Understanding Sample Rate

[Spacehound]

14 minutes ago, firedog said:

No, ridicule is just mean and diminishes the one ridiculing. At a certain point, they deserve to be ignored. People here have been incredibly patient with beerandmusic, actually. 

After all the time and personal attention that's been lavished on him by others - especially in this thread - he really should take a step back and try to understand the material without his false preconceptions.

If he isn't willing to do that, the best thing to do would be simply not to engage with him. No one should keep wasting their time on a person if he shows he isn't willing to listen to helpful, honest replies. 

Most people ask a question when  they want to know the answer. He doesn't, he actively rejects the answers he gets.  So it's just trolling.  (Though I don't like the word, as  it's used as a 'catch all' far too often. But I think it's true here, though of course that's just my opinion.)

 

Many here have got far more patience that I have when  dealing with 'wilful ignoranti'  but unlike in the 'open air' there are certain limits on what we can say.

[beerandmusic]

Ok, so i searched youtube for sample rates and found this video:

 

 

It states during the presentation, that not only does a higher sampling rate allow you to capture higher frequency rates, but it also allows you to take more samples to represent our audio (see time 2:50 seconds into the video).  
This is the part that I am concerned with...i don't care about higher frequency rates as a concern.

 

any way just now investigating if the logic I am trying to express is documented.

 

Again, i am NOT concerned with sampling rate to allow higher frequencies....

[beerandmusic]

Here's another question i have.

 

Does anyone here believe that SACD sounds better than CD, or does everyone that professes nyquist theorem applies suggest that nothing higher than 44.1khz is audible?

 

Is there anyone in this thread that believes that SACD is better than CD?

[Spacehound]

21 minutes ago, beerandmusic said:

 ...

Double post.

[Spacehound]

3 minutes ago, Spacehound said:

 Nyquist always applies.

 

On the rest I have not yet made up my mind and probably won't  try as I don't believe it is worth thinking about. To me 44.1 is an adequate minimum but   I'm not a 'minimalist' audiowise - what's a couple of dollars?..

 

[hsmeets]

2 hours ago, beerandmusic said:

 

It states during the presentation, that not only does a higher sampling rate allow you to capture higher frequency rates, but it also allows you to take more samples to represent our audio (see time 2:50 seconds into the video).  
This is the part that I am concerned with...i don't care about higher frequency rates as a concern.

 

any way just now investigating if the logic I am trying to express is documented.

 

Again, i am NOT concerned with sampling rate to allow higher frequencies....

 

By the nature of the signal, if that signal contains "smaller detail's"  these are always higher frequency components not some other information about the signal.

Before the signal is sampled it is send through a low-pass filter and the frequencies above 1/2 the sample frequency are removed

 

Assuming that the 'small detail' was above the 1/2 sample frequency the output after the low pass would be:

If the small details would be below it would have passed the low-pass filter unaltered and would be sampled (ADC) and reconstructed in the DAC.

 

So, yes, such "details" would be lost but you would not hear it as the ear basically is also a lowpass filter.

 

 

 

 

[Blackmorec]

Let me try and take a shot at this.

For audio purposes, frequency is basically the change in pressure P (soundwave), voltage V (analog audio) or word value N(digital audio) over time. A digital word contains no frequency information...only amplitude....how much voltage was present at a particular 'period' in time. I say period because a 'word' has a time period associated with it. Think of a word like a bucket with a tap. At time zero the bucket is empty. It starts to fill at time zero and its level is read and stored at the end of a single 'sampling rate' period, lets say 1/44,100th of a second.  The changing voltage within that 1/44,100th of a second defines how far open the tap is and therefore how full the bucket becomes. Details of any changes to the voltage within that 1/44,100 of a second are lost and only the cumulative voltage is stored.  So, after 1/44,100th of a sec.  you have a digital representation (a word or 'sample') that represents the total voltage that flowed into the bucket. As stated, that word contains NO frequency information, just a number.  Ideally that single word has to be an accurate representation of the voltage, ranging from silence to maximum amplitude. Being numbers, their accuracy depends on, among other things, the incremental difference between each number. To represent that number, standard CD uses a word length of 16 bits, the so-called bit depth.  

For perspective, each sample recorded at 16-bit resolution can contain any one of 65,536 unique values (216). With 24- bit resolution, you get 16,777,216 unique values (224)—a huge difference!

The most important practical effect of bit depth is that it determines the dynamic range of the signal. In theory, 24-bit digital audio has a maximum dynamic range of 144 dB, compared to 96 dB for 16-bit

In order to identify and store frequency we have to store a series or string of words, because frequency is a function of the change of P, V or N over time. So, if a digital word doesn't store frequency and is only a single number, how are we able to store a complex musical signal that comprises huge numbers of different frequencies?  This is where sampling rate plays a role. Let's think of a string of 44,100 words...i.e 1 second's worth of CD music.  In order to represent a frequency of 30 hz, the word values are going to rise and fall in a wave pattern with a frequency of 30Hz. To store a 200hz frequency the word values are going to rise and fall in a wave pattern of 200Hz. The amount by which words rise and fall at any given frequency represents the amplitude. So lets say, we have a loud bass note and a quiet 200 Hz note, the word numbers will rise and fall by large increments to describe the high amplitude 30Hz wave, while at the same time rising and falling by small increments to describe the 200Hz wave.

So, amplitude is represented by the size of the number in each word and frequency by the change in numbers over time. Dynamic range is governed by bit depth (16 or 24 bit) and upper frequency response is defined by sampling rate. Accuracy of the digital signal is defined both by word depth (accuracy of each word) and by sampling rate  (more finely graduated response to changes in the analog signal). A more accurate signal is typically a better sounding signal....high resolution audio isn't about extending frequency response, rather its the improved sound quality that's its important deliverable.

I know a lot of you already completely understand this.

[firedog]

3 hours ago, beerandmusic said:

Ok, so i searched youtube for sample rates and found this video:

 

 

It states during the presentation, that not only does a higher sampling rate allow you to capture higher frequency rates, but it also allows you to take more samples to represent our audio (see time 2:50 seconds into the video).  
This is the part that I am concerned with...i don't care about higher frequency rates as a concern.

 

any way just now investigating if the logic I am trying to express is documented.

 

Again, i am NOT concerned with sampling rate to allow higher frequencies....

This is getting tiresome. People have spent a lot of time trying to explain this to you. They've given you resources to read and to think about.

Instead you keep looking for reasons to double back on your mistaken arguments, and trying to find some link that enables you to keep arguing your baseless position.

 

STOP! You know, you are really coming off as a person with a lot of negative qualities. Do you not have any appreciation that people here are trying to help you, and of how patient they've been with you? Stop throwing their well intentioned efforts back in their faces.

 

Be quiet for at least a  FEW DAYS and study the material.

 

DON'T do more searches - you don't understand what you find anyway. You've already been given all the links you need. You will probably need to go over the material multiple times before you start to get it. 

DON'T post on this topic during that time. You won't ingest the material and understand it in just a few minutes. It's stuff that takes time to wrap your brain around - because it disproves your intuition (like a lot of math does). 
 

After you've truly ingested the material, then come back and ask questions if you have any. 

 

As far as your you tube video, he didn't say what you seem to claim. Of course the higher sampling rate allows you to take more samples - that is by definition true. The sticking point is that after you get to the minimum sample rate necessary, those extra samples provide NO useful information: the waveform drawn from them is exactly the same as the waveform drawn from a dataset taken with a lower sample rate, as long as both meet the minimal rate needed for the recording in question. That is what you need to learn to understand. 

 

Now stop writing about it and sit down and learn about it till you understand it.

 

No one here wants to keep repeating the same things to you over and over again. 

[mansr]

9 hours ago, beerandmusic said:

 

what i meant by gaps is that at t1 you have freq x and t2 you have freq y, and you must connect the dots, so the detail between the dots is the gaps that is estimated, calculated, averaged, or whatever terminology you use....and that is where the details and accuracy are lost....between the samples.

A frequency doesn't exist a single point in time, only over an interval. The longer the interval, the more well-defined the frequency becomes. I this might be the root of your misunderstanding.

[STC]

3 hours ago, beerandmusic said:

Here's another question i have.

 

Does anyone here believe that SACD sounds better than CD, or does everyone that professes nyquist theorem applies suggest that nothing higher than 44.1khz is audible?

 

Is there anyone in this thread that believes that SACD is better than CD?

 

SACD is better than CD. The only problem I can’t reap the benefit.

 

All I need is speakers that are capable of producing the dynamic range of SACD. So I am looking at speakers capable of at least 160 dB 3 meters away taking into consideration of noise floor of 40dB.  

 

I also need an amplifier that could deliver about 3.5 million watt per channel to drive a typical 94dB sensitivity speakers. Any amps you want to recommend?

 

 

 

 

 

[Blackmorec]

2 hours ago, hsmeets said:

By the nature of the signal, if that signal contains "smaller detail's"  these are always higher frequency components not some other information about the signal.

Before the signal is sampled it is send through a low-pass filter and the frequencies above 1/2 the sample frequency are removed

 

Assuming that the 'small detail' was above the 1/2 sample frequency the output after the low pass would be:

If the small details would be below it would have passed the low-pass filter unaltered and would be sampled (ADC) and reconstructed in the DAC.

So, yes, such "details" would be lost but you would not hear it as the ear basically is also a lowpass filter.

 

From which digital signal do you think your DAC would produce the most accurate analog rendition?

 

 

The reason for using higher sampling rates isn't to produce higher frequencies, its to produce a more accurate analog output that better matches the original analog input. In simple terms, your DAC is doing less guesswork

[beerandmusic]

So I just woke up, and haven't researched any more, and i see others have written more for me to read, but as of right now, this is my understanding.

 

Increasing the sample rate does 2 things.  ONE is related to norquist theorem and talks about the frequency range, which I have always stated that I don't have a problem with, because I am ONLY talking about what is in the audible range of hearing, and for the purpose of this argument, i am perfectly fine to just talk about frequencies between 600 and 700 hz.

 

The second thing that increasing the sample rate does, is along the horizotal axis and the accuracy of the resultant waveform by "connecting the dots".  This is what SONY was referring to about accuracy which i believe has nothing to do with the Norquist theorem.

 

So i have only studied for about 30 minutes, and about where i was last time i researched this and quit.

People can suggest I am thick, but I know otherwise, so it really doesn't bother me.  I know my IQ and ability to learn.

 

Logic also tells me if there was no possible truth, so many engineers would not even bother with sample rate higher than 44.1khz.  Clearly it is not a money making opportunity for everyone.

 

What i don't understand is why we can't stop talking about norquist and it's association with the highest frequency range, and lets just talk about the actual sampling on the horizontal axis and how more samples allows for more accuracy in the "connecting of the dots".

 

I will study more, but right now, i actually believe MORE about what i already previously believed.

 

If this is tiresome for anyone, i suggest you just bow out and ignore me and consider me ignorant...trust me, you won't hurt my feelings.  I would much prefer that, than to have to resort to name calling.

[beerandmusic]

7 minutes ago, STC said:

 

SACD is better than CD. The only problem I can’t reap the benefit.

 

All I need is speakers that are capable of producing the dynamic range of SACD. So I am looking at speakers capable of at least 160 dB 3 meters away taking into consideration of noise floor of 40dB.  

 

I also need an amplifier that could deliver about 3.5 million watt per channel to drive a typical 94dB sensitivity speakers. Any amps you want to recommend?

 

Clearly a mocking response, so i will give you a mocking answer....

 

You can't go wrong with McIntosh.

[esldude]

11 minutes ago, Blackmorec said:

 

From which digital signal do you think your DAC would produce the most accurate analog rendition?

 

 

The reason for using higher sampling rates isn't to produce higher frequencies, its to produce a more accurate analog output that better matches the original analog input. In simple terms, your DAC is doing less guesswork

This is definitively wrong. Do not listen to this.

@Blackmorec please don't post this misinformation.

 

Notice even in the illustration the waveform itself is exactly the same for both.  And that is there result you will get with both sample rates. Both will result in equal accuracy in the reconstructed wave.

[pkane2001]

4 minutes ago, Blackmorec said:

 

From which digital signal do you think your DAC would produce the most accurate analog rendition?

 

 

The reason for using higher sampling rates isn't to produce higher frequencies, its to produce a more accurate analog output that better matches the original analog input. In simple terms, your DAC is doing less guesswork

 

I can't believe we are still arguing about a mathematical concept proven nearly a century ago. Just like Beer, you are wrong if you are talking about a periodic signal. Sampling over the Nyquist rate does nothing to improve the accuracy of the reproduction.

 

[beerandmusic]

23 minutes ago, Blackmorec said:

 

From which digital signal do you think your DAC would produce the most accurate analog rendition?

 

 

The reason for using higher sampling rates isn't to produce higher frequencies, its to produce a more accurate analog output that better matches the original analog input. In simple terms, your DAC is doing less guesswork

 

From what i read yesterday, it does both....most of these people are tied to just part of what increasing the sample rate does, while totally disregarding the other thing it does....moving forward, i will only discuss what increasing the sample rate does across the horizontal axis.

[esldude]

14 minutes ago, beerandmusic said:

So I just woke up, and haven't researched any more, and i see others have written more for me to read, but as of right now, this is my understanding.

 

Increasing the sample rate does 2 things.  ONE is related to norquist theorem and talks about the frequency range, which I have always stated that I don't have a problem with, because I am ONLY talking about what is in the audible range of hearing, and for the purpose of this argument, i am perfectly fine to just talk about frequencies between 600 and 700 hz.

 

The second thing increasing the sample rate is along the horizotal axis and the accuracy of the resultant waveform by "connecting the dots".  This is what SONY was referring to and has nothing to do with Norquist in the statement I made previously about accuracy.

 

So i have only studied for about 30 minutes, and about where i was last time i researched this and quit.

People can suggest I am thick, but I know otherwise, so it really doesn't bother me.  I know my IQ and ability to learn.

 

Logic also tells me if there was no possible truth, so many engineers would not even bother with sample rate higher than 44.1khz.  Clearly it is not a money making opportunity for everyone.

 

What i don't understand is why we can't stop talking about norquist and it's association with the highest frequency range, and lets just talk about the actual sampling on the horizontal axis and how more samples allows for more accuracy in the "connecting of the dots".

 

I will study more, but right now, i actually believe MORE about what i already previously believed.

 

If this is tiresome for anyone, i suggest you just bow out and ignore me and consider me ignorant...trust me, you won't hurt my feelings.  I would much prefer that than to have to resort to name calling.

You are completely off base. Have a solid misunderstanding of digital audio.

 

I wouldn't mention it other than you have. If I were judging by this thread, your confidence and opinion of your ability to learn and IQ are much too high.

 

Stop all this. Watch the Digital Show and Tell video. You'll start to get somewhere if you'll understand it.

 

Writing about Norquist when it should be Nyquist is the closest you have come to being correct.  And it's wrong.

[beerandmusic]

 

 

8 minutes ago, esldude said:

\ Watch the Digital Show and Tell video. You'll start to get somewhere if you'll understand it.

 

 

 

provide the link again....does it also talk about the improved accuracy by the higher sample rate?  That is where my focus will be.  My focus will not be on the highest possible frequency range, but on the accuracy provided by the higher sample rate.

 

[opus101]

26 minutes ago, Blackmorec said:

From which digital signal do you think your DAC would produce the most accurate analog rendition?

 

Your pic isn't showing digital signals there, digital isn't defined anywhere but at the sampling instants.

[jabbr]

15 minutes ago, beerandmusic said:

 

From what i read yesterday, it does both....most of these people are tied to just part of what increasing the sample rate does, while totally disregarding the other thing it does....

Look you are trying to be understand this by “thinking about it” without understanding the underlying math. To me this all sounds like you are trying and trying to argue that 1+1=3 ??‍♂️

[beerandmusic]

3 minutes ago, jabbr said:

Look you are trying to be understand this by “thinking about it” without understanding the underlying math. To me this all sounds like you are trying and trying to argue that 1+1=3 ??‍♂️

 

I do NOT want to talk about the highest possible frequency range, i want to talk about the accuracy of transitions.

[jabbr]

4 minutes ago, beerandmusic said:

 

 

 

provide the link again....does it also talk about the improved accuracy by the higher sample rate?  That is where my focus will be.

 

Increased accuracy by higher sampling rate is not present in PCM encoding the way you are imagining. Until you have a solid understanding of the basics, topics like multibit SDM are going to be hopeless.

[esldude]

3 minutes ago, beerandmusic said:

 

 

 

provide the link again....does it also talk about the improved accuracy by the higher sample rate?  That is where my focus will be.  My focus will not be on the highest possible frequency range, but on the accuracy provided by the higher sample rate.

 

It provides the clearest, simplest, easiest to understand example of why higher sample rates only provide for higher frequencies and not improved accuracy.  Continuing to hang onto this incorrect idea will be stumbling block to ever understanding digital audio.

 

The link is in the first page of posts to this thread.

[Don Hills]

3 minutes ago, beerandmusic said:

provide the link again....does it also talk about the improved accuracy by the higher sample rate?  That is where my focus will be.

 

It actually shows just the opposite, and it does it with real hardware.

Increasing the sample rate does not capture an in-bandwidth signal more accurately. To take your "600 Hz to 700 Hz" tones, they can be accurately digitised with a 44.1 kHz sample rate. Digitising at, say, 96 kHz will not capture them any more accurately. "But surely," you say, "you are sampling at twice the rate so it must be more accurate."  Simple, intuitive, and wrong. When you understand how you are wrong, we can continue.

 

Watch this.

 

[beerandmusic]

Just now, jabbr said:

Increased accuracy by higher sampling rate is not present in PCM encoding the way you are imagining. Until you have a solid understanding of the basics, topics like multibit SDM are going to be hopeless.

 

ok, well i am willing to put a couple hours into it...and see if my understanding is any better....and if not, people that suggeset SACD is not superior to CD, can just consider me ignorant even though my ears and my logic tell me differently.