96k is better than 44.1k sample rate. why?

[altiris]

Something told me that sample rate was not audible frequency. So I searched for this. I found this article http://www.saecollege.de/reference_m...ders.htm#sampl
If you scroll down to the sampling rate where the sine wave is on this site, it will explain what that 44.1k sample rate actually means. So higher sample rate is higher resolution not higher frequency that the human ear can hear. So that means that the higher you sample at the closer to analog you get. Which means that 96K is better than 44.1k.
In conclusion you should record at 44.1k because “my DAW cant handle too many tracks at 96k” and not because “the human ear cant hear frequencies that high anyway”.

[mshilarious]

Originally Posted by altiris

Something told me that sample rate was not audible frequency. So I searched for this. I found this article http://www.saecollege.de/reference_m...ders.htm#sampl
If you scroll down to the sampling rate where the sine wave is on this site, it will explain what that 44.1k sample rate actually means. So higher sample rate is higher resolution not higher frequency that the human ear can hear. So that means that the higher you sample at the closer to analog you get. Which means that 96K is better than 44.1k.
In conclusion you should record at 44.1k because “my DAW cant handle too many tracks at 96k” and not because “the human ear cant hear frequencies that high anyway”.

But 10kHz will have 4.41 samples taken of each of it's waveforms. Now look at the first waveform we drew. In that drawing we took 6 samples of the waveform and got an amplitude reading saying 0,2,2,0,2,2. imagine how inaccurate 4.41 samples are of a complex waveform. That is why digital high frequencies sound harsh!! The industry has constantly denied this factor and even gone to the extent of saying the hear can't distinguish between a square wave and a sine wave above 7kHz. Pigs Bum.

At a sampling rate of 96kHz you get 9.6 samples of a 10kHz wave and believe me, you can hear it.

In an article by Rupert Neve, I read recently, he said that we should aim for 24bit resolution and 192kHz sampling rate if we want to equal the quality of high quality analogue recording. We will get there. DVD is already up to 24 bit 96kHz sampling so we are on the way. But if your 16bit, 44.1kHz CD sounds bright, consider what makes it bright and you will see that it's a false bright created by the high frequencies sounding like square waves!!

The author of that article lacks a fundamental understanding of digital audio theory. You can't simply reject Nyquist by saying "Pig's Bum".

Let's hear his audio files that "prove" he can hear it. Do the test yourself. Generate a 10kHz sine wave at 96kHz. Convert to 44.1kHz *using a high quality sample rate converter*. Now convert back. Listen to the two samples in a blind test and try to distinguish them. Report back with your results.

[altiris]

This is my research phase. And I will do the actual testing myself. But I’m mostly referring to “highest audible frequency” that I hear people say about the sample rate. I mean why even be capable of recording at higher frequencies if the human ear cant hear above the 20k or so? So to me I don’t believe the frequency it refers to is the “highest audible frequency” when we say sample rate.I’m open to clarification on this.

[bennychico11]

this has been discussed many times over on this site and millions of other forums. I suggest do a search to read everyone's opinion on this topic.

also, you're confused on what the "highest audible frequency" that can be recorded is. It's not the sample rate itself that denotes the highest frequency. In other words, 44.1kHz is not the highest frequency recorded using 44.1kHz sampling rate. The Nyuist theory tells us that HALF of the sampling rate is the highest frequency that can be captured. So it's actually 22.05kHZ.

To add to your research, find out what the highest reproducable frequency is on your typical microphone and/or speaker....and ponder that.

[mshilarious]

Originally Posted by altiris

This is my research phase. And I will do the actual testing myself. But I’m mostly referring to “highest audible frequency” that I hear people say about the sample rate. I mean why even be capable of recording at higher frequencies if the human ear cant hear above the 20k or so? So to me I don’t believe the frequency it refers to is the “highest audible frequency” when we say sample rate.I’m open to clarification on this.

The sample rate needs to be a little more than twice the highest audible frequency. I will be the first to admit 44.1 isn't perfect, because it cuts 20kHz a little too close, resulting in very slight attenuation of frequencies above 18.5kHz. But 48kHz is fine. And most playback systems (or listener's ears) cannot hear the difference anyway.

Where digital audio has run into trouble in the past has had more to do with quality of converters rather than a flaw in the underlying theory. Sadly, statements like those from that site just show that the author doesn't understand the theory.

Yes, it is true that a 10kHz waveform will be more crudely represented as a square wave at 44.1 than 96. But it doesn't matter, because those square waves are distortions that are occuring at frequencies above 1/2 the sample rate, which are then removed by the D/A converter's anti-imaging filter. So the reconstructured waveform is identical to its analog parent (assuming no limitations of hardware, etc.)

Here's an FFT graph of the exact test I described. The lower graph is the 96 to 44.1 to 96. Mind you it has been through TWO sample rate conversions. The top is the 96 kHz original.

So, if there is audible distortion, where it is?

As a note, if I showed you the same analysis at 18.5kHz (10kHz is really not that tough), you'd see exactly what I described: a slight attenuation, but no audible harmonic distortion. Unless you have a better SRC than I do

[snow lizard]

You might want to check out the Sampling Theory PDF at Dan Lavry's website.


sl

[altiris]

Originally Posted by bennychico11

To add to your research, find out what the highest reproducable frequency is on your typical microphone and/or speaker....and ponder that.

I’ve often wondered that. And I guess that’s why you would not hear the difference since the mic is limiting you to 20k on most mics Id even be able to afford. And your monitors and everything else that caps off under 44k.

[altiris]

Originally Posted by snow lizard

You might want to check out the Sampling Theory PDF at Dan Lavry's website.


sl

I will check that out

[pingu]

Originally Posted by mshilarious

The author of that article lacks a fundamental understanding of digital audio theory. You can't simply reject Nyquist by saying "Pig's Bum".

[crowning]

Originally Posted by altiris

Something told me that sample rate was not audible frequency. So I searched for this. I found this article http://www.saecollege.de/reference_m...ders.htm#sampl
If you scroll down to the sampling rate where the sine wave is on this site, it will explain what that 44.1k sample rate actually means. So higher sample rate is higher resolution not higher frequency that the human ear can hear. So that means that the higher you sample at the closer to analog you get. Which means that 96K is better than 44.1k.
In conclusion you should record at 44.1k because “my DAW cant handle too many tracks at 96k” and not because “the human ear cant hear frequencies that high anyway”.

It's quite easy: you won't hear any difference.
If you hear any differences at all it's because the filters/DACs are not as good
in the 44.1 kHz case.
It's much easier/inexpensiv to build filters for 96kHz sample rates.

Tom

[masteringhouse]

Originally Posted by mshilarious

Yes, it is true that a 10kHz waveform will be more crudely represented as a square wave at 44.1 than 96. But it doesn't matter, because those square waves are distortions that are occuring at frequencies above 1/2 the sample rate, which are then removed by the D/A converter's anti-imaging filter. So the reconstructured waveform is identical to its analog parent (assuming no limitations of hardware, etc.)

Well not entirely identical.

Let's take a theoretical worse case. You are sampling at Nyquist (40Khz) a 20Khz sine wave and the samples occurs at a zero crossing (twice). In this scenario you would get no signal at all and is one of the reasons why you should always sample above Nyquist (to get at least a few valid samples occuring during the cycle)

While the frequencies above Nyquist are syphoned off, there is still the possibility of quantization error (error in amplitude) at higher frequencies. Filtering doesn't take care of this, it just removes alias freqs.

I do agree however that his "pigs bum" comment in regards to undersampling and square waves is "horse shit" for the reasons that you have described.

[Teacher]

If I had the processing power, DAC's weren't limited to 48Khz(I mix OTB) and to a lesser extent storage space I'd do 96khz, I don't know if its cuz plug-ins sound better or if 96Khz just sound better then 44.1, but it does... its not a "o shit 96 sounds much better" but things just seem smoother to me when I mix 44.1 ITB and 88.2/96 ITB

[altiris]

Now could all this have anything to do with a problem I came across once. I was recoding some songs from the outs of my keyboard and 44.1k. I sounds just fine. Then I tried at 48.8k I noticed that at 48.8k some parts sounded muted. Not the whole song. like there is a part with a cello duel. At 48.8lk the duet is missing. I would go back to 44.1 and it could be heard again. So some frequencies are affected by sample rate?

[Ford Van]

Originally Posted by bennychico11

To add to your research, find out what the highest reproducable frequency is on your typical microphone and/or speaker....and ponder that.

With or without phase shift? Ponder THAT!

[bennychico11]

Originally Posted by Ford Van

With or without phase shift? Ponder THAT!

with AND without phase shift!

[bennychico11]

Originally Posted by altiris

Now could all this have anything to do with a problem I came across once. I was recoding some songs from the outs of my keyboard and 44.1k. I sounds just fine. Then I tried at 48.8k I noticed that at 48.8k some parts sounded muted. Not the whole song. like there is a part with a cello duel. At 48.8lk the duet is missing. I would go back to 44.1 and it could be heard again. So some frequencies are affected by sample rate?

Not in your situation. Especially not with a synth. The highest note on the synth wouldn't even come close to the highest reproducable frequency at 48kHz sampling rate. The highest note on a typical piano is C8 which is 4,186Hz. No where near the Nyquist Frequency of 24kHz.

[Ford Van]

Originally Posted by bennychico11

with AND without phase shift!

I think if you dig a little deeper, and talk to some "tech" people at some manufactures, you will find that most mics and speaker systems do indeed go beyond 20KHz! The specs above 20k are seldom shown, because usually they are not flat.

If things slow down, and I remember, I will find some stuff for you.

[altiris]

wow theres more to frequencies than I thought.

[bennychico11]

Originally Posted by Ford Van

I think if you dig a little deeper, and talk to some "tech" people at some manufactures, you will find that most mics and speaker systems do indeed go beyond 20KHz! The specs above 20k are seldom shown, because usually they are not flat.

which is one of the things I ponder
although I figured they may have left it off more because they just wanted to show the range of the human ear and how the mic responds within that range.

[Reggie]

This is some of the stuff that I've never been able to quite get my head around. Say you have a source that contains some 20k information. But as with many natural sounds it isn't perfect sine waves. Would a 44.1k sample rate be unable to convey the little imperfections since it is only sampling enough times to basically encode some sort of 20kHz sound, thereby causing the AD or DA convertors to sort of guess what shape the wave really is?
Aaaaand my head exploded.


BTW: I would worry a lot more about using 96K myself if people in the mainstream could listen to 96k CDs in their stereos.

[Robert D]

Originally Posted by Reggie

BTW: I would worry a lot more about using 96K myself if people in the mainstream could listen to 96k CDs in their stereos.

I'm all for it, and that'll be the day I start tracking everything at 96K.

[bennychico11]

Originally Posted by Reggie

This is some of the stuff that I've never been able to quite get my head around. Say you have a source that contains some 20k information. But as with many natural sounds it isn't perfect sine waves. Would a 44.1k sample rate be unable to convey the little imperfections since it is only sampling enough times to basically encode some sort of 20kHz sound, thereby causing the AD or DA convertors to sort of guess what shape the wave really is?
Aaaaand my head exploded.


BTW: I would worry a lot more about using 96K myself if people in the mainstream could listen to 96k CDs in their stereos.

you have to understand that 20kHz is very very high in the frequency spectrum of music. This is mainly just going to be upper harmonics of instruments and not the fundamental note. Again, not many instruments really play much higher than the C8 note on the piano (around 4000Hz). The upper harmonics of a fundamental note is what gives each instrument their own timbre. And there are nuances up there that strengthen the sound of the instrument.
The middle C on piano is where a lot of instruments reside, and this is around 261Hz. This frequency's 76th harmonic is where it nears 20,000. Not a huge deal if we are losing some of that. Now yes, around C8...the 5th harmonic is nearer to 20,000 and the instrument may start to sound different than when we listen to it in the room. But again, human hearing (perfect hearing) is between 20-20kHz...even though most people don't hear that high. This range is one of the reasons 44.1kHz was decided upon...and then we got a safety net of 48kHz to use as well if you so choose (most people do).

look into anti-aliasing to help with your converter question

[guitarfreak12]

But 10kHz will have 4.41 samples taken of each of it's waveforms. Now look at the first waveform we drew. In that drawing we took 6 samples of the waveform and got an amplitude reading saying 0,2,2,0,2,2. imagine how inaccurate 4.41 samples are of a complex waveform. That is why digital high frequencies sound harsh!! The industry has constantly denied this factor and even gone to the extent of saying the hear can't distinguish between a square wave and a sine wave above 7kHz. Pigs Bum.

At a sampling rate of 96kHz you get 9.6 samples of a 10kHz wave and believe me, you can hear it.

In an article by Rupert Neve, I read recently, he said that we should aim for 24bit resolution and 192kHz sampling rate if we want to equal the quality of high quality analogue recording. We will get there. DVD is already up to 24 bit 96kHz sampling so we are on the way. But if your 16bit, 44.1kHz CD sounds bright, consider what makes it bright and you will see that it's a false bright created by the high frequencies sounding like square waves!!

it looks like that guy has no idea what he's talking about, frequency is to sampling rate as apples are to CARS!!

[bennychico11]

Originally Posted by guitarfreak12

it looks like that guy has no idea what he's talking about, frequency is to sampling rate as apples are to CARS!!

well, balancing sample rate and frequency IS important. It helps avoid aliasing.

[guitarfreak12]

true, but I think that guy doesn't really understand what the relationship between them is.