25 khz. I think for most music the answer would be no. As far as I know there hasn't been formal testing of it however.

Hearing above 20khz is about as close to a moot point as we're going to get in a roomful of middle aged men. And the point of media and gear that goes to 40khz and beyond has yet to be made in a way I understand, anyway. Unless you like that IMD. I'll be the first to say I'm not the smartest guy in the room, but I understand this:

http://people.xiph.org/~xiphmont/demo/neil-young.html

Tim

 

[microstrip]

Isn't this what they did in Meyer and Moran? Convert high res to 44.1/16 and compare it blind to the unconverted files?

Tim

No Tim, they did not what I am asking - the only tests carried used an A/D/A loop. But my suggestion is that people do it in their systems, much better than debating for the 1000th time an old controversial experiment whose only current utility is consuming time, net bandwidth and unfortunately storage.

 

[esldude]

No Tim, they did not what I am asking - the only tests carried used an A/D/A loop. But my suggestion is that people do it in their systems, much better than debating for the 1000th time an old controversial experiment whose only current utility is consuming time, net bandwidth and unfortunately storage.

Yes, but their conversions should have degraded things even more than what you are suggesting.

 

[Phelonious Ponk]

Yes, but their conversions should have degraded things even more than what you are suggesting.

Correct. Less transparent. It should have been easier to tell the difference. But, no, micro, it was a sincere question. I don't care to debate Meyer & Moran again. What I really would like to see is someone make the case for the audible benefits of high-res (other than IMD) as clearly as the the case for it's lack of benefits was made in the article I linked here.

Tim

 

[rbbert]

No question that only a small percent of young people can hear much above 20 kHz, but I didn't the issue was only what a bunch of middle-aged men hear, but rather how to accurately record and reproduce music, and there are clearly plenty of people who hear higher. Dog whistles generally are at least 23 kHz, and I'm pretty sure everyone here knew at least one person who could hear a dog whistle when they were younger (and I don't know if dog whistles are as loud as 90 dB; they could be, but I doubt much louder). Assuming 150,000,000 people under 25 in the USA, even 1-2% of that is a very large number.

 

[Phelonious Ponk]

No question that only a small percent of young people can hear much above 20 kHz, but I didn't the issue was only what a bunch of middle-aged men hear, but rather how to accurately record and reproduce music, and there are clearly plenty of people who hear higher. Dog whistles generally are at least 23 kHz, and I'm pretty sure everyone here knew at least one person who could hear a dog whistle when they were younger (and I don't know if dog whistles are as loud as 90 dB; they could be, but I doubt much louder). Assuming 150,000,000 people under 25 in the USA, even 1-2% of that is a very large number.

It's a bunch of middle aged men on this forum (and many others) who are hearing a dramatic difference between hi-res and rebook, that's the point. We know they're not hearing above 22khz. The odds are very good that none of them are hearing above 18khz. So what are they hearing in those hi-res files? I'm not saying there's nothing there; I honestly don't know. What I'm saying is that in all my travels on internet audiophile discussion boards, I've yet to see it explained with nearly as much clarity, detail and credibility as the argument against it. I'd like to get that explanation.

Tim

 

[rbbert]

True, we're discussing two different things. This thread started by asking whether 16/44.1 was adequate for recording music. I guess it's expected it would devolve to the usual level, but I was still trying to look at it from a more scientific perspective. It's not too different (in that respect) from a previous thread which Amir presented, which made a very good (I thought) logical and scientific argument for why we need at least 120 db of dynamic range to accomodate even unamplified music, even though trying to then reproduce that in a home setting might be currently unrealistic and/or harmful to one's hearing. Likewise, there is just too much good evidence that there are a significant number of people for whom reproducing a frequency range of only 20-20 kHz will not replicate live music. Whether or not any of them are members of WBF is really beside the point, or should be. Furthermore, despite some assertions to the contrary, it's clear that digital filtering in both ADC and DAC has measurable anomalies which easily might (or might not, since the more aggressively vocal digital apologists refuse to admit there is a problem and therefore won't carefully investigate it) affect accurate music reproduction.

Try to remember that my (considerable) music collection (at least the part that I listen to) is all digital and probably 90+% CD, so I'm anything but a digital hater.

 

[amirm]

View attachment 13724

A story about a middle eastern country having a very long bridge built in the mountain which was designed I think by a French architect. The day came to test it, he made the French architect stand under it as the cars went by!!! Now that is one way to make sure there is extra safety margin in the bridge.

I cross over the new version of this bridge every week or two and I am always reminded of the demise of the original:

Thank heavens civil engineers finally learned about a thing called resonance .

 

[elcorso]

I believe the premises are poorly raised. It isn't only about human hearing capabilities, but about horrible brickwall filters in the regular CD format. If you truncate so heavy it will alter lower frequencies and music integrity.

I have a DSD DAC (Teac UD-501) with 3 possible filters when playing DSD: 50, 70 or 90 kHZ. By recommendation of one of the DSD authorities, the maker of HQAudioPlayer, I play always under the 90 kHZ filter. I own about 700 DSD albums that I play a lot, and never had a problem with my analogue gear nor with els speakers. Neither I got listening fatigue.

Why a lovely violin (and other instruments) should be reproduced high frequency truncated? This is not what happens in real live concerts.

Roch