1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

R2R/multibit vs Delta-Sigma - Is There A Measurable Scientific Difference That's Audible

Discussion in 'Sound Science' started by goodyfresh, Aug 31, 2015.
72 73 74 75 76 77 78 79 80 81
  1. gregorio
    1. Well that's your problem. Although to be fair, it's a common problem with many audiophiles: Just because they don't know some fact they falsely assume that we (science/humanity) also doesn't know it, but of course that's a fallacy, it's actually just a problem of their own ignorance (or worse, they're not ignorant of the facts but just pretend to be in order to sell some snake oil belief or product)!

    2. Nope, I'm not mis-remembering, I've never heard of the unit to which you're referring. As far as I'm aware, there was very little call for sample rate conversion in the early 1990's and therefore, I don't see how a unit that did that could be "popular"? Additionally of course, what they claimed in their marketing material and how proud they were of it, is irrelevant. Again, I bought a pro ADC in 1992 which had a jitter spec more that 1,000 times better than that, but I don't recall how proud they must have been about it!
    2a. As far as I recall, EVERY pro audio digital hardware unit I've ever bought (or seen specs of) always specified jitter. Of course these days there aren't many digital hardware units, with the obvious exception of ADCs/DACs but even going back to the 1990's, I don't recall a specification worse than about 100ps and by the early 2000's, even quite modestly priced pro/prosumer ADCs had jitter specs of around 50ps or so. Again though, in a pro studio setting, very low jitter could be an important factor, as jitter could accumulate due to clock signal distribution between numerous digital units and/or several round trips through ADCs/DACs (which was very common).
    2b. Agreed, the jitter of the clock is largely irrelevant, it's the jitter of the clock signal at the DAC chip's input that's relevant but this figure is pretty much never given.
    2c. And here we come to the crux of it all! Firstly, two orders of magnitude worse than 1 pico-sec is 100 pico-secs, which is still around 2,000 times below audibility when listening to music. Please explain how that's "an issue"! Secondly, you're very vague about what amount of jitter we actually end-up with at a DAC chip's input, I'd like to see evidence where the amount of jitter reaches even a magnitude below audibility.

    3. Ah, so you admit there are some "stone tablets" then!
    3a. Of course "not everyone agrees", not everyone even agrees that the Earth isn't flat!
    1. Hallelujah brother!
    1a. No they're not, the OP asked about "measurable scientific differences" and few hundred femto-secs worth of jitter is measurable.
    1b. True, in the context of what we end-up with "that's audible".
    1c. We don't have to assume anything, we can just measure the resultant distortion!

    2. A "very important point" to whom, snake oil salesmen? "In the context of our discussion" (measurable differences that are audible) it's NOT "very important", in fact the opposite, it's of less than no importance whatsoever because jitter artefacts even in cheap DACs isn't just inaudible but typically at least one (and not uncommonly two or so) orders of magnitude below audibility!

    Although padded out with some actual facts/truths, it's just the same old, tried and trusted, audiophile marketing BS again. "Very important points" and "real issues" which as far as audibility is concerned are neither of any importance nor "issues", but if we (falsely) state they are, then we can maybe sell them some snake oil product which allegedly addresses and fixes those "very important points" and "real issues"!

    Last edited: Aug 30, 2019
  2. KeithEmo
    (I may be somewhat off on the date there...)
    I believe we were also talking about hardware sample rate converters and NOT ADCs in that conversation.....
    The only reason I looked at them was that, around that time, I had a "pro mini-disc recorder" and a "pro CD recorder" - both really just expensive home units.
    The digital output on the mini-disc recorder ran at only a 48k sample rate while the CD recorder only accepted 44k.
    (I believe I ended up replacing the CD recorder with one that accepted both rates.)

    The only "consumer level sample rate converter" I see lately is the Behringer SRC2496 - which is apparently a current product.

    And, if you look in its spec sheet, you will find the following....
    Jitter permitted at input >40 ns
    Internal jitter at input <2 ns
    Internal jitter with external <20 ns, 10 ns typ.

    That is a cut-and-paste... they are actually specifying NANOSECONDS of jitter....
    (And that IS a spec that would be pretty bad on a $250 DAC...)

    And, yes, I am also trolling BigShot a bit...

    After all, whenever anybody ELSE makes a claim about something being obvious, he's usually the first to insist that OTHER PEOPLE's experiences are "just anecdotal data"....
    (Usually followed by a suggestion that everyone with any common sense would disregard any claims that aren't backed up by actual test results.)
    Therefore, it only seemed fair to expect him to follow the same "rules" he seems to expect everyone else to follow when making "claims"....
    Yet, apparently, we're expected to simply accept that "jitter is obviously a meaningless specification.)

    I am not personally aware of any well established "threshold of audibility" for jitter.....
    And, since jitter comes in a wide varieties of frequency spectra, waveforms, and levels of data correlation, it seems likely that any such spec would be quite detailed and complex.
    (I would expect the threshold of audibility for 0.1 Hz random jitter to be very different than the threashold of audibility for 440 Hz data-correlated sine wave jitter.)

    Of course, with a DAC, you also need to consider internal jitter and susceptibility to external jitter present on the input separately.
    Some DACs may have very low levels of self-jitter, yet be very sensitive to jitter present on their input signal, while others may be the opposite, and they are rarely if ever specified individually.
    (Note that the popularly quoted "J-Test" tests neither - it merely shows the amount of jitter-related distortion present at the output when you input certain specific test signals.)

    As for "we can just measure the distortion"..... there I would agree entirely.
    However, if you actually look at the spectrum of the distortion on the output of a particular device, you'll find that no two are exactly alike, and it varies depending on the conditions present at the input.
    Therefore, if you want to have useful information, you really do still need to specify your test conditions.

    And, no, I don't think there are any stone tablets....
    Although, since some people seem to believe that they do exist, I figured it was only fair to offer the opportunity for them to present those tablets for verification... :)

  3. bigshot
    I think it's deliberate muddying of the waters. Sometimes Sound Science is filled with Sound Sales Pitch. The only people that fall for this stuff are people who don't want to think for themselves, and they're a lost cause anyway. The Sound Science regulars and lurkers know the score.
    sonitus mirus likes this.
  4. Steve999
    I am thinking there might be some kind of stone tablets, although of course you can’t believe everything you read on the Internet.

    Here is some purported stone tablet where the ancient Babylonians were supposedly (if you know what I mean) doing trigonometry 3,700 years ago better than we can: :rolling_eyes:

    And here is something referred to as a “stone tablet” called the Rosetta Stone, although who knows, maybe it was just a publicity stunt by some software company a couple of thousand years ago. :deadhorse:

    I mean, if you want to believe in that kind of stuff.:unamused:

    IMHO, FWIW, YMMV, & etc. :thinking:
    Last edited: Sep 2, 2019
  5. gregorio
    1. Yes they are but there's two EXTREMELY OBVIOUS points here: A. That's both internal and external jitter and B. Even 20ns of jitter is still an order of magnitude below what has ever (reliably) be demonstrated to be audible with music recordings! Duh!

    2. Glad you're admitting it but not only bigshot of course, the whole of this sub-forum!

    3. No, we're expected to simply accept that jitter is obviously a meaningless specification not just because Bigshot says so but because ALL the reliable evidence says so, including the objective measurements of it!!

    4. Again, another circular argument. This isn't the "What KeithEmo is personally aware of" forum (or the "KeithEmo self contradiction" forum, as you've already admitted that "Obviously there is some level below which jitter has no audible effect").

    5. How can you possibly expect that? If ALL the jitter artefacts peak no higher than say -130dB then whether one of those peaks at 0.1Hz or 440Hz makes no difference whatsoever to the threshold of audibility because both are long way below it. (Except apparently on an office system that doesn't exist, playing a music recording without any music on it)!

  6. KeithEmo
    Yes, by today's standards, as applied to most other equipment, both are NOT especially good.
    More to the point, the "internal" measurement is the relevant one since it sets the minimum which you can ever expect that piece of gear to deliver.
    (More to the point, since you can only conduct valid tests using test samples whose quality is significantly better than what you're measuring, samples that had been through a
    piece of equipment whose quality was equivalent to that one would be useless for testing whether the lower levels present on another piece of gear are audible or not.)

    I only ever recall seeing a single test published about the audibility of jitter...
    And it was done a long time ago, using source content and test equipment that was itself somewhat dubious, and with no explanation of where and how they secured test samples with extremely low jitter to use....
    (Obviously, if you want to test whether adding 10 NS of jitter is audible or not, you must first start with test samples whose level of jitter can be demonstrated to be far lower...)
    Could you please present us all with some of that other "reliable evidence" substantiating that claim...

    There are obviously levels below which any type of noise or distortion is inaudible...
    However, since I seem to have missed the stone tablets containing that exact information for various types of jitter, I was thinking it might be nice to do a few tests...

    I would be inclined to agree that, if the total quantity of jitter artifacts sums to a level below -130 dB, then it will probably be inaudible...
    And, if the level of any individual sideband falls below 100 dB or so, then it too will probably be inaudible...
    Note how, since I haven't personally tested either, or seen any test results I would consider authoritative, I stated it as "my opinion" and "probably"...

  7. bigshot
    My opinion is based on doing the research and tracking down the tests and figuring it out for myself. It isn't up to anyone else to "present you" with anything. We've done that many times in the past on a variety of subjects and every time we did, you blew right past it and went back to your salesman double talk. If you really want to have an informed opinion, take Gregorio's post as a clue that perhaps you don't know as much as you think you do and go do your own homework. There's no reason to have a conversation with someone who won't acknowledge information when it's "presented" to them.
    Last edited: Sep 3, 2019
  8. KeithEmo
    Yes... we seem to always end up in the same place.

    When someone like myself makes a claim we are told "that's just anecdotal evidence" or "it's just a sales claim"...
    And you routinely follow that up by suggesting that, unless we provide documented proof to support our claim, then we shouldn't expect it to be taken seriously...
    However, when you make a claim, we're expected to accept that it has been fully documented and proven conclusively to be true...
    I should also point out that, while I have quite a bit of respect for Gregorio, I do NOT necessarily assume that he is necessarily always correct on technical matters either.
    (And I would not expect anyone to accept what I say as gospel truth either.... as we are ALL occasionally wrong. However, in general, I at least try to quality my proclamations.)

    As I said the first time around....
    I can't recall seeing published results of any tests on the audibility of various types and amounts of jitter....
    Conducted using test samples produced on equipment with known low levels of jitter....
    And with other equipment, also with low enough documented levels of jitter, so as not to obscure any possible results.
    If you actually have any such reports to mention - now would be a good time.

    Otherwise, we will consider your opinion, and the anecdotal evidence you seem to be saying you have, in their proper perspective.

  9. bigshot
    I'm more than happy to help research things for people who appreciate the help. But you have another kind of agenda. Every time I go fetch a stick for you, you change the subject and go back to the same old sloppy logic again. Sorry, but it just isn't worth my time. Feel free to look it up yourself. It's all at your google search.
    Last edited: Sep 3, 2019
  10. castleofargh Contributor
    I fully agree with this! 721%
    double standards are bad and all too common in this section depending on who's posting. it should not be so.

    consider what you'd think of somebody else making a statement, and when asked for supporting evidence, replied what you just posted. you've put people on ignore(and I don't blame you) for less.
    Keith just likes to explore possibilities, remote ones even more than big stuff, probably because for the most part in audio, we already know what there is to know about the big stuff. he never claimed that they were typical issues, never claimed that jitter had big audible impacts and that we needed super saiyan femto clocks. even his made up examples for audibility with THD and such are arguably no about music listening but involve test signals. so not much reason to declare WW3 just yet IMO.
    you on the other hand claim that jitter will not be audible. I tend to agree, but it remains that the burden of proof is on you. so you should be in charge of the googling and copy pasting as deciding by the knights of the round table, or science, or the internet of things... I'm not completely sure which one.

    yes jitter in modern products is expected to have no audible impact because of how small it typically should be. most DAC designers agree on that, even several MOT selling DACs here with more or less marketing free-styling habits, have said something along those lines about jitter in properly designed gears. and yes there is at least one paper pretty opinionated about jitter in consumer electronic, and several sound samples shared online over the years to try and get a feeling of what jitter(at least some form of jitter) might sound like and to check for ourselves what levels we can consider below hearing. and yes opening my window even in the dead of the night will affect the music I'm hearing in measurable ways probably magnitudes above what most stuff discussed in this topic ever will. I get that the constant distraction from actual audible issues in this hobby can be frustrating. but on the other hand, it's a forum, if someone is curious and wishes to discuss some oh so unlikely stuff possibly having oh so unlikely audible impact in music, that's his right. to quote a legendary philosopher, "it's the Sound Science forum, not the ________* forum". the important part being that it's a forum. a public one at that.

    * replace with the name of the guy annoying you.
  11. bigshot
    There's history here. I went out of my way many times to provide him with tests to take and backup to my comments, and he blew right past them as if they didn't exist. He launched into abstract theories that he thought hinted that documented evidence "might be wrong". I've seen enough UFO and bigfoot documentaries to know what that is. I've done my part and it was answered by complete blather. Is it a surprise I don't feel the need to provide anything now? I don't see a point in entertaining abstract fantasies of thought. If I wanted that, I would go to the head shop instead of Head-Fi. If it doesn't have any relationship to a person's stereo system, it seems irrelevant to me. And someone who says I'm not "scientific enough" just makes me want to blow them off completely. We've seen enough wastes of time taking that approach. He's parked on Sound Science to put on a self-serving show. And his example led to others doing the same thing. My patience for it has worn down to nothing. Gregorio is an energizer bunny and keeps answering and they blow by him too. He's a great resource being wasted on replying over and over to the same kind of grandstanding.

    A useful forum is being dominated by circular reasoning and pointless "what ifs". That isn't science. That's a waste.
    Last edited: Sep 3, 2019
72 73 74 75 76 77 78 79 80 81

Share This Page