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

Objectivists board room

Discussion in 'Sound Science' started by joe bloggs, May 28, 2015.
279 280 281 282 283 284 285 286 287 288
290 291 292 293 294 295 296 297 298 299
  1. Allanmarcus
    Man, you are tedious.

    While technically correct, you are also wrong. You don't need trained ears to hear differences. I never said that. Also, many people may not have the sensitivity in hearing to even hear the subtle, measurable differences in DAC.

    What I said was:
    "So, to sum up, a trained ear might be able to distinguish, and prefer, and describe, the subtle variations in DAC technology implementations (e.g., different DACs). Objectively, not all DAC implementations are the same. Measurements often prove this."

    I will expand this a little since, by your comments, I'm not making my point as well as I should (or you are just trolling, I cannot tell). A person with sensitive and trained ears may very well be able to distinguish, and prefer, and describe, the subtle variations in DAC technology implementations that are so subtle, that without trained and sensitive hearing, a person may not be able to distinguish and describe those variations.

    For example, there are clear measurable differences between a Bimby and a Gumby. I cannot hear the differences (other than the volume). I built a switch box and A/B tested (blind and sighted). Others claim to be able to, even in a blind test. The voltage is not the same coming out of the DACs, so the testers were able to determine that one was slightly louder, yet they described the sound from the slightly quieter DAC consistent with the variation in measurements.

    I'm a believer is science, but not dogma. This is at the sound science forum, not the sound dogma forum. Expectation bias is real, and it obvious in this industry. That doesn't mean there aren't people that can hear the subtle differences in measurable variations. DAC implementations, unless identical, often have subtle, measurable, variations that some people can hear, describe and prefer. Those with sensitive hearing are more likely to hear these variations. Those with sensitive hearing and trained ears are more likely to be able to hear and describe those variations. The is all I am saying. If you are agreeing with me, then ok. If not, I am not understanding what you are saying.
  2. cel4145

    Sorry. You didn't say between "expensive DACs." But you did claim,

    I'm not convinced at all that there are subtle audible differences between many DACs regularly recommended and talked about on Head-Fi, regardless of what kind of training and ear sensitivity one has. And the training and sensitivity argument seems used a lot to support subjectivism views that there are.
    sonitus mirus likes this.
  3. Allanmarcus
    Agreed. There are probably many DACs for which there is no audible difference. There are also many DACs for which there are measurable differences, and those differences may be discernible to some, and not to others. I also agree that expectation bias can play a huge role.

    Unfortunately, some of sound reproduction is still an art, and not a science. Until we can be sure we are able to measure everything, and that we even understand what we should measure, some of music reproduction is subjective and should be subject to DBT for validation. I think we are on the same page.
  4. castleofargh Contributor
    I can do with the rest of your post, it's hard to know if you're talking minute improvements around the usual threshold of hearing(maybe perception would be better than hearing for this), and then we're on a consensus basis. or if that argument would end up justifying Chord level stuff with allegedly audible changes down at -300dB. so maybe you could reassure the boys with some concept of magnitude, but otherwise I'm fine with what you said.
    but this specific quote smells a lot like confirmation bias to me. why would you care about the guy's impressions if you estimate the test wasn't properly controlled for loudness? (although, they both seem quoted at 2V in the specs, was one used balanced or are the specs only approximations?)
    the other issue is that if I know the DACs, have seen the measurements, and read reviews... how I will describe the "right" impressions without seeing which DAC it is, ends up having a 50/50 chance of success. it has nothing to do with passing the blind test.
  5. Allanmarcus
    Anything that is not-measurable but can be verified with a DBT is a valid difference, at least in my book.

    Volume leveling for DAC A/B testing, which the average person's equipment, is quite difficult. We gotta test with what we have. As for the test that atomicbob and his friends did a while back, I was just pointing out, well, you read what I was pointing out. Since the listeners didn't know which DAC was which, they only detected a small volume variation, I don't think that invalidates their other impression; especially since their impression are the opposite of the expected results (louder is better).

    As of the bimby and gumby output levels. I thought I recalled reading the gumby's actual output was a little less than max spec.

    Funny you mention Chord. I've did a lot of A/B testing with the Mojo and H2, and I passed on both. The Mojo because I was worried about the long term effects on the battery if I used the DAC as a desktop DAC. The H2 because it sounded the same, or slightly less enjoyable than the Bimby to me.
  6. cel4145
    I'm pretty sure that the problem is not that the science of measuring DACs is unable to accurately measure difference, but that the problem lies with the human condition that refuses to accept the measurements. But for that, certainly DBT is helpful.
    sonitus mirus likes this.
  7. castleofargh Contributor
    the difficulty involved in setting up a proper blind test is sadly very real. but to me it's just another reason to be extra skeptical when drawing conclusions based on partially controlled setups.
    you feel that a difference in loudness doesn't invalidate impressions, I think it does. or more precisely I think it massively lowers the confidence we can have in the experiment. which has the same result. obviously nobody is wrong all the time, and rejecting a conclusion doesn't mean it's a false one. but if something like a difference in loudness is introduced, how could I know how much of a role that variable played in the impressions? or maybe just in guessing which DAC was playing? and once the doubt is there, how can I call something conclusive?
    the level of confidence we put in our results should be directly proportional to the level of control in the experiment.

    sure. it's not like we measure everything all the time. and clearly a listening test is a perfectly valid way to confirm audible difference. that's not what I was pointing out.
    a blind test trying to detect a difference will show if a difference is detected. and that's about it. defining those differences is not the purpose of that test, nor can the impressions be taken at face value just because they occurred during a blind test. interpretations of sound are still interpretations and will involve all the annoying stuff in the brain. does this make sense? I generally suck at explaining things so maybe I just failed in the previous post. and maybe this one too? ^_^
  8. Glmoneydawg
    I(perhaps foolishly)stiil put a lot of stock in listening tests,but yeah 100% agree volume has got to identical.You are listening for minutiae in most cases.The slightest volume discrepancy will over shadow what we are listening for in the first place
  9. castleofargh Contributor
    if there is one thing I'm sure we all agree upon, it is that listening tests are important. even people forced into the objectivist box are rarely only focused on measured fidelity.
    now what is a proper listening test? answering that is when we see the world burn.
    and obviously massive differences don't require much effort to be observed or confirmed, it's when we approach the limits of audibility that precision and controls become real important.
  10. Glmoneydawg
    Whenever there are improvements in our little hobby i am sure they get there with measurements,tests ect...but when they hit the market i vote with my ears...and of course my wallet lol.I feel like there may be a little subjectivist in there my friend....anyone with a monty python reference in their name lol.
    Last edited: Mar 24, 2018
  11. bigshot
    These two comments seem to contradict each other.

    In a blind A/B comparison, trained ears shouldn't make any difference at all. If it's audible, anyone with normal hearing and an interest in doing the test should be able to discern a difference by simply comparing one against the other. If the difference is so small that it isn't audible to a person with normal hearing, it either doesn't matter or it's probably some sort of expectation bias.

    I think that people who claim to have "trained and sensitive hearing" are making that claim to bolster their ego. It doesn't mean that they have any superior abilities to hear or listen. In fact, it's been my experience that "golden ears" audiophiles generally perform worse in controlled listening tests than other people because they have something to prove. They double and triple think the tests and try to force themselves to come up with the right answer and end up barking up the wrong tree.

    People who learn about how sound works generally can puzzle out solutions to problems better, but that just makes sense. People with perfect pitch and experience in musical theory have an advantage in performing music. But listening to recorded music is a passive thing. It isn't a skill. Skills involve actually doing something. Perceiving just means that your hearing is normal.

    Exactly. If there are two DACs that sound different, please point them out so we can look at them and try to understand whether they actually do sound different- and if they do, why they sound different.

    I would like an example of this too. I don't know of any audible difference that isn't measurable. I would be very interested in hearing about this if it actually exists.

    I'm not trolling. I'm serious. But if my comments make you uncomfortable, you aren't required to respond.
    Last edited: Mar 25, 2018
    sonitus mirus likes this.
  12. Allanmarcus
    You ignored the emphasis I put in the work "need" in my original response. I will be more clear

    You don't need trained ears to hear differences.

    But in a situation where a person with untrained and/or not very sensitive hearing may not be able to distinguish a variation, a person with sensitive and trained ears may very well be able to distinguish... the subtle variations in DAC technology implementations that are so subtle, that without trained and sensitive hearing, a person may not be able to distinguish and describe those variations.

    As an example of person with trained ears vs not: Take a classical piece, say Mahler's 5th, recorded by two different orchestras in the same hall (hell, even its different halls). For the average person listening, the average person might very well not be able to tell the difference between the recordings. A person with trained ears, as described earlier, may very well be able to distinguish between the orchestras. The person with trained years may be able to identify specific differences between the timbre, pitch, pacing, interval of individual instruments, as well as the direction (how the orchestra was conducted). To the untrained person, the music may sound exactly the same.

    Now take the same recording played over two different DACs with different technologies. A person with sensitive and trained hearing may very well be able to distinguish and describe differences, while a person with less sensitive and/or non trained ears may not. Practically speaking, the person that cannot distinguish a difference should buy the lesser expensive of the DACs. The person that can distinguish a difference should get the DAC she prefers, which might be more expensive.

    Science is about theories as well as empirical evidence. Just study Einstein as empirical evidence of a theoretical physicist. Where I work, (the Los Alamos National Lab) we do lots and lots of empirical science, but we literally have a "Theoretical Division" that works on hypotheses and other stuff waaaay behind my understanding to push the envelope of what we think we know.

    I don't have the tools to measure my DACs, but in a blind test, I was able to distinguish between my Sony and my Schiit. My son was as well, as was my wife (who could not give damn). I would have preferred to have the Sony be my preference for many reasons. It was cheaper (I got a great deal), and it's also my desktop speaker amp. Having the Bimby sound better to me (and my son) was non expected (the Bimby's list price is cheaper than the Sony) and less convenient to me. Not being able to distinguish between the Gumby and the Bimby was also unexpected, but welcome.

    And once again, I'm not in the subjectivist camp. If something is not measurable but stands up to a blind test, then something is going on there. I don't need an empirical example of this for it to be true, but I have provided at least one.

    The Higgs boson particle, from what I understand, didn't exists until it was proved it did. For 50 years it was a theory until was a fact. Sure, lots of theories are proved wrong, but when evidence points to something, and theory points to the something, there is likely something.

    My point: If a blind test of two DACs with different implementations point to a difference in sound, even if that cannot be measured, then it's probable we just haven't figured out what to measure, or we don't have the right tools to measure. Expectation bias is very real, but assuming it explains all perceived differences is as dogmatic as blind trusting the reviews of audiophile pundits.

    In my example of the Bimby vs Gumby, all the measurements point to the Gumby being a better DAC. I could not hear that in my system. Others very well may be able to hear a difference, especially in different systems that may emphasize the particular areas of difference or if they have more sensitive hearing, or if they have trained their hearing to identify variations that I have not trained to hear. Also, many say DAC differences are easier to hear in speaker set ups. I was using headphones.

    Fun quotes:
    "It always seems impossible until it's done" - Nelson Mandela
    "Everything is theoretically impossible, until it is done. One could write a history of science in reverse by assembling the solemn pronouncements of highest authority about what could not be done and could never happen. " - Robert Heinlein
    "Clarke's First Law: When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong." - Arthur C. Clarke
    "All of physics is either impossible or trivial. It is impossible until you understand it, and then it becomes trivial." - Sir Ernest Rutherford
    "But by far the greatest obstacle to the progress of science and to the undertaking of new tasks and provinces therein is found in this—that men despair and think things impossible. " - Sir Francis Bacon
    "The wise man is one who, knows, what he does not know." -Lao Tzu, Tao Te Ching

    "Because it's not true today doesn't mean it will not be true tomorrow" - Allan Marcus
  13. bigshot
    This is interesting!

    What was the difference you detected between the Sony and the Schitt? What specifically were you hearing? Frequency response differences? Distortion? What frequency ranges? Have you done any controlled test with these two compared to other DACs or players? If so, do all of them but one sound the same? If so, my first guess would be that one of them is defective. Is the odd DAC still under warranty? If it was mine, I wouldn't tolerate a DAC that was out of spec. I would return it for a refund. But if it's out of warranty and that isn't an option, the next thing to do would be to compare it to other DACs of the same brand and model and see if it is defective by design.

    If DACs are supposed to be audibly transparent, then any DAC that sounds different is definitely out of spec. It's useful to document DACs like this so people don't get stuck with an inferior DAC.
    Last edited: Mar 25, 2018
  14. bfreedma
    Perhaps I missed it, but what type of controls were in place during the blind test. How were the units level matched and how close was that matching?

    It was already mentioned that there was a small volume variation - my guess is that explains the "differences".
  15. bigshot
    Oops! You are absolutely correct! I forgot that.

    If he still has the DACs, it would be a great idea to conduct a more careful blind comparison. I'm sure folks here would be happy to help him set one up. Then his trained ears can be put to work describing the differences if they still exist.
    Last edited: Mar 25, 2018
279 280 281 282 283 284 285 286 287 288
290 291 292 293 294 295 296 297 298 299

Share This Page