Jitter is RUINING your audio? The TRUTH about jitter!

listen to [Music] this that was Heavy Jitter on the music sounds bad right so Jitter should be taken very seriously when buying a deck audio interface or something similar right well actually no personally I don't care about Jitter anymore and here's why you shouldn't either hey Julian CR here and in this video we're going to have a look at Jitter what is it why can it negatively impact your sound quality why do audio files get scared when they hear this word why companies try to trick you into blowing this topic out of proportion what it really sounds like and why you can confidently ignore this whole topic hopefully after watching this video first of all what is Jitter Jitter is the variation in the time of an event such as a regular clock signal from nominal Julian dun imagine a metronome happily clicking away at a constant rate in theory the interval between each click is exactly the the same but in practice there are always some inaccuracies which lead to some clicks being a bit earlier than others and others are a bit more delayed that's Jitter essentially an inconsistency in clocking why is that an issue for digital audio well you might know that in digital audio the waveform is made up of many individual samples and in order to reproduce the audio Faithfully each sample needs to be played back at the correct time otherwise the original waveform will be distorted which can lead to AR effects like sidebands and noise like Distortion we will also have a listen to a few more audio samples of Jitter later but before that I directly want to address a misconception and that that the sound of Jitter is often compared to wow or flutter for my younger generation viewers here these are artifacts that you can come across when playing back vinyl records and these issues are often coming from offc Center rotation or warped records this results in audio that is fluctuating in speed and pitch and this is often used to describe what Jitter sounds like but in reality this is not really how Jitter behaves in digital systems the clocks in digital audio systems are very fast and the fluctuations do not result in a varying of speed or pitch in the audio jit introduces small amounts of timing inaccuracies for audio samples and this creates Distortion that is not harmonically related to the audio signal and you could actually get a taste of that at the beginning of this video now the big question is where does Jitter come from in the first place there are multiple areas where Jitter can occur for example in an audio device's internal clock signal degradation and cables and it can even be due to the data itself that's transmitted I don't want to go into detail on all of them otherwise we will be here until the cows come home but I want to highlight one difference in connections that can have a significant impact on Jitter I'm talking about synchronous and asynchronous connections big words I know but in a very basic sense you can think of them like this let's say you have a musician and they should play some music you can give them the complete sheet of music at once and then they can play back the music at their own speed that would be an asynchronous connection between you and the musician here your timing is not really relevant for the music as the musician goes at their own pace once you have handed over the notes or you could give the musician the song note by note which they play back the instant you give them the note but then of course your timing has to be perfect otherwise the song will sound horrible this would be the case for a synchronous Connection in asynchronous connections like USB and ethernet and yes USB can also be used synchronously but that's besides the point Jitter is much less of a concern when the data is in transport as there is no audio clocking signal which has to be preserved the audio simply sent in small data packages which are then put together in the receiving device and clocked with the device's internal clock this is also the reason why USB and network connections do not introduce any Jitter into the audio signal and in my opinion things like USB relockers and audio file Network switches are complete snake oil as they are addressing a non-existent issue I would be very suspicious of any of these devices and I have yet to see any scenario in which they make an audible Improvement you have been warned in synchronous data transmission like spff and as3 clocking becomes much more interesting because here the audio is sent as a continuous stream of data and clocking information so here if the sending device has Jitter issues the receiving device could have a problem with that too I know these days there are things in place like face lock Loops which can mitigate these problems but I don't want to go down that rabbit hole Additionally the signal can be negatively impacted by the cable which can introduce jit as well so with spit if or as3 connections there are simply more possibilities for jit to occur than with us or network connections by no means does that mean that synchronous connections are bad just that they have the potential to be more negatively impacted by Jitter as with many things in audio Jitter is not an onoff thing like whether you hear it or not it's more gradual and you might ask yourself the question now how strong does Jitter have to become in order to be Audible for that let's have a listen to a few audio samples as mentioned Jitter can have random components so let's listen to an increasing amount of random Jetta [Music] first you could hear that a random Jitter at low levels gives the recording a gritty underlying tone and the stronger the Jitter becomes the more it sounds like the distorted noise that is modulated with a signal the interesting thing is about a random jet a is that unlike noise it is not always there but only when a signal is present and then it adds this nasty noise-like artifact to the sound if we just play back a single tone and look at it with an fft you can see how this Jitter would look like you can see the 12 khz tone here presented as a single line with an increasing amount of Jitter Jitter can also be periodic which is where it introduces addition tones above and below the original frequency have a listen to how this sounds [Music] like [Music] you might have noticed that with a higher Jitter frequency you could detect the Jitter earlier this has to do with masking because when we have a look at an fft again you can see that with low frequency Jitter the created sidebands are quite close to the original tone and they have to become relatively strong in order to be detectable higher Jitter frequencies create side bands further away from the originating frequency and are thus easier to detect this is also the finding of Julian Dunn which shows the audibility of Jitter and you can see that with an increasing Jitter frequency the amount of Jitter needs to be lower in order to be not perceptible in these tests you might have been surprised by how severe really Jitter has to become before becoming audible of course this is YouTube and the audio here is compressed so I think under perfect conditions you can detect Jitter quite a bit earlier than in the test here but still this goes to show that with complex signals like music Jitter can sometimes be not so easy to detect still you could hear that random Jitter creates noise-like artifacts that are modulated with the audio and that give the music a grungy underlying tone periodic Jitter on the other hand creates Distortion side bands that at low frequencies can give the audio a garbled sound and at higher frequencies tend to result in a metallic buzzing type of distortion okay okay Jitter is bad I get it but didn't you say I don't have to worry about it after all this I think I care even more well let's have a look at a jit in practice remember when I said that Jitter introduces sidebands and noise-like Distortion this is very easily tested and visible in an F50 when playing back a single tone you essentially want to see a low noise floor and minimal sidebands something like this or this and as you can see I have also put the audibility threshold defined by Julian done on these graphs and you essentially don't want to see any signal above the orange line with the exception of the testone in the middle let me show you a few more devices to demonstrate how perfect audio devices are these days in terms of Jetta this interface is perfect this device is perfect this one is perfect as well this stack is totally fine this one is perfect again perfect performance technically not 100% perfect but totally fine really great performance this one is as perfect as it gets and this one is uh wait is my argument falling apart well let's be realistic here while technically speaking you can see quite some components above the threshold but these tones are still 100 DB below the test signal and in my opinion there is hardly any way you will ever hear that especially when you keep in mind that these are one of the least expensive devices which also have some other issues like increased noise and Distortion which are of an order of magnitude greater than the issues introduced by Jitter here so while this definitely does not look too great I still don't think that Jitter is an audible issue here if we have a look at the whole package but you might still not be convinced so let's step it up a notch as mentioned spidi has the potential to be more prone to Jitter so here are a few more devices this time I'm also using a specific signal called J test which is designed to be a worst case scenario again you can see that all these devices shown have a near perfect Jitter performance with sidebands just at or below the threshold affordability now one last example to drive home the point that I want to make this incredibly expensive $8 Deck shows hardly any signs of audible Jitter this VAR is tone here can be confidently ignored and I can say that because in other tests the deck has a high noise flaw and it distorts like crazy at high output levels so this deck definitely has some issues but Jitter is not one of them again this is an $8 deck that does not have any audible problems with Jitter under worst case conditions I rest my case now is Jitter never an issue no there might still be the one other device out there that is so badly designed it's still has some issues but with halfway decently designed devices my experience is that a Jitter is simply not the issue that it is made out to be yes Jitter definitely was a problem in the past but it has been solved with technology a long time ago and pretty much any device out there these days has perfect Jetta performance I also find it kind of sad that companies are still trying to cash in on the fear of Jitter with audio file Network switches USB relockers and ftoc reference clock BL for interfaces this stuff is just not needed anymore as I've demonstrated in this video Jitter is a solved issue period yeah that's why I don't care about Jitter anymore and you shouldn't either it just makes life easier all right if you found this interesting I would appreciate your thumbs up on this video subscribe for more and I will see you all in the next one

Channel: Julian Krause

Views: 42,511

Keywords:

Id: f_r33B4jptA

Length: 12min 31sec (751 seconds)

Published: Mon Jan 29 2024