Maybe you would get better results by rewarding the subjects, with a snack, when they guess correctly. That way they'd get fed 50% of the timeand would have incentive to continue testing.
I understand it's human to have a fleeting knowledge of auditory perception, considering it a something that we do every day & therefore as simple as a microphone (ear) recording sound to a storage device (brain) for playback. And to base a couple of years doing blind tests based on this wrong premise. I understand the back slaps & attention you can get on forums doing this - it's enticing.
When information is presented which undermines all of this, I understand your reluctance to accept it but you do have a chance to correct these misunderstandings by opening the door that you are being brought to.
Why are you afraid of doing a blind preference test if you are ACTUALLY searching for differences between DACs? It's blind so no cheating should be possible. If one DAC is preferred over another in a statistically relevant test what does this tell you - it's the "worlds best" (as a lot of your other threads are titled)? No, it doesn't.
Now is your chance to show that you are trying to find the audible difference between DACs as you stated & as some posters believe you are doing. Do you not want to find out if you & your participants "shock" at finding that you couldn't "spot-the-difference" between DACs was due to the nature of the test & the nature of auditory processing?
They only tell us about preference, which has little to do with high quality sound reproduction.mmerrill99 said:
It may be possible to say whether X is A or B without actually determining what the difference is. In fact, I suspect that in most cases where the test is on the threshold of what is audible this is what happens. If someone can identify and describe a difference then for him this not a threshold and he is likely to get a high score.
I am sure Harmon can take care of their own business. If the aim is hi-fi then preference is unhelpful; the test is 'does it sound like the real thing?'. I once saw someone on here say that he liked his sound system because it made music sound better than the real thing; that is someone who is not interested in hi-fi.
But I don't know that all these amateur ABX tests find a null result. Surely more common is a non-null result caused by poor level matching, for example?
All we have here is a result that four particular listeners in one environment could not distinguish between two DACs, which they say they expected to sound different because they were electrically quite different. It is a data point. It is consistent with the statement that all competent DACs sound similar (because they are all good enough), but does not prove this. It is also consistent with 'all except hugely expensive DACs sound similar' (because all less expensive DACs make exactly the same design mistakes, despite their big differences). It is consistent with 'most cheaper DACs sound similar because they are all good enough; more expensive ones may sound different because some of them contain glaring design errors'. See, we can pick the outcome which suits our prejudices.
As I keep telling you & others, auditory perception is not a microphone (ear) recording sound onto a playback device (brain) - we don't consciously hear all of the sound from the waves of air pressure that is impinging on the ear each moment in time - we select what to focus on.
If you would just for a moment think about that & think about what an ABX test is about, you will realise that it's far more complicated than your simple test involves.
If you are interested in discovering what's behind the "shock" you experienced with the first test, you owe it to yourself to do a BLIND preference test.
If not, carry on with the crusade - I'm sure the Spanish Inquisition will be arriving shortly & you'll all fit right in
Objectors wants to talk about LIES.
Very well, let's talk about LIES. The truth is the polar opposite of what mmerrill99 said previously.
The whole audiophile's parallel reality is based on perceptions that fails to pass most of the ABX testings. Cables, amplifiers, HD/lossy files and now converters.
For the past decades, that parallel reality was feeded by magazines, salesmen speeches, mouth-to-mouth and a market that was making good money, very good money, by making the illusion last for as long as possible.
Then, people started to ask questions.
One way to do that, in an effective manner, is by using ABX testing. Some basement tests first, there and there, and then some more serious, organized tests, such as the one from McGill university.
But the LIES continues to breath, because of the sustaining doubt, still floating around... ''But what IF...'' If, golden ears, if not high resolution enough, if that cable was not enough, if ABX testing method was flawed, if participants are biased, if ABX tests organizers are dishonest... IF. The doubt that floats. The fog that keeps coming back, over and over.
It's hard to accept that we are being fooled by our own ears, our own brain. It's even harder to accept for a seasoned audiophile, who is realizing that his past choices were partly based on illusions... Hard to accept, hard to swallow. Mindblocking is at doorstep.
But now that every serious studies show that our sensory system and brains are cheating on us (in a good protect-our-species way), that we may be mislead by so-called higher quality audio gears... What do we do about that ?
Keeping that big lie alive, by spreading it over and over, like a contagious disease ?
Or just accepting, once and for all, that our sensory system is limited by design... Therefore making some of the most popular audiophile's choices completely pointless ?
Very well, let's talk about LIES. The truth is the polar opposite of what mmerrill99 said previously.
The whole audiophile's parallel reality is based on perceptions that fails to pass most of the ABX testings. Cables, amplifiers, HD/lossy files and now converters.
For the past decades, that parallel reality was feeded by magazines, salesmen speeches, mouth-to-mouth and a market that was making good money, very good money, by making the illusion last for as long as possible.
Then, people started to ask questions.
One way to do that, in an effective manner, is by using ABX testing. Some basement tests first, there and there, and then some more serious, organized tests, such as the one from McGill university.
But the LIES continues to breath, because of the sustaining doubt, still floating around... ''But what IF...'' If, golden ears, if not high resolution enough, if that cable was not enough, if ABX testing method was flawed, if participants are biased, if ABX tests organizers are dishonest... IF. The doubt that floats. The fog that keeps coming back, over and over.
It's hard to accept that we are being fooled by our own ears, our own brain. It's even harder to accept for a seasoned audiophile, who is realizing that his past choices were partly based on illusions... Hard to accept, hard to swallow. Mindblocking is at doorstep.
But now that every serious studies show that our sensory system and brains are cheating on us (in a good protect-our-species way), that we may be mislead by so-called higher quality audio gears... What do we do about that ?
Keeping that big lie alive, by spreading it over and over, like a contagious disease ?
Or just accepting, once and for all, that our sensory system is limited by design... Therefore making some of the most popular audiophile's choices completely pointless ?
Last edited:
Haha, you jest, right? Do you 'hear' what you are saying? Just step back & look at it again. You are trying to say that ABX testing tells us which is high quality reproduction? If you ever read the Harmon/Olive work on blind preference testing you would realise that their conclusion was that all listeners favoured the speaker with the lowest measured on & off-axis distortion. So your attempts at twisting preference to mean just a "liking" for some flavour of sound is not borne out in true blind tests, properly organised.
Why do you think that these very well known & respected researchers in the field of audio didn't insist that blind ABX tests must be conducted before blind preference tests? Because they know such an insistence is ********.
Sure but in the little we know about the test from the o/p, he's using short music clips in order to be able to distinguish differences or didn't you pick that up?
More BS I'm afraid - you don't know what the 'real thing' is for that recording - you weren't at the sound studio when recorded (if it was recorded live) - most aren't recorded live so 'real thing' is completely meaningless & what Toole calls the "cone of confusion" - look it up.
Something is a data point when the data is gathered using valid techniques appropriate to the question being asked - we still are not sure of this from the o/p.
You are totally confused - one minute you suggest the question is "is it accurate" then you suggest "'does it sound like the real thing?'
And such confusion is typical of those who don't really want the truth just the version that will support their beliefs
How about trying to understand auditory perception as that is what you are testing & doing the correct test which has the best chance of answering a SINGLE question rather than a crusade hiding behind some amateur test?
Anyway, I can see a crusade is well under way - I can smell the burning crosses - so carry on!!
Reminds me of some lines from Yeats:
"The best lack all conviction, while the worst
Are full of passionate intensity."
Carry on crusaders!!
I am very serious. ABX testing cannot tell us what is hi-fi, but neither can preference testing. What ABX (and other ears-only) tests can tell us is whether two items (and hence many items if further tests are done) are distinguishable. Indistinguishability is a requirement for hi-fi; a necessary but not sufficient condition. This is simply because one sign that N items are all good enough at sound reproduction is that we cannot tell them apart. If we can distinguish them then at least some of them are not good enough.mmerrill99 said:
Speakers are about the weakest part of the chain, with high distortion and wobbly frequency response. Therefore it is likely that many of them have distortion so high that it is perceived as distortion, hence there is a preference for less of it. Smaller amounts of distortion may be misperceived as 'warmth' or 'musicality'.
Distinguishing a difference (indicated by ABX scores) and identifying a difference (describing what it is) are not the same thing.
This is precisely why many listening tests (even fully blind) are unhelpful for hi-fi. However, ABX can at least tell us whether items satisfy the indistinguishability requirement, which is necessary for hi-fi.
They mean the same. Sufficiently accurate reproduction means that it sounds like the real thing. That is hi-fi; everything else is just audio.
I note your repeated use of the word 'amateur' in a somewhat derogatory fashion. I remind you that 'professional' simply means that someone else is paying, not that there is increased competence in action. Payment can cloud judgement.
I said i'll dig deeper, well, i will.
Just took an appointment with an audiologist. Questionning ourselves starts with some audiology... I'll get my ears tested but i sure will ask some questions as well.
Audiology - Wikipedia
Just took an appointment with an audiologist. Questionning ourselves starts with some audiology... I'll get my ears tested but i sure will ask some questions as well.
Audiology - Wikipedia
Can't directly help much with D/A listening, but I have some 16/44 files of A/D converters. Some are easy to tell apart even at CD resolution, and some are harder. Here is a link: Dropbox - GQ
To get started with listening to the converters, listen to the cymbals first. Real? Distorted? White noise?
After that guitar double stops sound different (when two strings are plucked at once), but a little more practice may be needed to learn to identify those differences.
It can take a lot of practice to hear differences for some people, and maybe less for other people, but the point is it's not about listening to music or frequencies. It's about listening for something different that the mind has to learn to recognize, kind of like learning to recognize words spoken in new language. Hearing that there are frequencies, etc., doesn't help. It's about learning to hear something new.
The reason I mention this here is that practicing with the files I linked to may help people learn how to better hear D/A differences. Similar kinds of issues as with A/Ds.
To get started with listening to the converters, listen to the cymbals first. Real? Distorted? White noise?
After that guitar double stops sound different (when two strings are plucked at once), but a little more practice may be needed to learn to identify those differences.
It can take a lot of practice to hear differences for some people, and maybe less for other people, but the point is it's not about listening to music or frequencies. It's about listening for something different that the mind has to learn to recognize, kind of like learning to recognize words spoken in new language. Hearing that there are frequencies, etc., doesn't help. It's about learning to hear something new.
The reason I mention this here is that practicing with the files I linked to may help people learn how to better hear D/A differences. Similar kinds of issues as with A/Ds.
Last edited:
I have tinnitus, which was the main reason I saw an audiologist, I'd heard hearing aids may help with it. I met with a couple of them during my consultations, both were very understanding, calm and genuinely sympathetic. When you consider one of the questions I was asked was whether I'd had any suicidal thoughts, there comprehensive training is more understandable. I was also told that tinnitus can include aural hallucinations, to the degree of hearing music and even voices, some sufferers have literally thought they were schizophrenic or going mad.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.