High resolution audio. The science, or lack of...?

[fr0g]

First of all, I must apologize to Quadpatch for essentially wrecking his headphone thread...It wasn't the intention and in hindsight I can see that such topics can escalate...Probably best kept separate...

However, the topic interests me. WHF S&V themselves actively promote, or rather review and claim higher quality for HD music. whether it be 24/96 or higher.

Now, the last I had heard, and as far as I know, the good old CD and the theorum to which music data is recorded onto it, is more than enough for the absolute limit of human hearing.

ie Nyquist-Shannon states (and has been proven to be correct) to perfectly recreate an analogue signal, you need a sampling frequency of double the maximum sound frequency required. Or rather more accurately put (from WIki)

"If a function x(t) contains no frequencies higher than B hertz, it is completely determined by giving its ordinates at a series of points spaced 1/(2B) seconds apart."

Now some people here seem to believe that HD tracks will be better. Why? I have yet to see any links to any papers or proof on this, and in fact have seen some to studies that show the higher resolutions (192 KHz and above) can have an adverse effect on quality...

I do agree that many HD versions of music sound better. But experience tells me that these are simply better recordings...I used to buy HD and found it excellent...until I downsampled one to 16/44.1 myself and could not tell them apart...

As for the 16 bit v 24 bit for playback... Why does anyone think we need 24 bit? 16 bit gives us a 96 dB dynamic range (way higher than vinyl), and if we take an extremely well recorded CD with a DR of 30 dB (very unusual, they tend to be no more than 10-15), then we have a noise floor of -66 dB...which as far as I can see, is inaudible.

So please, without insults or petty put-downs, can someone give me any concrete,scientific information on why they think HD (for playback) is any better?

 

[BigH]

HDTracks are remastered so that maybe the difference. I thought bits was the noise floor, so more bits lower noise so more dynamic range? But then most cds have a DR of less than 20DB anyway.

 

[josh05]

Could it be what this article states "A vinyl record is an analog recording, and CDs and DVDs are digital recordings. Take a look at the graph below. Original sound is analog by definition. A digital recording takes snapshots of the analog signal at a certain rate (for CDs it is 44,100 times per second) and measures each snapshot with a certain accuracy (for CDs it is 16-bit, which means the value must be one of 65,536 possible values).

This means that, by definition, a digital recording is not capturing the complete sound wave. It is approximating it with a series of steps. Some sounds that have very quick transitions, such as a drum beat or a trumpet's tone, will be distorted because they change too quickly for the sample rate." and going to a higher quality gives a better approximation that is audible?

http://electronics.howstuffworks.com/question487.htm

 

[Native_bon]

I see were you coming from.. If most people cannot hear differences in sound then it must be the case.. You could hear differences but could be so liitle that its not worth the extra money paid for HD music. I always say believe your ears not what someone else tells you.

 

[josh05]

I see were you coming from.. If most people cannot hear differences in sound then it must be the case.. You could hear differences but could be so liitle that its not worth the extra money paid for HD music. I always say believe your ears not what some else tells you.

+1 whether its placebo or fact who cares if it sounds better too you.

 

[drummerman]

Dunno but if there is any doubt on how much difference there can be on so called 'HD downloads' ... a read of Paul Miller's measurements on monthly releases in HifiNews&RecordReviews will show that all are certainly not created equal ...

regards

 

[wilro15]

You want SCIENTIFIC information on this forum? :rofl:

Well . . . my tuppence is with my limited exposure to HD audio I could not tell the difference. From what I hear, the quality of the original recording and how it was mastered is more important than HD or not.

 

[fr0g]

Could it be what this article states "A vinyl record is an analog recording, and CDs and DVDs are digital recordings. Take a look at the graph below. Original sound is analog by definition. A digital recording takes snapshots of the analog signal at a certain rate (for CDs it is 44,100 times per second) and measures each snapshot with a certain accuracy (for CDs it is 16-bit, which means the value must be one of 65,536 possible values).

This means that, by definition, a digital recording is not capturing the complete sound wave. It is approximating it with a series of steps. Some sounds that have very quick transitions, such as a drum beat or a trumpet's tone, will be distorted because they change too quickly for the sample rate." and going to a higher quality gives a better approximation that is audible?

http://electronics.howstuffworks.com/question487.htm

The problem with that quote is it is nonsense. The original analogue wave can be "perfectly" reconstructed as long as we sample at 2x the maximum sound frequency. That's the whole point of the theorum, and it has been proven to be correct.

As for people saying "Just go by what you hear". That isn't the point...there are 2 reasons why we shouldn't do that.

1. We can be conned by our own brains.

2. We can be conned by recordings being made purposefully better in the higher resolution (read, more expensive) format.

ie. In the 1st instance we "expect" the HD track to be better...so it is, and in the second, it "is" better, but need not be, or rather, the SD track "could" be just as good.

The maths shows us clearly that a 0-22 KHz analogue signal can be digitised, then made analogue again and will not differ at all from the original for those frequencies.

Yes, the digitised signal is indeed snapshots, but once that digital signal goes back through the ADC, the signal should be identical to what went in.

 

[altruistic.lemon]

Trust your ears, and try these http://www.audiocheck.net/testtones_highdefinitionaudio.php

 

[fr0g]

Trust your ears, and try these http://www.audiocheck.net/testtones_highdefinitionaudio.php

Not sure what the point is. Most equipment won't play back such high frequencies anyway. So at the very best they won't sound like anything. At the worst they could cause distortion as aliasing, as they mention.

Either way, we can't hear them and attempting to recreate them is fairly pointless.

 

[altruistic.lemon]

The point is people can try and hear for themselves. Nothing wrong with that - you afraid the results may not be what you think, mate?

You talk big words about science but when it comes to applying it, you don't want to.

 

[JZC]

Could it be what this article states "A vinyl record is an analog recording, and CDs and DVDs are digital recordings. Take a look at the graph below. Original sound is analog by definition. A digital recording takes snapshots of the analog signal at a certain rate (for CDs it is 44,100 times per second) and measures each snapshot with a certain accuracy (for CDs it is 16-bit, which means the value must be one of 65,536 possible values).

This means that, by definition, a digital recording is not capturing the complete sound wave. It is approximating it with a series of steps. Some sounds that have very quick transitions, such as a drum beat or a trumpet's tone, will be distorted because they change too quickly for the sample rate." and going to a higher quality gives a better approximation that is audible?

http://electronics.howstuffworks.com/question487.htm

Whilst the graph in your link illustrates the point it exaggerates the difference between an analogue and digital waveforms to make it. The so called CD 'approximation' shown is more like a square wave and would not sound anything like the original. A better representation is shown here:

http://www.synaudcon.com/site/blog/digital-audio/digital-audio-at-last/

I think the fundamental question is has current digital sound quality reached a point where it's not possible for most people to detect any further improvement whether it be any increased sample rates or purely analogue recordings.

Even with anaglogue there are limitations in the equipment used to record and reproduce the original sound exactly and may or may not be better than digital equipment.

Also even contemporary vinyl records may contain recordings of sounds that were produced and recorded digitally.

Personally I find CD technology to be more than adequate and the important thing is to enjoy the music rather than trying to listen for imperfections.

 

[abacus]

The differences users hear between HD and CD is purely down to the mastering process, (One of the most important parts of the recording) and this is what you are paying for with HD recording. (The higher bit rate and frequency are of little relevance to playback in a domestic environment so long as the minimum CD level is met)

NOTE: Not all music in HD has been mastered properly (Some manufactures just use it as a selling point) and could be worse than the original CD, so buyer beware.

Bill

 

[fr0g]

The point is people can try and hear for themselves. Nothing wrong with that - you afraid the results may not be what you think, mate?

You talk big words about science but when it comes to applying it, you don't want to.

This post makes no sense at all Lemon.

"Listening" is NOT applying science. As I say, most speakers won't produce much anyway over 22 KHz. So feeding them such a signal won't have any effect, at best, at worst it will cause distortion.

And we can't possibly hear that high anyway.

I am extremely interested and open to other opinions that do indeed include scientific data. I am doing exactly what you accuse me of not doing...applying science. I used to buy HD music. I heard differences. But when I "applied science" and did my own downsample to CD quality, the difference was zero (audibly speaking).

I have no intention of saying the OP is absolute fact. I am open to changing my mind. Just not from meaningless "Trust your ears" types of response.

A lot of work went into the CD standard, and as far as I know (for now), it has yet to be debunked. And on the other thread, that was locked, Mr Dalethorn only threw insults and claims. Claims he did not once back up with anything remotely scientific or concrete.

That is all I am asking for. I'd be thrilled to see evidence to the contrary, but it's yet to arrive.

So for now, in my mind, CD quality is as high as it needs to be...any higher is a waste of money, bandwidth and storage space.

 

[fr0g]

The differences users hear between HD and CD is purely down to the mastering process, (One of the most important parts of the recording) and this is what you are paying for with HD recording

This is my current understanding.

Thanks.

 

[steve_1979]

First of all, I must apologize to Quadpatch for essentially wrecking his headphone thread...It wasn't the intention and in hindsight I can see that such topics can escalate...Probably best kept separate...

However, the topic interests me. WHF S&V themselves actively promote, or rather review and claim higher quality for HD music. whether it be 24/96 or higher.

Now, the last I had heard, and as far as I know, the good old CD and the theorum to which music data is recorded onto it, is more than enough for the absolute limit of human hearing.

ie Nyquist-Shannon states (and has been proven to be correct) to perfectly recreate an analogue signal, you need a sampling frequency of double the maximum sound frequency required. Or rather more accurately put (from WIki)

"If a function x(t) contains no frequencies higher than B hertz, it is completely determined by giving its ordinates at a series of points spaced 1/(2B) seconds apart."

Now some people here seem to believe that HD tracks will be better. Why? I have yet to see any links to any papers or proof on this, and in fact have seen some to studies that show the higher resolutions (192 KHz and above) can have an adverse effect on quality...

I do agree that many HD versions of music sound better. But experience tells me that these are simply better recordings...I used to buy HD and found it excellent...until I downsampled one to 16/44.1 myself and could not tell them apart...

As for the 16 bit v 24 bit for playback... Why does anyone think we need 24 bit? 16 bit gives us a 96 dB dynamic range (way higher than vinyl), and if we take an extremely well recorded CD with a DR of 30 dB (very unusual, they tend to be no more than 10-15), then we have a noise floor of -66 dB...which as far as I can see, is inaudible.

So please, without insults or petty put-downs, can someone give me any concrete,scientific information on why they think HD (for playback) is any better?

You are totally right of course fr0g. But this has been done to death and I fear that you're banging your head against a brick wall trying to explain to someone who doesn't understand Nyquist-Shannon theory why CD quality 16/44 is already past the limit of what humans can can hear and anything higher than that doesn't make any difference to the sound.

:wall:

 

[davedotco]

You are totally right of course fr0g. But this has been done to death and I fear that you're banging your head against a brick wall trying to explain to someone who doesn't understand Nyquist-Shannon theory why CD quality 16/44 is already past the limit of what humans can can hear and anything higher than that doesn't make any difference to the sound.

:wall:

You do not have to understand Shannon-Nyquist.

What you have to understand is that the conclusions of Shannon/Nyquist have been shown to be correct over and over again by scientifically verifiable methods. You need to understand that these conclusions are as close to concrete facts as make no difference.

This, and some comparable subjects, have been 'done to death', yet still people fail to understand.

In an ideal world I would like to know the real details of these people, I could make some money selling their details to anyone I know who has a spare bridge to sell, when we have run out of bridges we can return to that old favourite, 'oil of snake'.

 

[altruistic.lemon]

So the experts from the AVI forum don't want to try a simple test! Amazing, really, since it might prove what you think you know. Talk about closed minds.

Go on, give it a go, won't hurt and might even prove you right.

 

[chebby]

In an ideal world I would like to know the real details of these people, I could make some money selling their details to anyone I know who has a spare bridge to sell, when we have run out of bridges we can return to that old favourite, 'oil of snake'.

They are people who think that scientific discoveries (and the technologies developed from them) are a matter of belief or a matter of opinion rather than a painstaking discipline where every successful finding has to be rigorously tested, published, peer-reviewed and capable of being reproduced.

They are the people who say things like 'science doesn't know everything' (duh!) as if that qualifies them to fill in the gaps with any old 'woo-woo' (like 'directional' AC signal cables and magic rubber feet).

Unfortunately we still live in a society where - by and large - technologists, engineers, scientists, electronics experts etc. don't have the recognition or respect that is paid to them in a country like Germany or South Korea.

Even if they end up successful enough to get a bit of television time (to promote science) they only find themselves on an interview sofa with a witch or a spiritualist or a priest or astrologer whose opinions are all presented as equally valid! (None of whom would tolerate being labelled or addressed publicly as a 'Geek' or a 'Nerd' like many respected scientists regularly are in various media.)

 

[lindsayt]

It's a big fat marketing half truth to state that the 16 bit CD format has a dynamic range of 96ds, because it doesn't have a usable dynamic range of 96dbs.

Distortion in the CD format increases as the recording level decreases. By the time you get to 50, 60, 70 dbs below maximum recording level you start getting increasingly noticeable distortion, especially at higher frequencies.

Try it. Make a recording of a professionally played solo grand piano where the recording level ranges between -95dbs and -65dbs below maximum possible, using the CD format, and at the same time, make a recording onto 2 track 1/4" tape running at 15ips at -30 to -0dbs and then compare them using class A amplification and good speakers to see which sounds more like a real piano.

If CD format really did have a usable dynamic range of 96dbs it would sound at least as realistic as the tape recording.

 

[steve_1979]

So the experts from the AVI forum don't want to try a simple test! Amazing, really, since it might prove what you think you know. Talk about closed minds.

Go on, give it a go, won't hurt and might even prove you right.

I have done high rez vs 16/44 tests before and couldn't hear any difference.

If anyone is interested in trying it for yourselves first of all you need to start by converting the high rez audio files to 16/44 yourself for it to be a fair test. This is to ensure that both versions are from the same mastered version because many of the commercially available high rez audio files have been remastered to deliberately make them sound different to the 16/44 versions.

 

[johngw]

It's a big fat marketing half truth to state that the 16 bit CD format has a dynamic range of 96ds, because it doesn't have a usable dynamic range of 96dbs.

Distortion in the CD format increases as the recording level decreases. By the time you get to 50, 60, 70 dbs below maximum recording level you start getting increasingly noticeable distortion, especially at higher frequencies.

Try it. Make a recording of a professionally played solo grand piano where the recording level ranges between -95dbs and -65dbs below maximum possible, using the CD format, and at the same time, make a recording onto 2 track 1/4" tape running at 15ips at -30 to -0dbs and then compare them using class A amplification and good speakers to see which sounds more like a real piano.

If CD format really did have a usable dynamic range of 96dbs it would sound at least as realistic as the tape recording.

Perhaps you care to explain why you think this may be the case. And no, I don't have the equipment nor the grand piano nor the professional to play it at hand to do the actual experiment as you suggest.

Here's a decent reference by the way for those who genuinely want to understand the engineering background to all this.

 

[davedotco]

In an ideal world I would like to know the real details of these people, I could make some money selling their details to anyone I know who has a spare bridge to sell, when we have run out of bridges we can return to that old favourite, 'oil of snake'.

They are people who think that scientific discoveries (and the technologies developed from them) are a matter of belief or a matter of opinion rather than a painstaking discipline where every successful finding has to be rigorously tested, published, peer-reviewed and capable of being reproduced.

They are the people who say things like 'science doesn't know everything' (duh!) as if that qualifies them to fill in the gaps with any old 'woo-woo' (like 'directional' AC signal cables and magic rubber feet).

Unfortunately we still live in a society where - by and large - technologists, engineers, scientists, electronics experts etc. don't have the recognition or respect that is paid to them in a country like Germany or South Korea.

Even if they end up successful enough to get a bit of television time (to promote science) they only find themselves on an interview sofa with a witch or a spiritualist or a priest or astrologer whose opinions are all presented as equally valid! (None of whom would tolerate being labelled or addressed publicly as a 'Geek' or a 'Nerd' like many respected scientists regularly are in various media.)

It's called 'dumbing down', Chebby, this country has been working very hard at it for several decades.

Hi-fi reproduction is a hugely subjective process, I have a total understanding and respect for people who make their own choices, made on their own subjective evaluations, and buy their system on those criteria.

It is beliving that those criteria are valid and applicable to anyone else that is the problem.

 

[oldric_naubhoff]

It's a big fat marketing half truth to state that the 16 bit CD format has a dynamic range of 96ds, because it doesn't have a usable dynamic range of 96dbs.

Distortion in the CD format increases as the recording level decreases. By the time you get to 50, 60, 70 dbs below maximum recording level you start getting increasingly noticeable distortion, especially at higher frequencies.

Try it. Make a recording of a professionally played solo grand piano where the recording level ranges between -95dbs and -65dbs below maximum possible, using the CD format, and at the same time, make a recording onto 2 track 1/4" tape running at 15ips at -30 to -0dbs and then compare them using class A amplification and good speakers to see which sounds more like a real piano.

If CD format really did have a usable dynamic range of 96dbs it would sound at least as realistic as the tape recording.

this is quite correct. (as always with what comes from Lind). even though 16bit data depth gives you 96dB of dynamic range Red Book format has usable DR ofabout 50-60dB. (IIRC @ -60dB with 16bit material you reach THD level of 0.1% or more, rising rapidly as attenuation increases). still, 50 dB of DR is about right to capture even the most dynamic of live musical events IMO. it's like going from normal, close range conversation spls to ear drum splitting spls. and then you have the option to make use of dithering which buys you a few more bits of DR...

 

[fr0g]

So the experts from the AVI forum don't want to try a simple test! Amazing, really, since it might prove what you think you know. Talk about closed minds.

Go on, give it a go, won't hurt and might even prove you right.

Seriously Lemon, give it a rest.You are starting to look a bit silly. The "test" is useless and will prove nothing no matter what the result is. What frequency does your equipment play back ?

I know, you take a test to see how many litres of coffee you can fit in a spoon...It would be equally usefull.

And if you knew me, you would seriously not call me "an expert from the Avi forum"...to be honest I find that place one of the most close minded, authoritarian places to post. Agree with the consensus and you're "in", otherwise you are a trouble maker. It is a VERY useful place, but not what I would call "home", or in the slightest bit "fun".

But that is to be expected from a manufacturers forum. At least they don'¨t tolerate subjective nonsense about cables,isolation platforms, PSUs and other completely meaningless items that the delusional spend their money on....And that is the reason I use the place...not to mention that my speakers are rather good...