TIA
Nyquist Freq?
-
- KVRian
- 520 posts since 13 Aug, 2002 from Salzburg, Austria
-
- KVRAF
- 4738 posts since 20 Feb, 2004 from Gothenburg, Sweden
It's the highest signal frequency one can play back using a specific sampling frequency. This is always half of the sampling frequency.
So if having 44100hz sampling frequency the highest signal frequency that can be sampled / reconstructed is 22050hz.
Just use google.
So if having 44100hz sampling frequency the highest signal frequency that can be sampled / reconstructed is 22050hz.
Just use google.
Stefan H Singer
https://dropshotaudio.com/
https://dropshotaudio.com/
-
- KVRAF
- 4020 posts since 2 Sep, 2003 from Perth, Australia
The simplest signal you can represent in a discrete system is a sine wave, and that requires at least one "up" and one "down" value per cycle - hence why the Nyquist frequency is 1/2 the sampling frequency.
-
Stupid American Pig Stupid American Pig https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=4753
- KVRAF
- 7065 posts since 25 Nov, 2002 from not sure
above nyquist, you experience a phenominon called "aliasing". "Aliasing" is a subject of great debate- and while usually this should be relegated to the developers to hash out, you usually end up with end users trying to say how a certain synth/sampler is an Arsebiscuit(thanks for the word quincy) because it "aliases".
My general sentiment is that these threads usually pan out like this:
post 1"This synth sucks coz it aliases- what kind of synth engine is it? It doesnt sound like a virus or 303"
Post 2"this synth roxors, you are the one that sucks for not liking this synth
"
Post 3"everybody sucks, and I cant hear the aliasing"
Post 4"Its there plain as day when you hit note 127 on your midi keyboard and boost the treble 12 db- you suck for not being able to hear it, you must not be a pro- my ears are better than everyone elses, plus I think that Cubase sounds better than FL studio, I can prove it coz my ears hear it"
and so on...
sorry for the sarchasm, but look around and you will see Im not *that* far off base
My general sentiment is that these threads usually pan out like this:
post 1"This synth sucks coz it aliases- what kind of synth engine is it? It doesnt sound like a virus or 303"
Post 2"this synth roxors, you are the one that sucks for not liking this synth
Post 3"everybody sucks, and I cant hear the aliasing"
Post 4"Its there plain as day when you hit note 127 on your midi keyboard and boost the treble 12 db- you suck for not being able to hear it, you must not be a pro- my ears are better than everyone elses, plus I think that Cubase sounds better than FL studio, I can prove it coz my ears hear it"
and so on...
sorry for the sarchasm, but look around and you will see Im not *that* far off base
-
- KVRAF
- 1907 posts since 29 Oct, 2003
somebody mis-spelled "no quest" and so the *nyquist theorem* was born.
-
- KVRian
- Topic Starter
- 520 posts since 13 Aug, 2002 from Salzburg, Austria
Thanks guys 
So, when I get a sample, play very high notes and it sounds distorted, weird harmonics...thats because the sample goes above the nyquist freq right?
So, when I get a sample, play very high notes and it sounds distorted, weird harmonics...thats because the sample goes above the nyquist freq right?
-
- KVRian
- 574 posts since 6 Jan, 2003 from Somewhere between ))o Left and Right o((
Actually the Nyquist theorem says that you need at least 2 samples to store a sine wave on a digital medium, and then recontruct it with lowpass filtering in analog domain.
That is why the Nyquist frequency is half of the sampleRate.
The highest harmonic of a signal must be stored on at least 2 samples.
That is why the Nyquist frequency is half of the sampleRate.
The highest harmonic of a signal must be stored on at least 2 samples.
-
- KVRian
- 574 posts since 6 Jan, 2003 from Somewhere between ))o Left and Right o((
Yes that is called aliasing (already mentioned) and it's a harmonic distortion. The spectrum of the signal is mirrored back at NyquistFreq, than it also mirrores at 0Hz, and you will end up with a lot of inharmonic content.SPYRO wrote:Thanks guys
So, when I get a sample, play very high notes and it sounds distorted, weird harmonics...thats because the sample goes above the nyquist freq right?
You can verify a harmonic as beeing alias or not by divide Fnyquist / Fharmonic, if you get an integer number, it is possible that you have aliasing.
-
- KVRAF
- 2830 posts since 2 Mar, 2003 from The only civilized county in Texas
You got it. If the Nyquist frequency if f and you try to play f+x, then it comes out as f-x, which is almost always inharmonious.SPYRO wrote:So, when I get a sample, play very high notes and it sounds distorted, weird harmonics...thats because the sample goes above the nyquist freq right?
I think I posted the mathematical derivation of this in the aliasing thread.
V.
-
- KVRAF
- 1907 posts since 29 Oct, 2003
sortof similar to AM radio.
-
Voidoid Surrealist Voidoid Surrealist https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=41079
- KVRAF
- 4048 posts since 18 Sep, 2004 from Places far less tedious than this blue trainwreck...
Ok, that explains the Nyquist frequency, but can anyone here help me understand the Nyquil frequency? Or why the holy f**k would you make the standard flavor of your product black licorice? The mind boggles...
-
- KVRAF
- 2460 posts since 3 Oct, 2002 from SF CA USA NA Earth

