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.
838 839 840 841 842 843 844 845 846 847
849 850 851 852 853 854 855 856 857 858
  1. bigshot
    I can define a perfect amplifier easily... A wire with gain. If it takes the signal being put into it and raises the level and outputs it without audibly altering it, it's a perfect amplifier. Every amp I've ever run across that was in proper working order has fit that definition.
  2. bigshot
    I’ll tell you what... You go click through that link to Stereo Review and read about the other equipment in the chain for their amp test. When you’ve done that, I’ll take the time to reply to your irrelevant blather. There’s no point discussing this until you know what you’re talking about.
  3. analogsurviver
    Wrong. On both counts.

    A better built amp with the same schematics will sound better.

    It takes about 300 hours of normal playback for any newly built amp to settle down - sometimes more, but around 300 hours is the minimum I require before any serious listening should be attempted.

    If better build quality means film capacitors vs electrolitic, that most definitely does mean longer lasting amp. You can see in specs of any decent electrolitic caps the time they are kind of guaranteed to perform flawlessly - at given temperature. Now, most amps do not operate 24/7/365 - but those that do come close ( studios...) may well grew accustomed to ever slightly falling performance each day - those changes are gradual. That's why "recapped" mixing desks are such a thing in pro audio ... - if the before and after were not audible, nobody would be willing to part with cash while being at the same time incapable of working/earning money.

    Amps with mostly film caps require those initial 300 or so hours burn in - and then will most likely outlive its owner if operated within design envelope. With next to no change in either sound or measured performance troughout decades. It reduces maintenance to tending to the remaining electrolytics.

    I have been able to produce an audio chain entirely free from electrolitic caps - at least in signal path. And the difference film caps ( correctly chosen - NOT all film caps are good for all applications ... ) used in power supply ( where possible ) can make is nothing short of staggering.
  4. TheSonicTruth

    Nothing of manmade origin is 'perfect'. If it were, it would last forever, and never wear out or malfunction. Even if it passes sound 'perfectly', over time it will start to degrade in performance.

    The only perfect being is God Himself. Therefore it is good to aim for perfection, but remember, we will always fall short of that mark.
  5. Sgt. Ear Ache

  6. Sgt. Ear Ache
    I don’t believe in god but thanks for injecting your beliefs as much as possible.

    Nobody here is including, nor is interested in including durability and lifespan of components into the definition of perfection. We are discussing sound output. Nothing more.
    bfreedma and Bytor123 like this.
  7. bigshot
    I'll remind you what we are discussing here... We are talking about how amps tend to be audibly transparent. Many of them, if not all, sound exactly the same to human ears. There was no audible difference between amps of completely different design and price range in the Stereo Review test. Do you have anything pertinent to add to this particular discussion?
    Last edited: Jun 7, 2019
  8. KeithEmo
    Your first statement is not technically correct.
    There is no "precise curve that proves your results are random".
    At most, there is "a precise bell curve that is CONSISTENT with a random distribution".
    However, nobody who understands statistics would suggest that this proves that there are no other possible ways to produce that same curve...
    (Failure to see that curve suggest that your results are non-random; but seeing the curve does NOT suggest the opposite.)

    You seem to have missed the point of my second point.
    It doens't matter what shape curve you get or where it is centered.
    No matter how many students do or do not pick out the Strad... or how you analyze the resulting data...
    You WILL NOT learn with absolute certainty whether a certain specific student can or cannot do so.
    It is simply the wrong sort of test if that's the data you're looking for.

    You can test a dozen people, or a thousand, or a million.
    And it may turn out that NONE of them can hear a difference between Amplifier A and Amplifier B.
    However, even if that were true, you have not proven whether test subject #1,000,001 can do so or not...
    (At best, you may decide that your methodology has a small enough margin of error that you feel comfortable ignoring it...)

    I could test every single person I know....
    Or everyone in my high school graduating class....
    And I'll bet none of the can run a mile in under four minutes....
    And, from that, I might conclude that "human beings can't run a mile in under four minutes"....
    But all that conclusion would prove is that I don't understand statistics very well.

    Oddly, even after "decades of tests" "with countless subjects".....
    Nobody seems to be able to come up with a single one that is reasonably thorough, of sufficient scale, and sufficiently free of major flaws in methodology, to prove convincing.

    As you say, "audibly transparent" is a matter of experience...
    There is no single scientific definition for the term.
    It ALWAYS carries a qualification about what circumstances and what observers were involved.

    I'm totally confused a with your claim about nulls.
    Clearly, if we have two imperfect amplifiers,unless their flaws are perfectly identical, which seems unlikely, then they will NOT null perfectly.

    I have read about a variety of studies in which "the scores of audiophiles" were compared to "the scores of ordinary people"...
    Or "the scores of audiophiles" were compared to "the scores of professional musicians"....
    Generally with the apparent idea of demonstrating that "audiophiles who think they have golden ears really can't hear differences any better than anyone else"....
    (I don't keep track of them since I consider them meaningless... but, if you read through several tests, you will find similar references.)

    For the purposes of this discussion I am not in any way differentiating between "hearing ability" and "listening skills".
    I would expect both to contribute to "whether someone can hear something or not"...
    (But I would leave that to phase two. Once we establish that the subject can hear a difference then we can move on to whether some subjects are more likely to than others - and why.)
    However, when studies are PUBLISHED, it is pretty clear that the creators of such studies consider "using professional musicians" or "professional sound engineers" to grant extra credibility to their results.
    (The reality is that, on average, five year old children probably have better hearing acuity than professional musicians, but nobody really cares what stereo component they happen to like, since they have no purchasing budget.)

    Just out of curiosity....
    WHICH standard measurements do you consider to "tell us all we need to know about how something will sound".

  9. bigshot
    If you're going to support your opinion by arguing "we can't know everything, so we can't know anything", I'm just going to assume that you have no real point and you're arguing for the sake of arguing. Classic logical fallacy: Argument from ignorance.

    There is a finite limit to perception. A nod is as good as a wink to a blind bat. Audible transparency exists whether you want to admit it or not.

    None of us are here to entertain you. We aren't being paid to pay attention to you. If you want to discuss things with people, you have to earn their respect. You aren't doing that very well.
    Last edited: Jun 7, 2019
    Bytor123 likes this.
  10. TheSonicTruth
    And even that sound output is not 'perfect'. NOTHING we humans make is.

    PS: Don't choke on your pride, Ear Ache :wink:
    Last edited: Jun 7, 2019
  11. KeithEmo
    However, I need to point out an error.

    We seem to be talking about two different things.
    You seem to be defining "an audibly perfect amplifier"....
    Whereas I was talking about a functionally perfect amplifier....

    The concept of a "perfect amplifier", from a functional point of view, is relatively well understood, and is often used in circuit analysis.
    It includes a specified gain, zero noise, zero distortion, zero phase shift, and a perfectly flat frequency response from DC to infinity.

    When modelling a circuit we often start with "a perfect amplifier" - then consider separately the ways in which it deviates from that ideal.
    For example, an ideal or perfect amplifier also has a zero output impedance.
    And, when we model a real amplifier, we draw it as an ideal amplifier, in series with a resistor that is equal to the output impedance.

    (Since "audibly perfect" is simply a statement of your personal experience I have no comments about that either way.)

  12. bigshot
    Last edited: Jun 7, 2019
  13. analogsurviver
    Have you ever done any of such work by yourself - not read somewhere or been told by others ?

    You would not be laughing then.

    Although generally not publicly acknowledged, bettter manufacturers do burn in their electronics for an extensive period of time - those 300 hours roughly means two week of operation..It usually catches and weeds out units that would otherwise fail in the field in say half year or so after purchase - creating anger with customers and incurring costs of repair within warranty.
    It also means customer never gets to hear a unit that has not been properly burnt in - that's why reports on zero change between out of the box and final condition can actually be true. But you can not expect such burn in programme to be available in budget and lower priced equipment - it would drive the retail price too much up.

    I used to work at the microelectronics plant - and the cycling ( mainly temperature ) of integrated and hybrid circuits prior to final electric testing and shipping out differed from zero for low cost plastic package to you-would-not-believe-cycle-number-and-time for the military applications. Failure rate for the former has been low when used under reasonable conditions - and approaching zero for the high military grade under atrocious conditions. You do get what you pay for - even if specs "on paper" look exactly the same.
  14. Sgt. Ear Ache
    Oh brother

    are we talking about failure rate or are we talking about sound quality?
  15. bigshot
    How did this happen all of a sudden after such a nice long pause? I can't figure out if you guys are just attention starved and the subject is irrelevant, or if your intent is to shout down evidence you don't like with logical fallacies and made up stuff. In any other forum on this site, I would assume the latter. But I can't figure out how someone could come into the Sound Science forum and spend big chunks of their day arguing against sound science.

    Do all amps sound the same?

    "...the evidence would seem to suggest that distinctive amplifier sounds, if they exist at all, are so minute that they form a poor basis for choosing one amplifier over another. Certainly there are differences between amps, but we are unlikely to hear them." -Stereo Review January 1987

    ^ THIS ^ is what I'm talking about. How about the rest of you?
    Last edited: Jun 7, 2019
    sonitus mirus likes this.
838 839 840 841 842 843 844 845 846 847
849 850 851 852 853 854 855 856 857 858

Share This Page