Educate me about Jitter

Discussion in 'Audio Hardware' started by Strat-Mangler, Feb 13, 2018.

  1. Strat-Mangler

    Strat-Mangler Forum Resident Thread Starter

    Location:
    Toronto
    I've seen this term thrown around quite a bit. My limited understanding of it is that it represents the deviation from perfect accuracy in reading the digital data which could theoretically (in this application) lead to some inaccuracies in sound reproduction. Is this correct?

    What would the difference be on a technical level between ripping a CD to FLAC and using a DAC versus using a CD player or transport/DAC combination? Any advantage/disadvantage to going one way versus another?

    Any real-world night & day difference which can be heard when going for one setup versus another?
     
  2. BIGGER Dave

    BIGGER Dave Forum Resident

  3. Strat-Mangler

    Strat-Mangler Forum Resident Thread Starter

    Location:
    Toronto
    I know it says "English" in the link you posted, but it's all Greek to me.
     
    SandAndGlass and scobb like this.
  4. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    Jitter is a digital timing error measured in Femtoseconds, which are one quadrillionth of a second. In one femtosecond, light travels just 300 nanometers — about the size of the biggest particle that can pass through a HEPA filter, and just slightly larger than the smallest bacteria.

    Because Jitter is quantifiable, it's used as a "more better" attribute in digital audio. Now, this part is important - it is possible for some digital products to be better then others. A lot more goes into DACs then the chips themselves. At this point, in 2018, when we hear better digital it is unlikely reduction in audible jitter is the actual cause, it's just used as something objective that can be pointed to. Note that term - "audible jitter".
     
    thxphotog, Robert C, dartira and 6 others like this.
  5. chervokas

    chervokas Forum Resident

    Jitter is problem with the timing of digital audio -- when the samples being taken during a digital recording, or being decoded during D to A conversion, are done at uneven intervals, the result is in imperfect representation of the original waveform (one might argue AD/DA conversion always delivers an imperfect representation of the original waveform, but this creates a non-linear distortion of a particular sort). As with many things digital, noise and distortion levels with jitter are so low, that there's always been a debate about the threshold for audibility of jitter, and now that, as @Rolltide mentions, we're talking about such low levels of jitter and such small units of time, the threshold for audible impacts are probably far below those of other sources of distortion in you playback chain.

    Anyway, here are two links, the first a little less technical than the second, which may be of help. You need to dig into the tech a little bit to understand what it is, but I think these are in clear English with good visual aides.

    What is jitter? - Apogee KnowledgeBase

    What Is Jitter in Audio? | Headfonics
     
  6. Strat-Mangler

    Strat-Mangler Forum Resident Thread Starter

    Location:
    Toronto
    Basically, manufacturers of CD players and DACs use jitter numbers as a way of enticing customers to buy their products, based on so-called better accuracy but without the real promise of actual superior audible results?
     
  7. Strat-Mangler

    Strat-Mangler Forum Resident Thread Starter

    Location:
    Toronto
    Assuming this is the case, is a setup type inherently superior in terms of sound quality to another between a transport/DAC, a PC feeding FLAC files through a DAC, and a CD player?

    Or is it basically a matter of preference and convenience?
     
  8. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    I think the issue is that because jitter is quantifiable, and because yes, a 2018 DAC chip does have jess jitter then a 2015 DAC chip even if its , its something tangible they can present. Its also true that some very cheap devices have audible jitter (I want to say first gen Aiport Expresses measured horrible).

    And its important to note that if Gear A claims to be better then Gear B because of jitter, it doesn't mean Gear A doesn't sound better then Gear B, it just means more then likely a marketing department was either being lazy or just telling customers what they're conditioned to hear. For instance, the Stereophile measurements of the Audioquest Jitterbug, a device explicitly said to reduce jitter, confirmed it has zero impact on jitter one way or the other.
     
  9. Strat-Mangler

    Strat-Mangler Forum Resident Thread Starter

    Location:
    Toronto
    OK, pretty much what I thought. Numbers basically used purely for marketing purposes since our ears can't hear the difference anyway.

    Thanks for clearing that up. Glad it was that simple. :)
     
    basie-fan and showtaper like this.
  10. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    This is where it gets interesting. All audiophile purists on this forum still insist that CDs are superior to all methods of digital playback that don't involve mechanically spinning a plastic disc and reading it with a laser beam. The culprits that are said to spoil non-CD playback are jitter and electrical noise, both of which are totally quantifiable.

    I can say there was a time when computers and FLAC existed that I still preferred to play CDs. I could make the argument of a CD over airplay to an Apple TV. But today, when I can have a Raspberry Pi with an I2S connection to high grade clocks and SPDIF output, I can't make the argument anymore.
     
  11. tiller

    tiller Forum Resident

    Location:
    Montreal
  12. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    Then you get to the last page and notice he's selling you a better mousetrap. I wouldn't quite put this on part with getting an explanation on cables from Audioquest, but reading about jitter from the audiophile jitter reduction industry isn't the best way to actually educate ones self about jitter IMO.
     
    Robert C, sunspot42, jerico and 4 others like this.
  13. tiller

    tiller Forum Resident

    Location:
    Montreal
    Just saying the issue is not as black and white as it's being made out to be. Lavry is a polarizing guy for sure.
     
    Grant likes this.
  14. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    It may or may not be. The jitter landscape was likely much different when he wrote that sales pitch 20 years ago, but the biggest problem is that its data presented in terms of a sales pitch. He's describing a big problem that requires his products to correct. This doesn't automatically discredit him, but huge red flags appear. I'm comfortable saying that in literally any application, one should never educate themselves on a problem from a person selling a solution to said problem. Especially not in audio, where solutions tend to be invented first and problems second.

    Some specifics as to Larry's write-up -

    - He's using fake jitter, intentionally injecting it so he can talk about it. We don't get to see a real world scenario

    - this paragraph - "There are a number of articles written about "Jitter bandwidth". I have seen claims to the effect that jitter above Nyquist does not matter. Others claim that jitter frequencies above sampling causes no harm. Such claims may be true in some context (which I am still unaware of at the time of writing this article). In general, one has to be cautious about ignoring jitter at any frequency."

    I read this as being "there are folks who have quantified this. I'm not going to attempt to prove them wrong, only to declare I don't trust them".

    So in other words, I'm wondering if the water here isn't muddy because Larry muddied it himself.
     
    Robert C, sunspot42, Agitater and 4 others like this.
  15. oxenholme

    oxenholme High Quality Posts™ a speciality

    I use WASAPI PUSH (buffered) to play PCM files through DAC / power amplifier / speakers and I find the end result easily the best digital that I've ever heard. It presumably interpolates or whatever it does exceptionally well as 16 bit 44.1 sounds every bit as clearly defined as 24 / 192. It demonstrates admirably that there is a lot more to it than being able to achieve 20 KHz or higher.

    It intrigues me that people don't seem to attach much importance to either the mechanism for getting the digital file onto the computer or the actual computer hardware for getting the digital signal on its way to the DAC. Is it really 100% reliable?
     
    johnny q likes this.
  16. Ski Bum

    Ski Bum Happy Audiophile

    Location:
    Long Island, NY
    If you think that the effects of jitter are not audible, I suggest you try to arrange the following audition:

    Play your favorite CD or PCM digital file on each of the following:
    1. A dCS Vivaldi DAC without the dCS Vivaldi clock.
    2. A dCS Vivaldi DAC with a dCS Vivaldi clock.
    3. A dCS Vivaldi DAC with a dCS Vivaldi clock and with devices to deal with ringing and vibration isolation under each of them.

    2 is simply 1 with a somewhat better device at reducing jitter. Admittedly, 3 is not really resolving a jitter issue, but many DACs ring.

    Let us know whether you hear a difference.
     
    PhantomStranger likes this.
  17. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    Is reading a disc spinning at precise high speeds with a laser beam 100% reliable? It seems intuitive to me that removing mechanical aspects of digital playback make it more, not less, reliable. That's the part that confuses me - all of what you say makes sense until we get to the part where CDs are the baseline and solid state is scary and we should be worried.
     
    Robert C, Kiko1974, sunspot42 and 4 others like this.
  18. oxenholme

    oxenholme High Quality Posts™ a speciality

    What are the symptoms of a DAC that rings? Are you suggesting that we need a half inflated inner tube under the DAC or oak cones or something?

    I am not interested in forking out for a cryogenically treated 24 carat oxygen free gold USB lead if coat hanger wire is entirely adequate, but I am interested in eliminating any factor(s) that prevent(s) 100% of the ones and zeros (a) reaching the FLAC or WAV file on the hard drive and (b) leaving said FLAC or WAV file and reaching the DAC.

    What makes the output of one DAC sound different to the output from another? The sound wave form has been defined rigidly within the ones and zeros - surely the only correct analogue output corresponds exactly with same?
     
    Robert C and H8SLKC like this.
  19. Claude Benshaul

    Claude Benshaul Forum Resident

    Jitter is what was used 15 years ago to entice us to buy a product or scare us from buying another. It's one of these weird problems that only manifest with using HiFi audio equipment and are strangely absent in every other field or discipline, even when using the same components.

    Jitter problems in HiFi are usually solved by injecting money into the audio stream. These injections however are only a temporary remediation and may require repeated treatment through the lifetime of the user.
     
  20. oxenholme

    oxenholme High Quality Posts™ a speciality

    15 years ago? Interesting. So they didn't introduce it until sort of twenty years after the birth of CDs?
     
  21. Helom

    Helom Forum Resident

    Location:
    U.S.
    [​IMG]
     
    sunspot42, basie-fan, H8SLKC and 3 others like this.
  22. Grant

    Grant A Musical Free-Spirit

    Location:
    Arizona
    I hear jitter as a glassy or unfocused sound.
     
    jfeldt likes this.
  23. Claude Benshaul

    Claude Benshaul Forum Resident

    15 years ago was the time that digital audio became more common with the growth of the home theater market.

    It was then that using S/PDIF to connect components became standard and replaced the old Red/White RCA that many owners of CD and DVD players used. Evidently, it was also the time when discussions about jitter first appeared on AV forums, with about the same level of conviction from both sides as exhibited now in discussions about the effects of temporal blurr.
     
  24. House de Kris

    House de Kris Forum Resident

    Location:
    Texas
    Yes, numbers are often time used for marketing, it's how information is conveyed. Gives the illusion of easy comparison. But, you may believe that marketing, in and of itself, is sinister. In this case, jitter specs, power output of an amplifier, capacitive loading of a MM cartridge, and word width and sample rate of FLAC files, are all just marketing numbers and mean nothing to the end user. Simply a way of taking your hard earned cash away from you. In this way of thinking, the Kenner Close-n-Play had it right. "Close it, it PLAYS. Open it, it STOPS!" Oh, to have the simple life again.
     
  25. House de Kris

    House de Kris Forum Resident

    Location:
    Texas
    Jitter is battled in MANY other fields and disciplines. Has been for a long long time. I first ran into the equations to understand the effects of jitter back in the 80s. Jitter is not some made up catch phrase invented by audio companies to deprive you of your money.
     

Share This Page