Old 04-20-2017, 07:24 AM   #1
Tubeguy
Human being with feelings
 
Join Date: Sep 2016
Posts: 28
Default Frequency response over 20khz?

When I record with any daw and various sound cards, all I get get is frequency response to 20khz (usually around 18khz) testing with Voxengo Span. But when I load Wav from CD recorded in a proper studio, it usually goes to well over 20khz, maybe 25khz+. I can't measure that far but can see in window it goes way beyond what Span can measure.
So now I'm a bit confused about that since I've never payed attention to it. What does it mean?
Tubeguy is offline   Reply With Quote
Old 04-20-2017, 08:01 AM   #2
DVDdoug
Human being with feelings
 
Join Date: Jul 2010
Location: Silicon Valley, CA
Posts: 1,412
Default

What are you recording? ...Cymbals usually have high-harmonics, but vocals don't, etc.

I assume you are recording at 44.1kHz or higher, and I assume you are not using lossy compression (MP3, etc.)?

What kind of a microphone are you using? A good quality studio condenser will have more high-frequency output than a dynamic or ribbon.

Quote:
But when I load Wav from CD recorded in a proper studio, it usually goes to well over 20khz, maybe 25khz+.
CDs have a sample rate of 44.1kHz which means the audio can't go over 22.05kHz.* If you're seeing 25kHz that's an artifact of your spectrum analyzer.

FYI - Assuming you are young with good hearing and you can hear up to 20kHz, the highest frequencies in music are at a low-level compared to mid & bass sounds and they are masked (drowned out). So, you can generally filter-out everything above about 16kHz and you won't hear any difference.



* That's the Nyquist Theory, but an easy way to think about it is that you need at least one sample to represent the top-half of the wave and one sample for the bottom half... Two samples per cycle.

Last edited by DVDdoug; 04-20-2017 at 08:08 AM.
DVDdoug is offline   Reply With Quote
Old 04-20-2017, 08:09 AM   #3
serr
Human being with feelings
 
Join Date: Sep 2010
Posts: 6,004
Default

My first question would be what are you trying to record with frequency content above 20kHz?

Next, something is wrong with the metering you are doing because the CD format only supports 44.1kHz sample rate audio which has a max top frequency of 22.05kHz. 25kHz is not possible to resolve with a 44.1k sample rate.

Can I assume you are recording at 96k normally?

FYI on HD sample rates: We are still only working with the audible range of sound. That frequencies above the range of hearing are captured is an artifact of the system. The intention is to have a wide margin between the sampling frequency and the top of the audio band because converters work better (less distortion bleeding into the audio band) with a wider margin.

So there's that. (And we can ignore any strawman arguments against HD that try to claim it was intended to record audio above the range of hearing.)

Anyway, what content are you trying to record that is not being captured?
I would look to the microphone being used first.

Is this for music audio or something else?
__________________
Mac Pro 8x3.33GHz i7, 16GB, 256GB SSD(OS, apps), 3x2TB 7200 HD(data); MacBook Pro 2.8GHz, 6GB, 128GB SSD HD(OS, apps), 750GB 7200 HD(data); 2xTrue Precision 8; Apogee AD-16; 2xMOTU 828mk3, Evolution UC-33e; Faderport; WiRanger, iPad & the analog mixer has retired
serr is online now   Reply With Quote
Old 04-20-2017, 12:44 PM   #4
somebodyelseuk
Human being with feelings
 
somebodyelseuk's Avatar
 
Join Date: Jun 2011
Location: Birmingham, UK
Posts: 811
Default

Putting aside all the science, what is the equipment you are using capable of capturing?
If the frequency response of your mic/pre-amp/interface isn't up there, then you ain't capturing it...
...same applies at the other end - if your speakers/amp/interface ain't up there, you ain't hearing it.
__________________
"As long as I stay between the sun & my shadow, I guess I'm doing well."
somebodyelseuk is offline   Reply With Quote
Old 04-20-2017, 01:21 PM   #5
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 18,245
Default

Quote:
Originally Posted by somebodyelseuk View Post
If the frequency response of your mic/pre-amp/interface isn't up there, then you ain't capturing it...
Setting applicability and context aside, lots of analog gear goes anywhere from 10Hz to 150kHz - that doesn't account for mics etc. though which don't. Just the general point there is a bit of gear/circuitry that go pretty dang high regardless of usefulness. My Neve 511 is 10Hz to 31.5kHz for example (again, who cares about that last 10k).

Quote:
...same applies at the other end - if your speakers/amp/interface ain't up there, you ain't hearing it.
They, ain't hearing it anyway.
__________________
Your whole life people will tell you what you can't do. Getting past them is the first step to actually getting things done.
karbomusic is online now   Reply With Quote
Old 04-20-2017, 02:12 PM   #6
DVDdoug
Human being with feelings
 
Join Date: Jul 2010
Location: Silicon Valley, CA
Posts: 1,412
Default

P.S.
As long as the sample rate is high enough, and as long as you're not doing it with digital processing, we know it's NOT digital and it's NOT the software. It's something on the analog side (the sound source, acoustics, or electronics). It's possible that it's the analog side of a soundcard, but unlikely.
DVDdoug is offline   Reply With Quote
Old 04-20-2017, 02:35 PM   #7
bezusheist
Human being with feelings
 
bezusheist's Avatar
 
Join Date: Nov 2010
Location: Dummytown
Posts: 272
Default

Quote:
Originally Posted by Tubeguy View Post
I can't measure that far but can see in window it goes way beyond what Span can measure.
Span can show up to 96kHz. check the settings.


i would suspect your issue has to do with re-sampling, whether it's the converter's filters or the DAW, don't know. you need to do a few "tests" to find out.

Last edited by bezusheist; 04-20-2017 at 02:40 PM.
bezusheist is online now   Reply With Quote
Old 04-20-2017, 10:47 PM   #8
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Germany
Posts: 2,862
Default

Quote:
Originally Posted by Tubeguy View Post
But when I load Wav from CD recorded in a proper studio, it usually goes to well over 20khz, maybe 25khz+.
Audio from a CD that is higher than 20 KHz only can be an error.

Technically CD sampling rate is fixed to 44100 Samples per second an only can produce correct audio up to 22050 Hz. You do need an anti-aliasing filter that is located well below 22050 Hz to bring down any erroneous signal (above 22050 Hz) close to zero. as the filter can't do too sharp cut without audible side-effects, it will start to reduce all signs above 200000 Hz.

So, again, with a CD, everything you see above 20000 is Junk that might or might not reduce the audio quality.

-Michael
__________________
www.boa-sorte.de
mschnell is online now   Reply With Quote
Old 04-20-2017, 11:14 PM   #9
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by mschnell View Post
Audio from a CD that is higher than 20 KHz only can be an error.
I could synthesize a non-erroneous 21 KHz tone and burn it to CD.

Quote:
Originally Posted by mschnell View Post
... the filter can't do too sharp cut without audible side-effects....
?
TryingToMakeMusic is offline   Reply With Quote
Old 04-20-2017, 11:33 PM   #10
EvilDragon
Human being with feelings
 
EvilDragon's Avatar
 
Join Date: Jun 2009
Location: Croatia
Posts: 18,567
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
I could synthesize a non-erroneous 21 KHz tone and burn it to CD.
And you would get aliasing, unless it gets resampled with a good resampler to 44.1k before burning, which would then bandlimit that to get rid of the aliasing.


Quote:
Originally Posted by TryingToMakeMusic View Post
?
Filters with too steep transition bands can have ripples and other properties that would negatively affect the original signal.

Last edited by EvilDragon; 04-20-2017 at 11:38 PM.
EvilDragon is online now   Reply With Quote
Old 04-20-2017, 11:33 PM   #11
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by serr View Post
We are still only working with the audible range of sound.
The audible range of sound is what we hear, not what we work this. Those are different things.

Quote:
Originally Posted by DVDdoug View Post
Assuming you are young with good hearing and you can hear up to 20kHz, the highest frequencies in music are at a low-level compared to mid & bass sounds and they are masked (drowned out). So, you can generally filter-out everything above about 16kHz and you won't hear any difference.
no... record a 28,000 Hz tone + a 28,400 Hz tone (using sample-rate over 44 obviously).

A) filter everything above 16 kHz, and then add a distortion FX. You get silence.

B) start over. This time don't filter everything over 16 kHz, but add the same distortion FX. Now you have a 400 Hz tone which is perfectly audible.

Inaudible can become audible during processing. Processing is common in real-life, while often ignored by theoreticians.

Quote:
Originally Posted by karbomusic View Post
My Neve 511 is 10Hz to 31.5kHz for example (again, who cares about that last 10k). They, ain't hearing it anyway.
no... record a 28,000 Hz tone + a 28,400 Hz tone (using sample-rate over 44 obviously).

A) filter everything above 16 kHz, and then add a distortion FX. You get silence.

B) start over. This time don't filter everything over 16 kHz, but add the same distortion FX. Now you have a 400 Hz tone which is perfectly audible.

Inaudible can become audible during processing. Processing is common in real-life, while often ignored by theoreticians.
TryingToMakeMusic is offline   Reply With Quote
Old 04-20-2017, 11:38 PM   #12
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by EvilDragon View Post
Quote:
Originally Posted by TryingToMakeMusic View Post
I could synthesize a non-erroneous 21 KHz tone and burn it to CD.
And you would get aliasing.
I don't see why I'd necessarily get aliasing by synthesizing a 21 kHz tone.

Quote:
Originally Posted by EvilDragon View Post
Filters with too steep transition bands can have ripples and other properties that would negatively affect the original signal.
That's true of any filter, but it doesn't make them all completely useless.
TryingToMakeMusic is offline   Reply With Quote
Old 04-20-2017, 11:39 PM   #13
EvilDragon
Human being with feelings
 
EvilDragon's Avatar
 
Join Date: Jun 2009
Location: Croatia
Posts: 18,567
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
That's true of any filter, but it doesn't make them all completely useless.
A good DAC filter needs to have no rippling, so any filter with ripples used in a DAC is as useless as it gets.

Last edited by EvilDragon; 04-21-2017 at 06:23 AM.
EvilDragon is online now   Reply With Quote
Old 04-20-2017, 11:41 PM   #14
bezusheist
Human being with feelings
 
bezusheist's Avatar
 
Join Date: Nov 2010
Location: Dummytown
Posts: 272
Default

I got a headache from the misinformation here...
bezusheist is online now   Reply With Quote
Old 04-21-2017, 05:53 AM   #15
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Germany
Posts: 2,862
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
I could synthesize a non-erroneous 21 KHz tone and burn it to CD.
Only by some kind of Error. This is forbidden by Physics/Mathematics.

There are tons of videos in the Internet about Sampling, the Nyquist Theorem, and Aliasing, explaining that stuff in detail.

-Michael
__________________
www.boa-sorte.de
mschnell is online now   Reply With Quote
Old 04-21-2017, 05:58 AM   #16
bezusheist
Human being with feelings
 
bezusheist's Avatar
 
Join Date: Nov 2010
Location: Dummytown
Posts: 272
Default

Nyquist for CD audio (44.1 kHz) is 22,050 Hz.
the "limitation" to anything less would be at the ADC or DAC.
so yes, 21 kHz can be recorded and produced to/from CD.
It is capable of 0 Hz to 22.05 kHz. Hence the term "lossless".
__________________
the quim reaper

Last edited by bezusheist; 04-21-2017 at 06:08 AM.
bezusheist is online now   Reply With Quote
Old 04-21-2017, 06:05 AM   #17
Tubeguy
Human being with feelings
 
Join Date: Sep 2016
Posts: 28
Default

Oh...I never expected to get so many replies (thanks). I must say I've been recording for many years so I have some idea of digital limitations. I've actually started on analog systems than moved to DAW and just used 44.1k ever since. Anything higher and I can't hear any difference including noise floor so I don't bother.
But in the past year I've moved back to mostly analog gear,(tape,tube,etc)
and only using DAW for the recording. I just noticed the frequency difference by accident the other day while messing around. But it's not the analog gear doing that because I get the same frequency response with my old digital recordings too.
I use mostly SM57 which maxes at 15k so I just thought that was it but than again it is very common mic in big studios so maybe it's not that.
I think best I can do is to try the 96k Span, mine is the old one and post couple of pics of what I see. it's just a bit of a mystery to me I'd like to figure out.
Tubeguy is offline   Reply With Quote
Old 04-21-2017, 06:38 AM   #18
Tubeguy
Human being with feelings
 
Join Date: Sep 2016
Posts: 28
Default

Problems solved. I'm an idiot :-) I never noticed that my Span actually goes to 96K, I just had to slide it further. Shows how much I care about digital guess.
Still one question remains, maybe someone has an answer.
The pic-1 is of my own recording.
http://i.xomf.com/nqcfv.jpg

The Pic-2 is from CD 60's music similar to my recording.
http://i.xomf.com/skrlv.jpg
It seems to have much more at the high frequency end than my recording on the Span, yet audibly it sounds about 30% less bright - much more rolled off than my recording which on the Span looks to be reaching only about 18k. Possibly SM57 limitations? Everything is done via tube preamps that go 20k.

Last edited by Tubeguy; 04-21-2017 at 06:53 AM.
Tubeguy is offline   Reply With Quote
Old 04-21-2017, 06:55 AM   #19
Softsynth
Human being with feelings
 
Join Date: Jun 2015
Posts: 3,005
Default

Quote:
Originally Posted by bezusheist View Post
Nyquist for CD audio (44.1 kHz) is 22,050 Hz.
the "limitation" to anything less would be at the ADC or DAC.
so yes, 21 kHz can be recorded and produced to/from CD.
It is capable of 0 Hz to 22.05 kHz. Hence the term "lossless".
Lossless does not refer to frequency response, as implied here. It means compression without information loss. In other words the data is simply stored more efficiently. ALL of the original data is still there.

Lossy compression like MP3 (depending on level of compression) uses psychoacoustic models to get rid of supposedly inaudible sounds that are masked by other sounds.
The greater the compression the more aggressive this psychoacoustic model has to be. Lowest bitrate compression is quite audible to everyone, not only wiping out detail but introducing strange artifacts.
Softsynth is offline   Reply With Quote
Old 04-21-2017, 07:01 AM   #20
serr
Human being with feelings
 
Join Date: Sep 2010
Posts: 6,004
Default

Quote:
Originally Posted by bezusheist View Post
It is capable of 0 Hz to 22.05 kHz. Hence the term "lossless".
Lossless doesn't refer to the level of quality or fidelity a format is capable of. It refers to weather data compression throws away data or not. Lossless data size reduction preserves all the original data and can be expanded back to an exact copy of all the original ones and zeros. (Regardless if those ones and zeros describe a 24 bit HD audiophile recording or an iphone recording of farts.)

You can make an example of a 44.1k wav file being lossy if it is a reduced (sample rate converted) copy of an original 96k recording as well. (The term is normally only used to describe data compression schemes. But if you consider sample rate reduction data compression, then it fits here too.)

Lossy/lossless isn't a format. It describes how you preserve (or not) any number of formats.
__________________
Mac Pro 8x3.33GHz i7, 16GB, 256GB SSD(OS, apps), 3x2TB 7200 HD(data); MacBook Pro 2.8GHz, 6GB, 128GB SSD HD(OS, apps), 750GB 7200 HD(data); 2xTrue Precision 8; Apogee AD-16; 2xMOTU 828mk3, Evolution UC-33e; Faderport; WiRanger, iPad & the analog mixer has retired
serr is online now   Reply With Quote
Old 04-21-2017, 07:01 AM   #21
DrKev
Human being with feelings
 
DrKev's Avatar
 
Join Date: Mar 2015
Location: Paris, France
Posts: 131
Default

Quote:
Originally Posted by Tubeguy View Post
It seems to have much more at the high frequency end than my recording on the Span, yet audibly it sounds about 30% less bright - much more rolled off than my recording which on the Span looks to be reaching only about 18k. Possibly SM57 limitations? Everything is done via tube preamps that go 20k.
Most of the content above 10kHz in the 60's recording is barely heard because limitations of ears. In your recording it's a) the stuff from 3kHz to 10kHz that gives the brightness, AND b) the slope of the spectrum: yours is relatively flat, the 60's recording slopes gradually down from 400Hz as frequency increases.
__________________
Irishman in Paris

Musician / Guitar Teacher/ Guitar Tech / ex-Physicist (hence the Dr in DrKev)
DrKev is offline   Reply With Quote
Old 04-21-2017, 07:22 AM   #22
serr
Human being with feelings
 
Join Date: Sep 2010
Posts: 6,004
Default

Quote:
Originally Posted by Tubeguy View Post
Possibly SM57 limitations? Everything is done via tube preamps that go 20k.
A '57 is a pretty lo-fi mic. It's more about what it doesn't pick up than what it does! Hence the live stage uses because this gives you more isolation than you'd normally be getting away with.

Any perception of bright vs. not bright with a '57 recording must surely be in the 2k - 6k range.

That doesn't invalidate that critique! Just that the issue is not related to the high frequency limit of the system but something else.
__________________
Mac Pro 8x3.33GHz i7, 16GB, 256GB SSD(OS, apps), 3x2TB 7200 HD(data); MacBook Pro 2.8GHz, 6GB, 128GB SSD HD(OS, apps), 750GB 7200 HD(data); 2xTrue Precision 8; Apogee AD-16; 2xMOTU 828mk3, Evolution UC-33e; Faderport; WiRanger, iPad & the analog mixer has retired
serr is online now   Reply With Quote
Old 04-21-2017, 08:54 AM   #23
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by bezusheist View Post
Nyquist for CD audio (44.1 kHz) is 22,050 Hz.
the "limitation" to anything less would be at the ADC or DAC.
so yes, 21 kHz can be recorded and produced to/from CD.
It is capable of 0 Hz to 22.05 kHz.
+1.

Nyquist Theorem shows you can record all frequencies up to half the sampling rate, which allows you to record 21 kHz on CD.

Quote:
Originally Posted by mschnell View Post
Only by some kind of Error. This is forbidden by Physics/Mathematics.

There are tons of videos in the Internet about Sampling, the Nyquist Theorem, and Aliasing, explaining that stuff in detail.
I'm willing to bet $200,000 that those videos are wrong and that I can store a 21 kHz tone in Audio-CD format.

Last edited by TryingToMakeMusic; 04-21-2017 at 09:00 AM.
TryingToMakeMusic is offline   Reply With Quote
Old 04-21-2017, 09:39 AM   #24
hopi
Human being with feelings
 
hopi's Avatar
 
Join Date: Oct 2008
Location: Right Hear
Posts: 12,234
Default

what is the highest freq you can hear?

I'm old so the hi ends are sadly gone for me... but even my kids who have super hearing don't hear much above 20 k....

so OK you may see it on a scope but do you hear it?

If a CD falls in the forest and there is no one there does it make the top 40?
__________________
...should be fixed for the next build... http://tinyurl.com/cr7o7yl
https://soundcloud.com/hopikiva/angel-rain
hopi is online now   Reply With Quote
Old 04-21-2017, 09:54 AM   #25
Softsynth
Human being with feelings
 
Join Date: Jun 2015
Posts: 3,005
Default

Quote:
Originally Posted by hopi View Post
what is the highest freq you can hear?

I'm old so the hi ends are sadly gone for me... but even my kids who have super hearing don't hear much above 20 k....

so OK you may see it on a scope but do you hear it?

If a CD falls in the forest and there is no one there does it make the top 40?
Gnat farts are very musical dontcha know!
Softsynth is offline   Reply With Quote
Old 04-21-2017, 11:56 AM   #26
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by hopi View Post
what is the highest freq you can hear?

I'm old so the hi ends are sadly gone for me... but even my kids who have super hearing don't hear much above 20 k....

so OK you may see it on a scope but do you hear it?

If a CD falls in the forest and there is no one there does it make the top 40?
Inaudible can become audible during processing. Processing is often ignored by theoreticians on DAW forums---as if recording audio into your DAW is the final step and that's all anyone will ever hear?---but in real life, processing recordings is routine.
TryingToMakeMusic is offline   Reply With Quote
Old 04-21-2017, 05:36 PM   #27
gpunk_w
Human being with feelings
 
Join Date: Dec 2010
Posts: 3,120
Default

Quote:
Originally Posted by hopi View Post
what is the highest freq you can hear?

I'm old so the hi ends are sadly gone for me... but even my kids who have super hearing don't hear much above 20 k....

so OK you may see it on a scope but do you hear it?

If a CD falls in the forest and there is no one there does it make the top 40?
Doubt I get anywhere close to 20 all the gigs and crap, hi end studio mix for whales lol
__________________
Reaper scripters, that is all !!!!!!!!!!!!!!!!
gpunk_w is offline   Reply With Quote
Old 04-21-2017, 06:04 PM   #28
Bri1
Human being with feelings
 
Bri1's Avatar
 
Join Date: Dec 2016
Posts: 556
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
I'm willing to bet $200,000 that those videos are wrong and that I can store a 21 kHz tone in Audio-CD format.
Show and tell please?
People only have to amplify to hear certain frequencies.. they are not invisiaudible.
Bri1 is offline   Reply With Quote
Old 04-21-2017, 07:15 PM   #29
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by Bri1 View Post
Show and tell please?
Just sample a 21 kHz sine wave. The rumor that it doesn't work seems odd to me.

Last edited by TryingToMakeMusic; 04-21-2017 at 08:08 PM.
TryingToMakeMusic is offline   Reply With Quote
Old 04-21-2017, 07:17 PM   #30
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
Just sample a 21 kHz tone. The rumor that it doesn't work seems odd to me.
Perhaps some were thinking CD sample-rate is 41.4 kHz? That would preclude 21 kHz recording. But CD sample-rate is actually 44.1 kHz.
TryingToMakeMusic is offline   Reply With Quote
Old 04-21-2017, 10:41 PM   #31
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Even if you can't hear that high, mute your master before doing this. It is full-blast:

Code:
in_pin:left input
in_pin:right input
out_pin:left output
out_pin:right output
@init
i = 0;
@sample
spl0 = sin(i * 2 * $PI * 21000 / 44100);
i += 1;
You all owe me $200,000.
TryingToMakeMusic is offline   Reply With Quote
Old 04-21-2017, 11:20 PM   #32
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Germany
Posts: 2,862
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
But CD sample-rate is actually 44.1 kHz.
Please do respect that with physical sampling, the (analogue) syntheses and analyzing filters need some (frequency) range to work. Hence the range between (some) 20,000 and 22,050 Hz does not provide any decently usable information. What might be found there will (increasingly with the frequency visible in digital spectrum diagram) result in distortion rather than a reproduction of the originally sampled signal.

If you use the CD as an example this is 16 Bits which theoretically is 20*log(1/(2^16)) = -96 dB digital noise

20,000 to 22,100 Hz is 0.144 of an Octave. So a filter allowing for 20,000 Hz and forbidding for 22,100 Hz with the same grade of distortion as the 16 Bit resolution would need to feature a steepness of 96 / 0.144 dB/octave. But some 700 dB/ Octave in fact is impossible to do, so the A/D and D/A converters need to provide a compromise, anyway.

-Michael [edited two typos regarding erroneously set Zeros]
__________________
www.boa-sorte.de

Last edited by mschnell; 04-22-2017 at 09:01 AM.
mschnell is online now   Reply With Quote
Old 04-21-2017, 11:27 PM   #33
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Germany
Posts: 2,862
Default

Quote:
Originally Posted by TryingToMakeMusic View Post
de]You all owe me $200,000.
What you programmed is a non-anti-aliasing tone generator. It generates nothing but wrong artifacts in the frequency range above (and near to) Nyquist.

If you use it to create a frequency well above half the sampling frequency, you will hear a tone well below the sampling frequency, the generator was not intended to create. Tis is called aliasing.

(Shot in your food )

-Michael
__________________
www.boa-sorte.de
mschnell is online now   Reply With Quote
Old 04-21-2017, 11:46 PM   #34
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

The only thing special about the value 20 kHz is that it was chosen by Voxengo as the default value of a parameter affecting display only, leading to the OP's temporary (now resolved) confusion.
Quote:
Originally Posted by mschnell View Post
Please do respect that with physical sampling, the (analogue) syntheses and analyzing filters need some (frequency) range to work.
This is the case with every single filter ever, isn't it?

Quote:
Originally Posted by mschnell View Post
Hence the range between (some) 20.0000 and 22.05 Hz does not provide any decently usable information.
How are you arriving at the figure "20 kHz"?

Quote:
Originally Posted by mschnell View Post
What might be found there will (increasingly with the frequency visible in digital spectrum diagram) result in distortion rather than a reproduction of the originally sampled signal.
I understand that ideal low-pass are unattainable in practice, but that doesn't imply that 20 kHz is anything special, and it also doesn't mean that 21 kHz tones alias. You could use the same argument to conclude that everything over 10 kHz is bad, or everything over 5 kHz is bad, because something, something filters.

OP just needed someone to tell him to change a setting in a Voxengo product. Instead we got a lot strange theories about why it would be fine for Reaper to mess up the high end.

Last edited by TryingToMakeMusic; 04-22-2017 at 12:24 AM.
TryingToMakeMusic is offline   Reply With Quote
Old 04-22-2017, 12:23 AM   #35
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by mschnell View Post
It generates nothing but wrong artifacts in the frequency range above (and near to) Nyquist.
Prove it. (You can't.)

Quote:
Originally Posted by mschnell View Post
What you programmed is a non-anti-aliasing tone generator.
No; a continuous 21 kHz sine wave transforms to a Dirac-Delta function in the frequency domain shifted to 21 kHz, which is absolutely unaffected if you then multiply by a Rectangle function to apply an ideal low-pass at Nyquist. Your "anti-aliasing" design is unnecessary here and would only turn a perfect non-aliased signal into something worse.

You won't get aliasing unless there's frequency content over Nyquist to begin with. There's no frequency content over Nyquist to begin with in my example; it is band-limited below Nyquist from the start. You're confusing it with naive implementation of a Saw or Square wave; I didn't do that though.

Quote:
Originally Posted by mschnell View Post
If you use it to create a frequency well above half the sampling frequency,
And you believe that 21,000 is well above Nyquist (22,050)? I'll leave you to that.

Last edited by TryingToMakeMusic; 04-22-2017 at 01:15 AM.
TryingToMakeMusic is offline   Reply With Quote
Old 04-22-2017, 12:47 AM   #36
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

If you want to see Shannon's proof of the Nyquist Theorem, I don't think it's on YouTube, but it's here (p. 448):

http://www.ee.oulu.fi/~kk/dtsp/tutor...annonpaper.pdf
TryingToMakeMusic is offline   Reply With Quote
Old 04-22-2017, 02:01 AM   #37
bezusheist
Human being with feelings
 
bezusheist's Avatar
 
Join Date: Nov 2010
Location: Dummytown
Posts: 272
Default

Quote:
Originally Posted by mschnell View Post
What you programmed is a non-anti-aliasing tone generator. It generates nothing but wrong artifacts in the frequency range above (and near to) Nyquist.
it generates nothing but a "pure" 21 kHz sine wave...
__________________
the quim reaper
bezusheist is online now   Reply With Quote
Old 04-22-2017, 03:11 AM   #38
Bri1
Human being with feelings
 
Bri1's Avatar
 
Join Date: Dec 2016
Posts: 556
Default

Quote:
I'm willing to bet $200,000 that those videos are wrong and that I can store a 21 kHz tone in Audio-CD format.
Quote:
Show and tell please?
Quote:
Code:
in_pin:left input
in_pin:right input
out_pin:left output
out_pin:right output
@init
i = 0;
@sample
spl0 = sin(i * 2 * $PI * 21000 / 44100);
i += 1;
You all owe me $200,000.
This shows not what you claimed.
Can you show the recorded cd-then on a scope please?
Nobody owes anybody anything-we all want to learn by sharing.
Bri1 is offline   Reply With Quote
Old 04-22-2017, 03:23 AM   #39
TryingToMakeMusic
Banned
 
Join Date: Nov 2016
Posts: 416
Default

Quote:
Originally Posted by Bri1 View Post
Can you show the recorded cd-then on a scope please?
Nobody owes anybody anything-we all want to learn by sharing.
I want to learn by sharing, so please show me on a scope from a recorded CD the aliasing from my implementation?
TryingToMakeMusic is offline   Reply With Quote
Old 04-22-2017, 03:27 AM   #40
Bri1
Human being with feelings
 
Bri1's Avatar
 
Join Date: Dec 2016
Posts: 556
Default

Quote:
I want to learn by sharing, so please show me on a scope from a recorded CD the aliasing from my implementation?
I see a tone,but it's the burning process i'm interested to see how u burn that cleanly..
I'm not suggesting your 'wrong' in anyway-just curious of debate.
Bri1 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 02:02 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2017, vBulletin Solutions, Inc.