White Noise vs Pink Noise [Audio Engineering & Music Production]

Рет қаралды 67,532

Күн бұрын

Пікірлер: 49

@breernancy 5 ай бұрын

Never heard the uniform energy per octave explanation for pink noise before; always just that its more realistic from a human hearing perspective. Thanks.

@SashaGarcia 3 жыл бұрын

The ability to smile with your eyes is an underrated art.

@curtvincent3728 3 жыл бұрын

Brilliant! Really well done! Thank you. I am an old audio engineer, just a hobbiest these days. But I swear I did a better job back on analog boards as today I tend (like anyone else) to mix with the eyes using a graph. This is especially true with parametric eq. Bad, bad habit.

@leftlane5085 3 жыл бұрын

i'm actually working on a project for school relating to the generation of white/pink noise. Basically, a white noise source is totally uncorrelated. Think of a spinner divided into equal parts with each part corresponding to a different tone. Then, a "white melody" can be generated by spinning the spinner many times and recording the tones the spinner lands on, one after another. Any given tone will be totally uncorrelated to the tone that came before it. This can be contrasted with a brown noise (1/f^2) source which is highly correlated. Think of starting at a certain tone. Then, the next tone may only deviate slightly higher/lower. Continue this process to generate a "brown melody". You'll notice that a spectral analysis of brown noise has an even steeper negative slope than pink noise. Pink noise is somewhere in between white and brown noise. It is not as correlated as brown, but it is more correlated than white. Pink noise is also called "1 over f" noise. A "pink melody" can be generated using the Voss algorithm which is actually fairly easy and elegant to describe, but I don't have the energy for that right now.

@AudioUniversity 3 жыл бұрын

Thanks for sharing this information, therealcarlgauss!

@andybedingfield7035 3 ай бұрын

Higher frequency waves hit you more ‘frequently’ so they carry more energy than low frequency waves at the same amplitude. It makes sense that you need more amplitude in the low end to balance the delivery of energy. Thanks for the great video!

@paulo_00 3 жыл бұрын

The content you share is extremely good. Thank you.

@macronencer Жыл бұрын

5:00 Wow, nobody has ever explained it to me this way before. That's awesome!

@VAMSIMNS 3 жыл бұрын

Hi Bro, I really like the way you put the theory onto the board. Crisp & covers all the necessary information.

@joho-z Жыл бұрын

This is very well done and simply explained for a complicated subject. Thank you!

@AudioUniversity Жыл бұрын

Thanks for watching!

@sammiller9855 3 жыл бұрын

For reference in mixing (using a match EQ like TDR's Nova GE), I often use -4.5/oct noise, between brown (-6db/oct) and pink (-3db/oct), which is closer to most modern music, and dial in around 30% or less of the match and then make manual adjustments. (For those that don't know, the negative slope of pink noise, -3db/oct, is relative to white noise, which is considered 0db/oct, Pink noise can be created from white noise by applying a -3dB/oct filter). Sometimes, I will use a noise reference match only as a guide to know where further down in my mix (sub-buses and individual tracks) I may want to make some EQ adjustments to minimize EQing on the master bus. Pink as a reference, although much better than white, still tends to be too bright, lacking lows. As stated in the excellent scientific article, Long-term Average Spectrum in Popular Music and its Relation to the Level of the Percussion, available free online, the frequency spectrum slope has been estimated across genres to be approximately -5 dB/octave on average with the slope steepening in the far upper higher frequencies, which is what I have often found through analyzing various songs with a spectrogram. While pink and brown noise is easily available online, other slopes are not. To create my own custom noise slopes in a DAW such as Ableton (e.g., -4.5db/oct and -5db/oct), I use a noise generator plugin such as the free Pink plugin, a tilt EQ , such as the free stiltV2, and a spectrogram such as the free Span. After selecting the target slope in the spectrogram, I adjust the tile EQ until the frequency profile display flat, horizontal in the spectrogram. By the way, excellent content! Please keep the videos rolling out.

@AudioUniversity 3 жыл бұрын

Thanks, Sam! This is great info! Here is the article Sam is referring to for anyone interested... Long-term Average Spectrum in Popular Music and its Relation to the Level of the Percussion (AES): www.aes.org/e-lib/browse.cfm?elib=18638

@tunemxr480 3 жыл бұрын

Awesome as usual! Have you heard of John Meyer’s latest concoction between doobs? M-Noise. He swears it’s a superior FFT analysis noise than pink?!?! (He smokes a lot, but he is an audio genius. I would have had no better side fills than UPA-1’s thanks to him back in my live sound days!)

@AudioUniversity 3 жыл бұрын

Thanks, Steve! I’ll have to check out M-Noise. I hadn’t heard of that before.

@nickpenacl_ 4 ай бұрын

Excellent, where did you get those nice plugs-ins ??

@vonnieh7259 2 жыл бұрын

Can you make an example attaching the piano to device that creates Cymatics using the sand with hertz to expose each image it creates w each note as in creating alphabet images thru the frequency sounds 🤔

@_abhinavprajapati 3 жыл бұрын

Hi Kyle ! I'm following you since the beginning of the channel and really I've learned alot from you ! Thanks for sharing the Information. Can you Give detailed difference between all the synthesis like wavetable, Modular etc. ?

@AudioUniversity 3 жыл бұрын

Thanks for supporting the channel from the beginning, Abhinav. I’ll try to make videos that explore these subjects!

@XRaym 3 жыл бұрын

Well explained! It worth noticing that white nosie being flat on spectogram is only true if the Slope of the vizualizer is set to 0dB... which is not the case in most EQ/vizualizer VST oriented to ward mixing (Span, Fab Filter Pro-Q etc), which in this case, have 4.5dB slope per default, for which white noise isn't flat.

@AudioUniversity 3 жыл бұрын

Good point! Thanks.

@peace4531 3 ай бұрын

And ironically I could hear pink noise in the background of your voice for the entire video is it an Easter egg or just a coincidence?

@jitendradewangan9485 2 жыл бұрын

thank you, kyle. you have done such a fantastic job and I really like the you way teah us. The audio engineering easy with kyle.

@lapetitemortbis Жыл бұрын

Hi. Great content! I use Audirvana in my system to play through Qobuz. In any case, sometimes music suddenly gets interrupted by white noise or pink noise ( I honestly don’t know which one). Is this damaging to speakers?

@sonofguns1016 8 ай бұрын

Ok I’m subbed now! I wasn’t sure at first but yeah I think it’s worth it lol

@OneStepToday Жыл бұрын

Great video. You should have explained the theory, why exactly the distance between frequencies keeps increasing, or that it is a linear curve, that would be interesting. I just downloaded a good pink noise, saw about it on the focus, motivation videos of Andrew Huberman who recommended that noise naturally generate or increase alertness in our mind. So they are great tool for studying or work focus enhancement.

@jirlaaaay9051 2 жыл бұрын

Great video, thanks a lot

@adi.sax.ardelean 2 жыл бұрын

Hello. if I want to linearize my system which is more suitable pink noise or White noise? if, for example, white noise is linear, so should the linear system? if pink is slightly oblique then do I have to keep the slope that pink noise gives? if I linearize with pynk noise then the upper middle part will be louder than the lower part under the bass! so which is easier to use and respect white or pink?