Hi resolution audio and extended treble response!

[scrubb]

It's almost as if the people who developed the CD 40(?) years ago knew what they were doing when the chose a 16bit/44.1kHz sampling rate. Who would have though? Bits define dynamic range and sampling rate defines frequency bandwidth. The typical "stair-step" representation of digital audio is very misleading. Here's an engineer's description of how digital audio works which finally made me understand it. And yes, for standard listening at home, I don't think there's much value in anything higher than 16/44.1

 

[Geert]

It's almost as if the people who developed the CD 40(?) years ago knew what they were doing when the chose a 16bit/44.1kHz sampling rate.

And they didn't even listen

 

[AdamG]

Where does it show in the Klippel test? Please I am genuinely curious.

I didn’t see it in the compression test at least the non linearity is probably less than 0.1dB at 20kHz.

I also had a look at distortion at 96db SPL. Yes distortion increases but the chart only goes to 10KHz. And probably still at non audible levels considering also MunsonFletcher.

View attachment 261340

Thanks. Appreciate it.

Full results here. Just serving as an example. Pick any other speaker if needed.

KEF Reference 1 META Bookshelf Speaker Review by Erin's Audio Corner

Parting / Random Thoughts (of Erin) As stated in the Foreword, this written review is purposely a cliff’s notes version. No order; just some random notes from my listening. For details about the performance (objectively and subjectively) please watch the YouTube video. But a few notes: All...

www.audiosciencereview.com

And sure. I for sure can not hear 20k anymore, so overall I think it is a non starter.

Hi Harmonic,

Well I’m not entirely certain myself here. But I thought I remembered seeing measurements way out into 40k Hz. At the moment I don’t remember exactly where, but I think I was Amir doing some explaining of how speaker testing works in a video perhaps? Maybe he was demonstrating how High Frequency Roll Off Filters work? It could also be related to how the Frequency Range Axis is set. Most graphs are cut off at 20k Hz. I know the burden to prove my inquiry is on me and I will go digging for the data/graphs.

 

[kemmler3D]

Hi Harmonic,

Well I’m not entirely certain myself here. But I thought I remembered seeing measurements way out into 40k Hz. At the moment I don’t remember exactly where, but I think I was Amir doing some explaining of how speaker testing works in a video perhaps? Maybe he was demonstrating how High Frequency Roll Off Filters work? It could also be related to how the Frequency Range Axis is set. Most graphs are cut off at 20k Hz. I know the burden to prove my inquiry is on me and I will go digging for the data/graphs.

Some DAC and amp measurements go beyond 20khz because 1) the DAC or amp claims a bandwidth up to 50khz or 100khz or something, and so there's a claim to test and 2) Class D amps have very high switching frequencies (many times higher than the audio range) and so they actually output ultrasonic noise to speakers, which is worth checking on in case it's insanely high.

 

[HarmonicTHD]

Hi Harmonic,

Well I’m not entirely certain myself here. But I thought I remembered seeing measurements way out into 40k Hz. At the moment I don’t remember exactly where, but I think I was Amir doing some explaining of how speaker testing works in a video perhaps? Maybe he was demonstrating how High Frequency Roll Off Filters work? It could also be related to how the Frequency Range Axis is set. Most graphs are cut off at 20k Hz. I know the burden to prove my inquiry is on me and I will go digging for the data/graphs.

Nah. That’s ok. I don’t know myself. Was just thinking that the Klippel most likely would also put out some ultrasound from its DAC and yet nothing to see what could be a problem. But again. Probably looking at it from the wrong side or not seeing the full picture (yet).

 

[Blumlein 88]

Some DAC and amp measurements go beyond 20khz because 1) the DAC or amp claims a bandwidth up to 50khz or 100khz or something, and so there's a claim to test and 2) Class D amps have very high switching frequencies (many times higher than the audio range) and so they actually output ultrasonic noise to speakers, which is worth checking on in case it's insanely high.

I think it was mansr who did a careful analysis of switching frequencies and how it might effect speakers. Somewhere here on ASR. Bottom line is, there is no reason for concern in terms of endangering or overloading or otherwise causing issues with tweeters.

 

[kemmler3D]

I think it was mansr who did a careful analysis of switching frequencies and how it might effect speakers. Somewhere here on ASR. Bottom line is, there is no reason for concern in terms of endangering or overloading or otherwise causing issues with tweeters.

Good call out, that's why I said "insanely high" because it'd have to be very unusual to have any effect on tweeters.

 

[Sokel]

How high a typical measurement mic can measure?
Cause I can try,I have sacrificial speakers,an amp that goes up to 500Khz so I suppose I can do a sweep up to 100Khz and see what REW will report.

 

[Blumlein 88]

How high a typical measurement mic can measure?
Cause I can try,I have sacrificial speakers,an amp that goes up to 500Khz so I suppose I can do a sweep up to 100Khz and see what REW will report.

SDC mics will have some response past 20 khz. Maybe to 30-35 khz at reduced level. You need some special mics to get 100 khz. And your speaker won't put it out anyway. And if it did it still wouldn't matter. 100 khz gets absorbed passing thru the air at not inconsequential levels.

 

[AdamG]

I think it was mansr who did a careful analysis of switching frequencies and how it might effect speakers. Somewhere here on ASR. Bottom line is, there is no reason for concern in terms of endangering or overloading or otherwise causing issues with tweeters.

I wish to reiterate in my original post about this (post 32, pasted below for convenience), I wasn’t suggesting that anything being produced above 20k Hz is dangerous or could cause overloading damage to a tweeter in of itself. I was asking if on a small scale, there could be negative effects/consequences. Some speaker manufacturers declare that their speakers can go up to 25k Hz and some even higher. My distillation of this peaked my curiosity about “What do we know about this” and are there possibly negative consequences of very high frequency sound on the drivers. No matter how minuscule. I mentioned this for conversational purposes in the context of the Threads subject (Hi resolution audio and extended treble response!) If we combine this higher resolution Audio that might have music content above 20K Hz being fed to our speakers. Are we possibly (a big stretch I realize) setting up a cascading event that might somehow have negative consequences upon our speakers/amps?

A follow on question. Does it make sense that by sending inaudible content to our speakers, and assuming that our speaker tweeters and our amplification are capable of reproducing this sound (only Bats and Dogs can hear). That we could be doing several unintentional things to our speakers transducers and amplifiers.

1. Causing them to work harder and experience additional wear and tear on content we can not hear or enjoy ?
2. Placing additional audio load on the tweeters and amplifiers that could be reducing the transducers/Amps ability, even if in a minuscule amount, to fully reproduce content we can hear? To reduce its dynamic range and capacity ?
3. Causing the transducer/Amp to work trying to reproduce, again sounds we can not hear, but enough activity as to cause a heat load on the transducer/Amp ? Resulting in increased cumulative heat operational environment ? For no benefit.
4. Is it therefore possible, that playing content that contains sound above human hearing capabilities we are wasting electrical energy, accelerating heat load related degradation, which will eventually leads to early failure mode of the tweeter and/or the Amplifier ?

What other possible negative or unwanted effects could be caused by this inaudible sound ?

 

[Sokel]

SDC mics will have some response past 20 khz. Maybe to 30-35 khz at reduced level. You need some special mics to get 100 khz. And your speaker won't put it out anyway. And if it did it still wouldn't matter. 100 khz gets absorbed passing thru the air at not inconsequential levels.

So what's all the fuss about it?
Pff,I get bored if not applied practically

 

[BeerBear]

Are we possibly (a big stretch I realize) setting up a cascading event that might somehow have negative consequences upon our speakers/amps?

Sure, it's possible. That's why some people in production choose to filter out ultrasonic and subsonic frequencies (mostly the latter, though).
Filtering has its own challenges, but that's besides the point. The reasoning is that if you can't hear some frequencies and they're not contributing anything useful, you might as well remove them and prevent potentially bad stuff. But there's also potentially good stuff there, so it's not always an easy choice.

 

[AbbeyRd1969]

I have heard Sony ES speakers on several occasions. They are equal to any $20,000 speakers out there, period! HiRes is is irrelevant at this price point... I love that certain speaker companies use second rate components and still charge a premium. Quality is quality all the way to the depths of the speaker...

 

[kongwee]

We cannot sample 20 to 20khz more than 44khz?

You can have 192,000 point to draw 20Hz sinewave instead of 44100 point. That what the Sony is talking about on the purple graph.

 

[voodooless]

You can have 192,000 point to draw 20Hz sinewave instead of 44100 point. That what the Sony is talking about on the purple graph.

But we only need 41 points to perfectly recreate it...

correction: we only need 2 points at 41 Hz sampling.

 

[danadam]

The idea more samples per time and extended response meant deeper details for us to hear.

Well, it does mean that, but only up to a point and we reached that point long ago

The resolution bit was one of those bad analogies. Resolution like in pictures like in megapixels. Yet there is not a correct version of that in audio.

Arguably, images also have frequency domain and the more pixels per unit space, the more bandwidth it has.

Understanding Frequency in Images

In the previous blog, we discussed filters and convolution operation. Before moving forward, let’s discuss an important concept “Frequency”, which is widely used in spatial filter…

theailearner.com

What does "frequency" mean in an image?

I don't understand how frequencies are defined in images/photographs. As far as I understand it now, high frequencies are like sharp things in images, like edges or so, and low frequencies are kind...

photo.stackexchange.com

 

[sergeauckland]

It is the first one (But up to a bit below half the sample rate - so around 90Khz for the 192kHz sample rate). But - Hi Res means delivering frequencies higher than the human ear can hear. It is basically pointless.

And in my view, not only pointless but a bit of a fraud. High Quality studio microphones rarely have much response above 18-20kHz, so there's precious little of audio content past there anyway. Use small omnis to get up to 50kHz+, and you'll pick up the whistle from SMPS, line whistle from video screens, small rodents' conversations and weird noises from VAC systems. Not a lot of musical interest, but yes, you'll turn on the 'Hi-Res' light or whatever.

S

 

[kongwee]

But we only need 41 points to perfectly recreate it...

correction: we only need 2 points at 41 Hz sampling.

Sampling rate are points that is 41 point per second. Sampling rate and frequency waveforms are two different thing. You can have 192k sampling rate to capture 20Hz sine wave or any complex waveform at 20Hz. It will not shoot above magically above 20Hz cycle.

And in my view, not only pointless but a bit of a fraud. High Quality studio microphones rarely have much response above 18-20kHz, so there's precious little of audio content past there anyway. Use small omnis to get up to 50kHz+, and you'll pick up the whistle from SMPS, line whistle from video screens, small rodents' conversations and weird noises from VAC systems. Not a lot of musical interest, but yes, you'll turn on the 'Hi-Res' light or whatever.

192k sampling rate is not frequency response. You have miss the point.

 

[voodooless]

192k sampling rate is not frequency response. You have miss the point.

So what is it then?

 

[kongwee]

So what is it then?

Like I have said later sampling rate you are taking 192000 point per second to capture at 20Hz cycle waveform. Of course, you can say capture to 192kHz cycle waveform w/o consider the distortion against a sine wave at 192kHz cycle.