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

Testing audiophile claims and myths

Discussion in 'Sound Science' started by prog rock man, May 3, 2010.
First
 
Back
763 764 765 766 767 768 769 770 771 772
774 775 776 777 778 779 780 781 782 783
Next
 
Last
  1. castleofargh Contributor
    do you imagine being completely aware of all the data coming from our senses at all time? it would be... Hermagerd staaahhppppp!!!!!!
    and then to add some fun, having to actually remember all the data+interpretation instead of how close something is from already vaguely memorized patterns. again in quasi real time because new data keeps coming. that would be crazy and most likely not viable. instead we have perfect creatures like me, who can hardly follow a conversation if there's a TV turned on in the room, and can't remember what someone requested this morning on the phone.
    flawless!
     
  2. james444 Contributor
    Change of subject. [​IMG]


    I'm putting this up for discussion, because I honestly have no particular idea what to make of it. Also, I don't know enough about DACs to even make an educated guess. So, here's the story...

    There's this ES9218P DAC + AMP chip, which seems to be pretty popular with "audiophile" portable gear manufacturers. I have it in my LG V30 smartphone and in a tiny DAP/USB-DAC, the Shanling M0. As a peculiarity of this chip, it features a user-programmable FIR filter with different presets. Some of these presets are exposed via user interface, my LG V30 offers three different settings and the Shanling M0 eight. The general consensus among users both on Head-Fi and on other forums seems to be that these filter settings do indeed sound different from each other.

    So I decided to measure and compare four of them...

    I used the Shanling M0 as USB-DAC, and since I don't have gear to measure DACs/AMPs directly, I connected my IEMs and measured the entire chain with my miniDSP EARS. I did not remove the earpiece from the artificial ear between measurements, nor did I change anything else except the FIR filter setting.

    Here's a frequency response comparison of 4 filter presets (linear fast, linear slow, minimum fast, minimum slow):

    [​IMG]

    The same graph zoomed in:
    [​IMG]

    And here's an animated comparison of impulse responses (in between spoiler tags, to avoid annoyance):
    [​IMG]

    Subjectively, I'd say these presets do sound slightly different to my ears (in a sighted test). Objectively, I'd think the changes in FR are too small to make an audible difference. Last but not least, I don't know what to make of impulse responses... There seems to be quite some variance, however from what I've read, FR and IR should be just two different visualizations of the same data.

    What's your opinion on that? :)
     
    Last edited: Dec 16, 2018
  3. castleofargh Contributor
    not sure I'd try to answer this question that way. it would probably be more interesting to record a short passage of music with various filters, edit the hell out of them in audacity or whatever to try and have them start at the same time. then just ABX the various tracks in foobar.
    I'd suggest to record straight out of the DAP, even if it's with the computer's soundcard, to limit the potential variations outside of changing the filters(ambient noise in the room with the EARS, or tiny change in position as the tips maybe flex a little). and also because the EARS is fixed to 48khz I think. so while that may leave enough room to show differences on 16/44, it will still band limit however it wants for 48kHz/s.so I'd rather record at 96khz(or more) to make sure we're getting FR roll off profile from the DAP.

    about impulses in REW, I would suggest to set to %FS in the top left corner of the graph, and then set the limits on the resizing setting (somewhere top right above the graph) to maybe -0.001s and 0.001s at least as a start to find the impulse ^_^. and you can normalize the vertical axis in the controls if you want but then you might not notice a change in amplitude of the main impulse, if any. you can also just zoom while holding the MOUSE3(wheel click) down, but on the impulse graph it's usually a nightmare.
    anyway that should give you something looking a lot more like the impulses on innerfidelity and IMO they're more relevant to you than the noise in your room :wink:

    about impulse and FR, the FR is in the impulse, but not every information in the impulse is in the FR(like reverb).
     
    james444 likes this.
  4. james444 Contributor
    Thanks, I've changed the GIF in my post to a more meaningful presentation.:wink: Only bothered with IR because these filter settings are said to change impulse characteristics,

    Don't think I own equipment up to this task, just an old notebook and no external soundcard, I'm afraid. I thought, if anything, variations outside of changing the filters would excacerbate variations on the EARS, but since resulting FRs look near-identical, I felt the measurement must have been pretty consistent.

    Now that's interesting. Is it possible that two different filter settings might both conform to DAC specs [20HZ~20KHz (-0.5dB)], look near-identical in the FR-graph, yet make an audible difference in impulse response?
     
  5. KeithEmo
    I've finally figured it out... we seem to have some sort of language barrier.
    I have been talkng abut what's POSSIBLE... in terms of science... and not what already exists.
    You are talking about "what's available today in a typical high end audio recording studio".
    (interspersed with your opinion about what we actually need to produce a commercial recording.)
    So, in fact, we don't disagree; we simply aren't talking about the same thing at all.)

    If you look up "possible" in the dictionary you will find that "someone already did it" is NOT a requirement.
    In fact, you will also find that "you can currently purchase equipment that can do it" is also not a requirement.
    In fact, the definition is quite specific that it even applies to "whatever can be conceived or is within the limits of ability".

    https://www.merriam-webster.com/dictionary/possible

    However, in this case, it has in fact been done.
    The scientists who ran those tests already provided the specific example you keep asking for.

    I have no idea if any existing audio track on a commercial recording meets those requirements. However, since I simply said that it's POSSIBLE, I have no obligation to prove that someone has already done it. Several scientists have obviously produced a track that meets that requirement - because they used it to obtain their results. And you seem to agree with me that it would not be even difficult to "record a 1 kHz tone in two tracks and pan it hard left". I could certainly do it in about five minutes in Audition. Therefore, as far as I can tell, you AGREE with me that it's possible. So how can you argue the exact opposite? If it makes you happy, I can record a track with that test tone on it, name it "Keith's strange music", and e-mail it to you and two other friends, or post it on this site. We will then have a distributed track with that requirement. And, if it really makes you happy, I'll get a friend of mine to pay me $1 for a copy, then we'll have a commercially distributed track that does. (And, in the mean time, if you want to claim that "no existing commercial track has it", then feel free to provide a signed affidavit that you've screened all million existing tracks, and can actually assert that none of them contain it.) Alternately, I'll be perfectly happy to stipulate that it very rarely comes up in commercial recordings and that, if it does, odds are nobody would be likely to notice that tiny discrepancy anyway. (However, just to be fair, if there something that I know my system isn't doing right, and which I know could cause an audible eror, I'd prefer to fix it.)

    Your second question is also simple to answer....
    You either invent it yourself or pay someone to invent it for you.
    In science we are often required to invent new technology to meet new requirements.

    1) You decide that your current microphones are inadequate and you decide what you actually need.
    2) You carefully write down those specifications as a technical requirement.
    3) You pay someone to design and build you a new microphone that meets your requirements.
    4) After they deliver it you use it to record your band.

    The current level of technology is simply the result of a balance between what's available and what's wanted or needed. Your recordings are limited by the quality of your equipment; the quality of your equipment is limited by the requirements you set for it; those requirements are set by the requirements of your customers; and the requirements of your customers are set largely by what they know is available. Technology evolves... driven by need. It is very rare to find a case where the technology is actually limited by the laws of physics in any practical way.

    The technical limitations of the vast majority of equipment in your studio is NOT set by "the limits of physics". In most cases it is simply "the level of performance that someone was willing to pay to achieve - so far".

     
  6. KeithEmo
    I've read estimates that we visually actually process about 5% of the available data... and even that needs to be qualified because we actually often "pre-screen" the data we acquire. For example, when you walk down a sidewalk, you don't actually "see" much of the sidewalk. Your brain, at various levels, screens out most of the input, and "only shows you the interesting parts". You notice the rock that's sticking up, and the hole you might trip in, but you do not see the millions of "ordinary" rocks and bits of cement.

    An important thing to realize, and one which may well be relevant to how we process audio, is that this mechanism operates at many VERY different levels. A simplistic view might be that "your eyes continually take in a picture of the sidewalk in front of you, which your brain then processes, and discards all but the information it considers important". However, the reality is far more complex. For example, your eyes actually have a rather limited visible field, so they can only focus on a small area at any given time, and the mechanisms that control eye movement also have limitations, and different areas and types of data are processed differently. (Your peripheral vision responds to motion, but not color, while your central vision responds better to detail and color.)

    So what really happens is something like this. Let's say you see a movement out of the corner of your eye. The sensors and parts of your brain dedicated to detecting motion do so relatively quickly, but without much detail. So, when that motion is detected, your brain sends an instruction to the muscles in your eye to start turning the central more-detailed area of your vision towards it. However, while your eye is still moving, other areas of your brain continue to process that original input. As they complete their slower but more thorough processing, they may reinforce the command to "focus on the interesting area", or they may countermand it after "deciding it wasn't interesting after all". This process occurs in a varety of different areas in your brain - with more complex processing occurring more slowly and simpler processing more quickly... and these processes tie in to other areas of your brain.

    So, for example, your brain notices "motion out of the corner of your eye" relatively quickly, and starts your eye turning in that direction. At the same time it may turn up your adrenaline to enable your muscles to respond more quickly if necessary. While this is all going on, the more complex processing necessary to decide whether that motion was a bird or a lion continues, as does the even more complex processing to decide whether it's an edible bird, or a hungry lion, and the yet more complex and slower processing to decide whether, if it is a hungry lion, you should try to run away, stand and fight, or hold very still and hope he doesn't notice you. And, at any point in the process, the results of certain processes can also oppose others. For example, you may recognize that it's a butterfly, at which point your brain "countermands" the adrenaline, now that it's decided no physical response will be necessary, and may even direct the eyes to stop wasting effort and go back to scanning for important targets.

    Assuming that the other areas of our brain are wired and operate in similar fashion, this may explain how we are able to discern so much about what we listen to at some times, while being insensitive to other details, or to the same details at other times. Once you know that a musician hit a certain clinker, whether because you heard it the last time you listened to that song, or because you read it somewhere, a relatively complex portion of your brain has driected your attention towards it... and, since the total amount of processing and sensory capability is limited, this pulls attention away from other details. This is also a neurological explanation of "training" - which simply means that we have learned to preset our attention in certain ways in certain situations based on previous experience. (So, for example, if you believe that "cymbals are more difficult to reproduce accurately than other instruments", more of your attention is focussed on those details, and less of your attention on other details.)

    However, because of this, we need to accept that, when listening to the same piece of music, or the same audio system, others may "be hearing very different things than we are". And, in fact, under different conditions, the same may even be true for us. I'm sure everyone here has noticed how, sometimes, "everything just sounds good", or the opposite. This simply means that, at those times, our attention was focussed on aspects of the music that we found enjoyable - or the opposite. The important thing to note, though, is that it is not just a surface experience. Since we cannot separate our braisn from the process, we may LITERALLY be more or less able to hear certain details depending on whether our attention is focussed on them or not. (So, for example, people who are "better trained to listen for details" may in fact actually have the equivalent of physically more acute hearing, and people who are absolutely certain something doesn't exist may LITERALLY be unable to hear it. A visual analogy is simply that "you actually won't be able to see the bird if you aren't looking up".)

     
    Phronesis likes this.
  7. KeithEmo
    Interesting indeed....

    I think it's worth mentioning that, while you made an excellent attempt with what you have at hand, the impulse response of mechanical transducers like IEMs is usually quite limited.
    Likewise, the impulse response of microphones, even very good ones, isn't especially good either.
    Therefore it makes sense that they would not convey the actual electronic output of the device very accurately.
    (of course, many folks take this as an indication that "we shouldn't be able to hear it either".)

    However:
    I found THIS recent review of the LG v30+ phone:
    https://musicphotolife.com/2017/11/lg-v30-review-best-audio-video-and-photo-experience/
    (Note that, being a phone, it might incorporate additional filtering outside the DAC chip itself, although not necessarily in "hi-fi mode").

    If you scroll down about halfway you'll find THIS set of screenshots showing the claimed impulse responses of its three filter options...
    Note that they are quite different... and rather typical of many DACs (Sabre and otherwise).
    (Also note that they may be theoretical drawings rather than actual 'scope captures.)

    upload_2018-12-16_14-0-48.png

    https://i1.wp.com/musicphotolife.com/wp-content/uploads/2017/11/Screenshots153.jpg?resize=800,486

    Interestingly ESS (who makes the chip) doesn't seem to be providing detailed information or data sheets....

    It is also worth noting that the ES9218 chip itself is quite complex - far more complex than "a simple DAC chip".
    For example, the internal "jitter reduction technology" it incorporates is actually a novel type of circuit that operates somewhat like sample rate converter.
    (The exact internal workings are immensely complicated - and probably prorietary - but it does alter the actual digital audio data as part of the process.)
    However, because the chip is so complex, it's quite possible that more significant differences between the filter options might present themselves only with more complex test signals - like actual music.

    I should also point out that some of the ESS DAC chips, beyond offering selectable internal filter choices, allow the device designer to create their own custom external filters.
    (I'm not sure if this particular chip does or not.)

    FInally, I would expect that most devices like your Shanling don't apply any odd external processing to the digital audio signal- so you are seeing and hearing the actual performance of the DAC chip.
    HOWEVER, anyone attempting to do this sort of comparison using a phone should be aware that MOST phones apply A LOT of extra processing to the audio signal.
    With many modern phones this includes:
    - built in EQ to correct for deficiencies in the internal speaker
    - optional user-controlled EQ
    - dynamic processing and limiting to protect their tiny speakers from overload (far beyond simple limiting)
    - dynamic processing to make voice more intelligible
    - high pass filtering, combined with artificial addition of extra synthesized harmonics, to provide a convincing illusion of improved bass response


     
    Last edited: Dec 16, 2018
    james444 likes this.
  8. bfreedma

    Congratulations- you’ve now found 100 different ways to say “science doesn’t know everything, so it know nothing”

    Continually creating irrational and unlikely in the extreme “possibilities” is an exercise of the absurd, not an example of proper science.

    Pigs may fly one day - have you equipped your car with anti flying pig screens to protect your windshield? If you really believed in your view of science, you need to do so if you’re going to be consistent. You never know - it could happen tomorrow- it’s possible...
     
  9. KeithEmo
    Luckily for the rest of us, science isn't limited to what you or I think is important.
    And the laws of physics nowhere state that "if you or I can't do it today then it must be impossible" either.
    (And, as any real scientist will tell you, "the laws of physics" are really just the best model we have - for now.)
    It really is just that simple.
    Why do you insist on trying to make it more complicated?

    I still haven't figured out where you've gotten that silly "science doesn't know anything so it knows nothing" meme from.
    (But it certainly wasn't from me.)
    Science knows a lot, but science doesn't know everything, and everything science "knows" is subject to revision.
    However, that doesn't mean that what science does know isn't USEFUL... whcih is all we can reasonably ask.

    Here's a link to a flying pig (he didn't fly very far):


    And here's one who completed his flight:
    https://abcnews.go.com/US/story?id=95217&page=1

    And, contrary to that last story, the FAA says that pigs may continue to fly (even in first class):
    https://abcnews.go.com/US/story?id=94861&page=1

    And, while I certainly hope nobody would be mean enough to do it, you could surely launch a pig from a cannon.
    And I'll be that, even though perhaps it wasn't recorded, pigs have opccasionally been picked up by tornadoes.

    So, yes, in fact pigs CAN fly.
    The next time I encounter a pig flying on a plane I'll say "hello" for you.
    And, if I'm unlucky enough to be caught in a tornado, and see a pig flying towards me, I'll duck.
    (I would suggest that, if it ever happens to you, you abandon your preconceived notions long enough to duck as well.)
    And, no, I doubt I'm LIKELY ENOUGH to have a pig fly into my car to justify the cost of a "pig-screen".
    Have we cleared up the difference between "impossible" and "unlikely" yet?
    Can we stop now?

     
  10. bfreedma

    That’s exactly the point Keith - you’re using absurd examples as a way to avoid acknowledging current scientific models. In your universe, we need to test every possible scenario, past, present, or future prior to establishing a rational model. That’s simply not how science “works”. No one is disputing that new knowledge may change applied science, but in order to move forward, science relies upon testing existing and rational potential use cases which may deliver consistent and repeatable results. Those results become applied science which is used to define operational parameters and functionality. When new discoveries require change, then the models are updated or potentially replaced.

    If your vision of science were real world, than we would have no applied science as there is no field where every conceivable past, present, and future scenario has been tested, vetted, and correlated. You only weaken your argument with the childish and obviously irrelevant examples used above.

    You’re playing games, continuously throwing extremely unlikely use cases at the wall, providing no actual evidence, then claiming they somehow represent rational doubt of our current knowledge base. It’s whataboutism of the worst and most unconstructive kind. It’s the equivalent of saying everything we know about physics is wrong because who knows, the earth could be pulled inside the Schwarzschild radius of a black hole next month. After all, we can’t say it’s impossible for that to happen.

    Last response can be yours. I’m no longer willing to participate in your gaming of the topic and the damage it’s causing to the actual discussion of audio science in terms of real world application.
     
    Last edited: Dec 16, 2018
  11. Phronesis
    Speaking as an engineer who likes science, I think we need to keep in mind the distinction between the two.

    The models used in engineering are intended for practical purposes, and they need only be accurate and reliable enough for those purposes. Sometimes those models aren’t very “scientific” and involve very substantial empirical data fitting.

    Scientific models aim towards capturing “truth” as much as possible, and they need not be suited for practical use at all, though often they are.

    So, for example, we might find it appropriate to include ultrasonic effects in scientific models of perception because there are observable differences in activity of the auditory cortex, but we might find that those effects are negligibly small for listener experiences, so we neglect them in audio engineering models.
     
    Last edited: Dec 16, 2018
  12. Steve999
    Point A: You keep changing your picture. :deadhorse::beerchug:

    Point B: Speaking as an economics major who is incompetent in economics, engineering and science, this all reminds me of an econometrics class I took, and the right side of every big long equation ended in "plus or minus a portmanteau variable, epsilon," a.k.a (in my mind) the fudge factor. It looked kind of like this on the blackboard: "bla bla bla bla bla = bla bla bla bla bla bla bla bla +/- E." Except the E was a fancy E with curves on the top and bottom corners. I joke about the "portmanteau variable, epsilon" to this day--I can still hear the lecturer's phrase in my mind as if it were yesterday. It has become a personal cliche (if there is such a thing) for me for whenever someone isn't exactly sure about something.
     
    Last edited: Dec 16, 2018
  13. bfreedma

    No disagreement that pure science is different than engineering and the resultant engineered devices based upon science. That said, pure science and the pursuit of the “truth” consists of more than continually tossing out random cases with no supporting evidence, then refusing to develop that supporting evidence due to “lack of interest”

    Regardless, the title of this thread is “testing audiophile claims and myths”, which by definition can only include the evaluation of claims and myths based on known science based engineered solutions, i.e. claims and myths based on existing product utilization. The debates here involve existing DACS, cables, etc., not theoretical, yet to be developed devices.

    Perhaps the solution would be for those wanting to discuss pure science to create a thread dedicated to that discussion.
     
  14. sonitus mirus
    I mostly agree with what you say...122%

    We can all be a superhero.

    Humans can't fly today, yet only a little over a century ago, few people thought we could ever use machines to aid in flying, and now flying in various types of vehicles is commonplace. wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text.

    wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text.

    Only crazy folks think we can fly without an assisted wing and power. People dream about flying. Some dreams come true. We might be able to fly without machines soon. It only has to be discovered, similar to airplane flight. wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text.

    wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text.

    So, some people should be able to fly to the sun and back just like Superman, in theory. wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text wall of text.

    If someone wears blue tights and a red cape, they might be able to fly and are more powerful than a locomotive, emphasis on loco.

































    Just



    to



    be



    annoying,



    I'll




    place





    your





    comment




    that




    I




    am




    replying






    to




    way





    down





    here




    for




    your






    inconvenience.








     
  15. Steve999
    On the one hand you have a good point about the title of the thread. On the other hand look at the title of this forum. :)

     
    Last edited: Dec 16, 2018
First
 
Back
763 764 765 766 767 768 769 770 771 772
774 775 776 777 778 779 780 781 782 783
Next
 
Last

Share This Page