Audio Dither Explained

This video explains dithering in digital audio-what it is, and why we use it. Additional articles on dither, including examples, are available at earlevel.com: www.earlevel.com/main/categor...
Please visit EarLevel Engineering at earlevel.com for more tutorials on digital audio signal processing and sound synthesis.

[@OysterWallace]

Your demonstration and final statement were the best explanations of dither I've come across. The only thing I'm concerned about is this video possibly perpetuating the myth that final converted audio signals look like their digital stair-step counterparts.

[@nigel_redmon]

Kali Doom Thanks Kali. However, the "stair-step" idea is a misinterpretation-I think you understand that. We're looking at digital values in this demonstration, not the output of a D/A converter, and these values can quantize into repeated distinct levels that look like stair steps. Granted, "stair-step" automatically conjures up that poor and incorrect description of D/A output for some, and I could have used something like "plateau" instead. But I don't think it will perpetrate a myth, because it's an accurate description of what I'm showing, and if people get confused, they don't have a grasp of what's being demonstrated anyway. But, I know now to avoid the words "stair-step" in digital audio should the need arise again. Actually, I'm working on a sampling theory and sample rate conversion video (possible as separate videos, depending on how long each part runs) that will make it abundantly clear that samples represent impulses, and exactly why the output of a D/A is a smooth analog waveform. I have older articles on these topics on my website, but video will give the opportunity for a more animated demonstration, with audio.

[@bassjio]

Even 8 years on this video is fantastic. I think an auditory example of the effect of dither would have been nice, but the excellent visual examples more than sufficed.

[@nigel_redmon]

Thanks for saying. There are examples in my other video on dither: kzbin.info/www/bejne/gXTccmmCoadmd9U

[@mikkokylmanen9296]

What a great visual demonstration of the concept in the beginning!

[@cultserpent4501]

The best explanation of dither i've ever seen. Thank you so much!

[@Illjustwait]

Nigel, you are a humble yogi imparting wisdom amidst a vast ocean of misinformation. I love you ❤️

[@nigel_redmon]

Thank you Sergio!

[@infrasonic436]

As someone studying digital design this was insanely helpful. The example at 1:15 is nothing short of mind bending.

[@nigel_redmon]

Thanks for letting me know-glad it as helpful to you!

[@promusic2k6]

Nigel Redmond, we at 2600Hz Brazil would like to thank you and congratulate you for making this video! It is the best video on dithering and truncation we could ever find. We always wondered why dithering went up to 3 bits (as opposed to only 1 bit), and it got cleared for us when we learned about Rounding from you. "Dithering smooths out the transition between bits". That is amazing! And your widget is also incredible. Thank you for that work!

[@Meknassih]

The visual analogy at 0:35 with pixels has blown my mind 🤯. I've been wondering what dithering is for about 10 years now, it always seemed something too difficult to understand for a regular musician, but thanks to you, I know what I'm doing now haha.

[@pinkponyofprey1965]

Never ever have I heard a better description of dither! Thank you, Sir! :D

[@iambaney]

I've scoured the internet for resources on how to explain this most clearly to my student and this is easily the best explanation of it I've found. I hope you find time to do more videos on digital audio!

[@tofu.delivery.]

Thank you so much! I export my audio every time without knowing what the dither option does, and I checked the box anyway, but now I know to always use it!

[@_thom-ass]

Bro thank you so much. This is an AMAZING explanation of dither, you are an amazing teacher. Turning such a complicated subject into something easy to understand is insanely difficult, yet you managed to do it. Thank you

[@nigel_redmon]

Thank you so much, it helps to know-be sure to subscribe if you haven't so you don't miss my (long delayed, but I won't let another year go by) intuitive explanation of sampling theory...

[@QuabmasM]

Very well done video. This explained dithering like none other...I truly truly understand it now to the point where I can explain it clearly. Thank you for your contribution to musical education. I hope everybody checks this one out.

[@nigel_redmon]

I want to make one thing very clear: The animated waveform graph does NOT depict a "staircase" wave. It simply connects the dots in order to make the display comprehensible. Without the lines, it's just a huge mess with the dots dancing around (and even worse if drawn as impulses). And note that adjacent dots at different sample values are connected *diagonally*-it's NOT a staircase, just "connect the dots". I noted this in the video, expressly to make this point clear. Samples are impulses. In this case, I've simply connected their tops to give an outline that's easier to follow visually.

[@JustinNewton]

Hey, I got your message, actually I was replying to a Google+ post by Sigurdór Guðmundsson sharing this video. That's probably why it was confusing to reply to. Anyway, I appreciate that you get that the staircase isn't really what's going on, but since I was taught this this misconception in college, and had to correct it countless times since then, (interns, artists who know enough to be dangerous, etc.) I speak up when I see digital audio visualized with the zero-hold graph. We engineers need to stop putting that visualization out there so we can start to put an end to that idea. Really "diagonally" is not strictly correct either, it's a parabola most accurately. Only one possible solution for the D/A converter to find. But anyway, I'm not meaning to come down on you, Nigel Redmon like it must seem like I was --sorry about that! It's a great visualization of bit depth and dither, the grey/black graphics, I just think drawing the graph like you did leaves people with an exaggerated idea of what these distortions would look like in a waveform, when really they'd be much smoother than all those squared off points and lines.

[@nigel_redmon]

Justin Newton No, I don't feel at all that you're getting on my case, so to speak. I'm just as sensitive to the point as you are, I'm sure. But to be clear, I didn't connect the dots to be "correct" in any manner at all. The connected line is a visual aid, like circling a point on a chart with a red marker to highlight it. Believe me, if I turned off the connecting line, people would not be able to make heads or tails of the chart-it would be just points dancing around. I tried it. The line is just there so that you can follow from one sample to the next clearly, especially as I change the amplitude.

[@nigel_redmon]

Justin Newton And yes, I did consider doing band-limited reconstruction to simulate what would come out of an analog converter, but the visual difference for the staircase-ish part is relatively minor with the samples that close together (the dots would be obscuring most of it-I'd have to hide the dots, then I'd have lost what it is I trying to illustrate), and in the animated noise-added view it would be just messy and distracting, zoomed in that closely. The object is to let the eye follow that sample points, not to render an analog view.

[@iblesbosuok]

You remind me of _'Sweet Dither Bitter Jitter'_ paper which published many years ago. Wonderful video.

[@nigel_redmon]

I couldn't find that paper, but thanks!

[@lucr1369]

Great explanation!

[@numbersix9477]

I found your video inspiring. I've never seen a better lay level presentation of a technical topic. WELL DONE! I fully intend to watch your other videos. I watched and enjoyed despite having learned everything you covered several years back when I was in the process of capturing to digital from a number of family reel to reel and cassette recordings. LOL! I captured at 88.2 kHz, 32 bit floating point, edited and did mild hiss reduction then used noise shaping dither when normalizing and transcoding to 44.1k (end product-CDs) - all on material with a noise floor 30 dB or more above the dither. It was extreme overkill but I wanted an absolutely minimal quality impact as I was only going to do it once. The process was very educational and I'm happy with the result.)

[@Noone-of-your-Business]

Mind BLOWN. I have never seen dither explained so _precisely_ and well illustrated! Finally, something I can *understand!* Your explanations *easily* beat anything on Reverb's channel. Subscribed!

[@nigel_redmon]

Thanks! I do appreciate the feedback. And I'm currently working on a couple of major videos after being too busy for a few years, so I'm glad you subscribed.

[@FarhatGuitarOfficial]

The best explanation in youtube !! Thanks!!!

[@nigel_redmon]

Thanks, I appreciate you saying!

[@RichardPhilbin]

Wow, great explanation!

[@brankoskara7201]

Nice work !

[@TLOEric]

This was incredible. Thank you.

[@xguidosan]

Best explanation ever!

[@WFNKcom]

OUTSTANDING VIDEO!! Misunderstanding of dither lead to the horrible lossy decades we've just experienced. From 0:15 to 0:25 you lay out the fundamental point that every digital music listener and user needs to understand...... the native resolution is the BEST. Dither is *only needed* if you must reduce that native resolution for convenience (final format size restriction) reasons. Every person who's ever said 24bit audio is a scam can't deal with the first 30 seconds of this video. Re-watch it and focus on every word. If there' was nothing we can hear beyond what 16bit word length could capture we wouldn't need dither for music production. But no mastering engineers work without it b/c it is necessary to cover up LOSS.

[@CapsAdmin]

Nice explanation. Although it's not centered around graphics I would add that pure randomness is the "naive" way when dithering an image. There are other less random algorithms that make the perceived image quality much better by placing the pixels in specific patterns. You briefly touched on this point for audio at the end.

[@reayxx]

best explaination ever!

[@novivi4390]

Best videos about dithering short but complete thanks

[@LokeybeatsInstrumentals]

Thanks for the clear explanation!

[@SveinOlavGlesaaenNyberg]

Excellent explanation of dither. Thank you!

[@nigel_redmon]

Appreciate it, Svein

[@VirtualMark2]

Great article, thanks.

[@shayanja6749]

Wow this is beautifully explained

[@celopezmusic]

Best explanation I´ve seen so far, congrats!

[@keanumanning]

thanks man, im a music tech student and this helped

[@esmaeilhashemi1551]

So useful. Thank you so much.

[@Svm777]

Really good info and video, Nigel! Thank you very much!

[@averywang6922]

Nice video. An interesting visualization to add to this video would be to show the FFT of a sine tone several bits below the LSB. And then with shaped noise.

[@jay2xtremefy]

Best explanation. Thank you

[@LawrenceAaronLuther]

very helpful, thank you

[@davidevans1420]

Excellent. Best I've ever heard...

[@nigel_redmon]

Thanks, David

[@rampagetv]

Feel free to start uploading again if you have the time at all. Your videos are very easy to watch while learning a lot. Hard to nail that one. Have a great one!

[@nigel_redmon]

Thanks! I know it may seem impossible, but I'm so close to finishing a new video-and it's prerequisite to another. maybe I'll get faster at doing this with practice...

[@StringerNews1]

Awesome visual aids! One of the best.

[@VAMSIMNS]

Super explained

[@j.thorgard]

Thank you very clear

[@th1alb]

great visuals dude

[@hukl3945]

Well done!

[@zga8]

Great explanation Thanks !!!!

[@davidm.johnston8994]

Great video, man! Thanks

[@otavakarjalainen]

There is so much erroneous or misleading information on bit depth and dithering.on the web. This is best explanation I have seen on this topic which is at the end of the day quite easy to understand. Thank you!

[@WFNKcom]

Business interests want LOSSY. It's how they profit. They have been trying to convince us we can't hear the difference for decades. If you could sell 10% of something and make the same, you probably would. I understand. But full, native resolution recordings are the best, This is lossless, and if you dither from 24bit to 16bit, you technically have loss. No one wants to admit it b/c $$

[@michaelcardoso5914]

Very helpful, thanks!

[@testpattern701]

A nicely done video.

[@nmitchellfinch4208]

Fuck you

[@codemiesterbeats]

thanks for that explanation... It makes sense now I just have to figure out how to use it properly

[@ACDCBoy62]

Burger = flipped

[@JoeySavage]

Best in I found in music will always put this to good thinking n doing use thsnk you im blessed to have a understating mind thsnk God I got my mind

[@artemonstrick]

Awesome!

[@sb-di3of]

thank u senor

[@eheadj5712]

Awesome!!

[@davedilena3214]

That is an amazing explanation; I think the initial explanation could have very far reaching affects explaining probability distributions and why Quantum Mechanics leads to statistical distributions, it may simply be the information that persists the longest.

[@age365]

always wondered what dithering actually did instead of people saying to just make sure you dither.

[@Robil63]

Woah, Dither rules, man!

[@VSARecordsDelhi]

Great

[@johneygd]

It’s mind blowing that we can do more with less.

[@nigel_redmon]

I've added the widget to my website-you can use it with or without "connecting the dots": www.earlevel.com/main/2014/03/15/dither-widget/

[@HealthyBodyForLife]

Gosh, I just wish you knew what you were talking about, can demonstrate it with visual examples, and clearly explain what the heck dithering really is! :D

[@andrewandrosow4797]

This method also use to improve of PWM resolution

[@DANNFIGDESIGNS]

For one, I will have to re-wind a couple of times. 2, I may need to run this through my "Laymen's Term" machine. JK. But yeah... pretty scientific. So far... adding noise helps. Got it. Love noise. But I really need to grasp what the heck is up with Rectangular, Triangular, POW-r 1,2 and 3. I'm sure it's simple once I wrap my head around it. :P But seriously... thanks for this video.

[@cyrusworldmusic]

very cool explain !(python notebook +Bokeh flavour!)

[@refaelshansi]

Hi, Thanks for the video. I have a question regrading the noise that you add in the dithering: should the noise level PSD should be by the value 0.5X(q/srt(12))^2? Since for my understanding, when adding to much noise, higher than ( LSB/2), you can degrade the ADC performance. Will appreciate your explanation. Refael

[@nigel_redmon]

Hm, I think that's right except for the factor 0.5. The noise amplitude is ±1 LSB (2 LSB peak to peak). That's the minimum dither level to ensure no signal correlation and no noise modulation. Offhand, I'm not sure about the LSB/2 reference, but take care that sometimes dither levels are expressed in rms.

[@zlbdad]

THis is excellent, thank you! Do you have an appetite to tackle aliasing?

[@nigel_redmon]

Thanks, Jonathan. Yes-I've been wanting to do a video on sampling theory, expect to get to it this year. I wrote somewhat of an outline for it, starting here: www.earlevel.com/main/2017/08/16/sampling-theory-the-best-explanation-youve-ever-heard-part-1/

[@euphoriess]

nice explaination, I understand it immediately but why I can't access your website? I want to read more

[@nigel_redmon]

Thanks! You should be able to reach the website at earlevel.com...it's probably because the link has "http", from back before the site was secure-only. I'll update it...

[@reinforcedpenisstem]

There was a bit of noise-shaping at the the end there, wasn't there?

[@ZipplyZane]

The main reason I wanted to listen is that I wanted an example. I can read this information everywhere. But I want to hear the difference.

[@nigel_redmon]

You'll get that very soon! I've been terribly busy, but hope to finish my follow-up video in the next week or so. You'll heard comparisons of dithered and truncated digital audio at different bit levels.

[@michaelgraflmusic]

Now I know where to direct people to when dithering audio comes up again and everyone spirals into a confused panic.

[@bubbapinks5175]

this is so complicated. where does one go to learn such a thing

[@zia-i]

1:33 why we don't use the 8 bit instantly?

[@nigel_redmon]

In the case of black and white print media (newspaper, magazine, b&w photocopies for handouts) such dithering is done because there are no gray levels of ink available.

[@CSparzo]

im so fucked

[@meowchin]

So, why do we even use 24 bit depth in studios? Wouldn't it be better to just master everything in 16 bits?

[@nigel_redmon]

To keep it brief: It's true that 16-bit is enough for audiophile quality, when fully utilized and with linear processing, but it lacks flexibility. But it's not nearly enough for heavy non-linear processing for example (for instance, say you record clean electric guitar, and want to apply an amp simulator for a overdriven sound-it will raise the 16-bit noise floor to an intolerable level). Maybe a simpler way to view it is that for only 50% more data storage (memory is cheap!), you get 256 times the dynamic range to work with (48 dB more).

[@meowchin]

I was thinking more about the necessity of dithering when downsampling from 24 to 16, rather than storage constrains. If everything was mastered in 16 bits, there would be no such necessity.

[@nigel_redmon]

You wouldn't save on dither: When you master to 16 bits, that's when you'd have to dither. In fact, any time you apply processing-including a minor level adjustment-then save to 16-bit, you'd need to add dither. The key point is that a multiply expands resolution. For instance, if you have two fractional decimal numbers, hundredths precision (0.00 to 0.99), and multiply them, the result has a resolution of thousandths. 0.93 x 0.96 = 0.8928, for instance (perhaps that's a sample value and a gain fader value). If you try to fit that back into hundredths (0.89), you've truncated-you need to dither for a better error distribution.

[@meowchin]

Ok, thanks for explaining!

[@JVonD]

So if your using a project that is in 16bit 44.1 then you don't need to use dither at all when making the cd quality wav files right?

[@spritemasters9719]

socool

[@zimouspero]

You never came again as you said

[@nigel_redmon]

I did post the follow-up video, kzbin.info/www/bejne/gXTccmmCoadmd9U. Regrettably, in that one I also said there would be a follow-up video to it, but it didn't happen because I never received some professionally recorded tracks that were promised. I do have another video planned on digital audio sampling (you can read the gist of it on earlevel.com), just been tied up with work for a while. It will happen!

[@zimouspero]

Nigel Redmon hello nigel. Ok, i understand. Your explanation it was very usefull, so i was waiting for more. Thank you for the video.

[@christopherjolly3380]

Is this why dj dither is called dither ?

[@2ice715]

can i Like Twice?! Thank you

[@BENX545]

and then I knew...

[@CurvedSlightly]

1 dude didn't understand.

[@1xgt]

10

[@chanceneck8072]

I still don't get it... Lol

[@nigel_redmon]

Watch it one more time :-D

[@nigel_redmon]

Short version: Limited resolution (sample size) results in a grid of possible values. The grid can possibly result in patterns in certain audio material that sound like the audio is distorted. By jiggling the grid a tiny bit, randomly, we don't get error patterns that line up with the audio signal-instead of a distorted signal, it sounds like the clean signal plus a tiny bit of hiss. Analogy: Close one eye and spread your fingers in front of your view, and read this. Your fingers will block some words. Jiggle your hand side to side, quickly, a small amount. Now you can read all the words, even though there is a little blur in front of them. The blur is better than the blocking.

[@ezrazski]

all dither is damage. all dithers sound different. it is not needed if you don't downsample in the first place. sell 24bit masters to consumers (we have the bandwidth for it now) and stop the bad audio. #savetheaudio

[@TangerineTux]

You mean quantize, not downsample. And 24 bits for the final master is a waste. 16 bits already give 96dB of dynamic range _without_ noise shaping. With noise shaping, it rises to approximately 120dB, which is more than enough. To quote xiph.org/~xiphmont/demo/neil-young.html: “120dB is greater than the difference between a mosquito somewhere in the same room and a jackhammer a foot away.... or the difference between a deserted 'soundproof' room and a sound loud enough to cause hearing damage in seconds. 16 bits is enough to store all we can hear, and will be enough forever.”

[@isidroco]

Great explanation!

[@nigel_redmon]

Thanks!