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.
867 868 869 870 871 872 873 874 875 876
878 879 880 881 882 883 884 885 886 887
  1. AudioThief
    Thank you for the thourough posts @KeithEmo & @castleofargh

    If we say we have a "continuum of reality", where on the left side we have essentially the reality being that only the results of blind testing and measurements being legitimate - i.e most, almost all amps/dacs/cables and what have you would be a waste of time and money. While on the right side, the reality was that there are subtle, percievable differences from DAC to DAC and so on, where bias didn't play in at all to anyone.

    I'd wager to say that the truth would be found somewhere in between those two extremes. I'd say that eliminating bias is indeed the most important thing, and the one thing that affects us most. But I am worried that while blind testing is better than sighted testing because of those biases, it doesn't strike me as perfect.

    Is the general consensus around these parts that any competently put together DAC will sound the same? Because if that is the case, I'll need to go take a look in the mirror. I think that this is a very interesting subject that I find there is a wide spread within the community as to what people believe to be true, so I want to home in on the seemingly most correct beliefs..

    edit: I am trying to read up on the topic. So I have read the first post in this thread and I'll read around more on this sub forum as well, to get a good overview.

    A thought popped into my head. In the matter of blind testing, say you were to have two vocalists do covers of a song, but you didn't get to know who sang, or who the vocalists even were. They performed the same song i.e no differences except the voice.

    Clearly, in a sighted test it would be very easy to distinguish who sang what, and you could probably tell which one you liked more.

    But suppose you didn't get to see or know anything. Most would be able to hear a difference, easily. But I would imagine that judging who were the technically most adept vocalist would be near impossible (I am here comparing it to which DACs was the most costly, for instance).

    Now lets say you had a computer program that could take their performances, and equate them i.e they had the exact same timing, volume, etc - the only thing different now being their raw voice. I would imagine that telling them apart in a blind test would be near impossible, but I would also imagine that telling them apart with the help of knowing who sang when would be much, much easier. And not only because of some sort of bias towards the one or the other. Or say you never got to know who sang what, but you got to listen to song A or song B for several years.. I'd think you would pick out a favorite, one you liked more than the other. Even if telling the difference in the blind test was impossible.

    I am sure you have been exposed to examples such as these a million times, so I am just wondering what is wrong in my logical chain here? To me, blind tests just doesn't sway me as much as I suppose it does to a lot of others. At the same time, I don't want to be essentially believing in shamanism.
    Last edited: Jul 6, 2019
  2. gregorio
    1. Not just subreddits but various other forums as well, this sub-forum for example, pro-audio forums and probably most famously/strictly, HydrogenAudio.

    2. Not really, most headphones don't score particularly well, the balance between the two drivers is typically rather approximate, the linearity can be rather poor and the frequency response very variable.

    3. Arguably the single biggest problem in the audiophile world is the almost complete lack of understanding of the difference between a property of a sound signal (be it a digital, analogue or acoustic signal) and a perception. While we can measure all the properties of a sound signal, we can't generally measure perceptions. What makes the situation complex is that differences in the properties of a sound signal affect our perception (given a sufficient amount of difference) but that our perception can change (dramatically in some cases) even when the sound signal is absolutely identical IE. This different perception can be due to something we see and/or some cognitive bias but have absolutely nothing to do with the actual sound waves entering our ears. Using your examples, transients are a property of a sound signal (which we can therefore measure) and therefore prove scientifically/objectively that one device has a faster transient response than another. "Clarity" and "Detail" on the other hand are not audio properties, they are perceptions! And in these two cases, there are sound signal properties which can affect that perception, a significant amount of noise in the signal will reduce the perception of clarity and detail for example. Therefore, if you hear (or more precisely, "perceive") greater clarity/detail with the Stax there are 3 possibilities: A. The greater clarity/detail you are hearing is entirely due to some actual difference in the sound waves being reproduced by the different headphones, or B. The greater clarity/detail you're hearing has nothing whatsoever to do with the sound waves being reproduced by the different headphones and is entirely due to your perception, or C. Some combination of "A" and "B".
    3a. There's three problems here: Firstly, "much better" is a value judgement and therefore by definition a perception, which brings us back to the problem above. Secondly and related to the first problem, there are numerous individual factors/variables which are combined to give an overall perception of "better" or "worse", so you're going to have to define "better" much more precisely. Lastly, in a science/fact based forum (such as this one) if you can't "prove it" or at least provide reliable supporting evidence, then what you "know" is in fact just an unsupported belief/impression and not an actual fact or science.
    3b. Again, there's two problems here: Firstly, "quality" is again a human value judgement, is therefore a perception and there is no way to measure "quality", so we need to be more precise. If for example we define quality as "fidelity", then that is a property of a sound wave and we can measure it, compare those measurements for the different headphones and objectively/scientifically state which is higher fidelity. However, that won't necessarily tell you much about soundstage, because soundstage is one of those perceptions which is a combination of a number of different sound properties, plus factors unrelated to sound properties. On the other hand, it would tell us a lot about "timbre", because timbre is directly related to a single sound property, frequency response.
    3c. They are right, "everything and anything" can be proven about the properties of the sound waves the headphones or other equipment are reproducing. What generally can't be proven is what happens after those sound waves have entered your ears, how your brain changes and interprets the information from your ears and combines it with other information to create a "perception".
    3d. No, that's incorrect. Headphones are entirely based on science, the science of electrical/analogue signals and their conversion to acoustic sound waves. However, there's certainly "art" too, visual appearance for example and marketing.
    3e. Again, how do you define "better". In the case of an amplifier, it's job is obviously to amplify an analogue signal. So, if we define "better" as higher fidelity amplification then we can relatively easily prove which amplifier is better and if there is no measurable difference or it's a tiny difference below audibility then the two will sound identical (though not necessarily be perceived as identical)! Same for a DAC, whose job is just to convert digital audio data to an analogue audio signal.

    1. You need to be very careful how you phrase things here in this forum, such as; "definite" assertions of fact and the use of condescension. Even worse, is the use of condescension in conjunction with a "definite" fact that's actually wrong! Music these days is typically NOT "recorded with a higher overall volume than decades in the past", in fact it's often recorded with a lower overall volume than decades past! It's typically mixed and mastered with a higher overall volume but not recorded.

    2. The transition to digital started in the late 1970's, which is four decades ago, not one and a half and the consumer transition was largely complete three decades ago. Streaming is much more recent though.
    2a. Yes but those advances in processing power were achieved many years ago and digital/software EQ was available by around the mid 1980's.
    2b. No, the way we did it was the way we'd always done it. There is no need to first drop the track volume by 5-10% to give headroom, where did you get this from?
    2c. No, the use of compressors/limiters to make music louder was not in response to this.

    3. There's two problems here: Firstly, that is a massively over-simplified article and as such there is a great amount of factual detail missing. Extrapolating other facts/conclusions from it will therefore almost certainly result in false conclusions/facts. Secondly, it appears to contradict what you're stating, not support it. For example "The 'Loudness Wars' have gone back to the days of 45s..." - that's seven decades ago and long before either digital recording or digital EQ.

    1. Your view is common amongst audiophiles but if you think about it, it contains a serious logical contradiction. If you are hearing a difference due to your other senses, for example your sight, then by definition you're not hearing a difference, you're seeing a difference! What you're actually doing is experiencing a perception, rather than just hearing and a perception is an impression manufactured by your brain which is a combination of all your senses, including your hearing, plus various cognitive biases. What a blind test attempts to achieve is the elimination of your other senses and cognitive biases, leaving you with only what you're actually hearing.

    2. That depends on what you mean by "relatively similar". For example, differences between headphones are relatively large and well within the range of audibility. Differences between say amps and DACs though are relatively small and inaudible (assuming they're competently designed for high fidelity and used appropriately).
    2a. Agreed, if two things sound similar but look different, then it's much easier to distinguish between them. But again, that additional easiness is down to what you're seeing (and the resultant cognitive biases), not down to what you're actually hearing! In other words, let's say we have two DACs, a $60 DAC and a $2,000 DAC but this more expensive DAC is just the exact same $60 DAC placed in a much more expensive case. It would obviously be easy to distinguish between them in a sighted test but in a blind test it would be impossible (because the actual sound is identical). So the question is; would you be prepared to pay the extra $1,940 for no audible difference, just a nicer case? For some audiophiles the answer might be "yes" but the vast majority would expect $2,000 DAC to have some audible benefit over the $60 DAC, not only a visual benefit.

    3. Just because there is a consensus of belief doesn't mean that it's right/factually accurate. At one time there was a consensus of belief that the Earth was the centre of the universe, that blood letting with leeches cure almost every illness, that witches were real evil beings, etc. That's effectively why science was invented, to separate the consensus' of beliefs into a facts or a false superstitions/myths. A consensus of belief can just as easily be fact or myth and therefore it tells us nothing!
    3a. Well firstly, a decent double/blind test should be designed to eliminate the other senses and biases, while optimising the conditions for hearing the difference. Secondly, if those people could not hear a difference in such a test it is either because there are no audible difference or those particular people do not have the hearing acuity to hear them. Either way, for those particular people, if they can perceive the difference in a sighted test but not a decently designed blind test, logically it must be due to the factors the blind test has eliminated; sight and cognitive biases, rather than sound and hearing. Which brings us back to the question in 2a above!

    4. There are quite a few areas in science (and the facts) which appear to contradict intuition. But "education" can be described to a large extent as the bringing of intuition in line with science. Fortunately, although quite complex, it is within the grasp of most people to educate themselves enough in the science/s of audio for those intuition contradictions to disappear, unlike certain other areas of science, such as quantum mechanics for example, which is completely counter-intuitive for all but a small number of massively educated physicists. The problem is, most audiophiles simply can't be bothered, which is understandable because it's still quite an undertaking, so instead they just take the most readily available and easily digestible information and then extrapolate conclusions/facts from that. Unfortunately, almost all the "most readily available and easily digestible information" isn't science or the facts, but marketing materials!

    5. No problem. None of us were born with the knowledge of how audio works, so it's not a case of you being dumb, just a case of you being at a different place in the journey, a place we've all been. That's one of the reasons this sub-forum exists, to separate the facts from the intuition (or marketing) and hopefully help people along that journey. Unfortunately, you'll find this forum can often/sometimes get very heated, in pretty much every case it's because many audiophiles are only interested in justifying their intuition at all costs, even if it obviously contradicts the science/actual facts and is completely illogical!

    Last edited: Jul 6, 2019
  3. TheSonicTruth
    So you're admitting it!

  4. AudioThief
    @gregorio That is the single best post I've read on this forum. I am at work so I could only skim it, I will read it again later today (EU time). I appreciate it a lot, thank you
  5. gregorio
    There is a loudness war, I and others have presented reliable evidence that proves it (but that you "pay no mind"), I've been commenting against it for probably at least a decade before you even knew it existed and I've repeated this god knows how many times! Honestly, what is wrong with you?

    sonitus mirus and bfreedma like this.
  6. TheSonicTruth
    Then how come all of you are asking me to use a meter to prove that a remastered version of an album on CD is louder than the original issue on CD, or modern CD releases in general, when my own DAM EARS tell me the remastered is louder? Then you go and tell me I used the "wrong meter" - the SAME TYPE enforcers use to check volume compliance at clubs and concerts?
    Last edited: Jul 6, 2019
  7. gregorio
    When is enough, enough? Again, what is wrong with you? I never asked you to use a meter to prove anything, I never mentioned remastered versions of an album on CD and the meters they use to check compliance at night clubs and concerts are NOT measuring or checking volume/loudness, they are measuring sound pressure level compliance, there is no loudness compliance for clubs or concerts. It's all complete nonsense you've just made-up!

    I'm glad it was useful. If you find something that doesn't make sense or doesn't agree with your intuition, please don't be afraid to ask. Don't let the heated exchanges put you off. Heated exchanges never occur if you ask questions, only if you repeatedly make false assertions of fact.

  8. TheSonicTruth
  9. gregorio
    Exactly! Now where in that document does it say that you measure loudness with an SPL meter? Also, the tech specs it contains have only been implemented for TV broadcast in some/most European countries. It does not apply to music, either music recordings, clubs or live gigs. The legal restriction for clubs and live gigs, if there is one, is NOT based on loudness but on peak level (sound pressure level, SPL), which is what you measured!

    It's good that you've finally posted some actual reliable evidence, rather than just "because your wife said so". Clearly though, you obviously haven't understood the document you've posted, as it contradicts your assertions!

  10. TheSonicTruth
    Gregorio: Open your narrow mind and try to apply this excerpt on page 9 to the music business - the PRINCIPLE of peak vs loudness norm.


    Again, what's on the left corresponds to what has happened, in part, due to peak metering throughout much of the era of digital audio recording, mixing, and mastering - peak-based metering. On the right wouls be both the idea, and APROXIMATELY: what the situation was back during the analog/VU meter era, pre digital. Loudness metering would allow us to get back to that.

    Now go ahead, demonstrate to everyone here what a j*rk you are by saying I've "got it all backwards"!

    Seriously! That's all you're good at.
  11. bigshot
    I think it's hopeless. I don't think he understands what anyone else is saying. He's just reciting routines that he's memorized.
    bfreedma likes this.
  12. bfreedma
    Insults are the arguments employed by those who are in the wrong.

    Jean-Jacques Rousseau
  13. TheSonicTruth
    BS.. I understand a LOT more than you all care to admit or give me credit for.
  14. AudioThief
    Can someone give me a quick rundown on the major differences between a typical objectivist and a typical subjectivist?

    I am having difficulty placing the exact stance of objectivists.. On reddit, you will hear people talk about zero audible gains above HD650 (in price), for instance. That doesn't make sense to me. At the same time, cables making a difference doesnt make sense to me either.
    Last edited: Jul 6, 2019
  15. dprimary

    That is a broadcast standard. It does not apply to SPL or loudness in a live venue. It would only apply to the broadcast feed of the live event which is whole different mixing desk and processing chain.
867 868 869 870 871 872 873 874 875 876
878 879 880 881 882 883 884 885 886 887

Share This Page