Double Blind Test (DBT)

[Ambio]

Re: Double Blind Test (DBT)
Mods have bifurcated the thread and posts have gone haywire. Hence quoting all the threads that are relevant to my post below:-

Quote:
Originally Posted by Ambio
Some would rather use their ears to judge sound but when it comes to buying gold or 1kg onion or 1 liter petrol they would rely on measurements rather than their senses. Dr.Bowes described this as a new form of neurosis known as audiophilia. He discovered that long before I was even born!
Quote:
Originally Posted by captrajesh
Forget apple to apple comparison this is not even fruit to any other fruit comparison. Do you sincerely believe stereotyping all of them as one bunch is a correct thing to do!
Quote:
Originally Posted by Ambio
Why not? It applies to all who rather rely on their senses rather than science. To illustrate how poor is our hearing. Ask someone to play the same track everyday at 0.1dB difference and you try to guess the correct level. 0.1dB is too small? Ok...make it 1dB . Than, at least we can believe that the person who could hear the difference indeed in possession of super hearing. Ever seen any audiophiles tried to prove their hearing superiority by identifying pitch? phase? loudness? highest audible frequency?
The topic of discussion is whether one should rather use his senses or the so called specifications to base ones judgement on which sound is pleasing to ones ears. Hope we are on the same page here. What has science got to do here!! FYI I've graduated with pure sciences as majors so there is no confusion on my part.

Just because an average person's hearing cannot differentiate 0.1 dB / 1 dB change in amplitude (as per your post) it doesn't mean that he would be unable to perceive what pleases his ears. It is like saying, a person who doesn't have 6/7 eyesight cannot be a marksman.

It's a totally different ball game.

Quote:
Originally Posted by jls001
The next significant sound level is +/- 3 dB (from a given reference level). Human hearing acuity is not blessed with the ability to distinguish 0.1 dB difference (or even 1 dB), but most people can make out a difference when difference is of the order of 3 dB.
Quote:
Originally Posted by Ambio
Actually, you could tell 0.1dB after some training and side by side comparison. There must be a few recording engineers in this forum. Anyone?
And a few posts later you say this!! I'm not sure what your point is!!
__________________
LIVE & LET LIVE

GR Research speakers; build threads

Managed to find holes in my posts ah?

You are still confused and your tertiary education is no help here. Let me explain to you in a more simplistic way.

Take two glasses of water. Add few granules of salt in one. Let's say your taste buds too are as sensitive as your ears. Under DBT you may able to tell the difference. Just like the 0.1dB. Now ask your wife, mom or anyone to make a nice meal. Split the meal in to two and add few extra granules of salt in some of the dishes. I mean a few granules. Do you honestly could tell the difference? Or does it matter?

When you listen to music, it is a mixture of many sound. You listen to the rythmn not a single fraction of a second. When it comes to the so called difference you hear that cannot be identified when it forms part of the music unless you have a broken equipment that changes make such a profound difference.

In those days, when someone claims that they could hear sound that other couldn't, the villagers would exorcise him or her to get rid of the demon. Maybe, it is time to revisit such practice in audiophile world. If they say they could hear the difference then why can't they prove it? Either they are hallucinating or not being truthful to cover their guilt for spending money that only elevates you in your social standing.

 

[jls001]

@Ambio: I'm not arguing with your belief in DBT - that's entirely your prerogative. My quarrel is only with your number (0.1 dB). It's simply not a realistic change in loudness level for humans to discern, whether he/she is a recording engineer or a regular citizen.

PS: I would love to understand more about DBT and I had asked a set of questions in an earlier post but no one has bothered to answer them. May be some educating (on the part of those who have been following this discipline in theory and in practice to help discern the grain from the chaff) is called for? To my untrained mind, the prime reason why the uninitiated find DBT hard to accept is because the practitioners do not bother to explain the nuts and bolts of the procedure, while also adopting a holier-than-attitude to those who doubt or disagree. Not exactly conducive to convincing others, I'd say I say do educate us. Let others believe and follow (or otherwise) on the basis of its merits/demerits.

 

[Thad E Ginathom]

You might be able to get some more practical advice on DBTs on Hydrogenaud.io. Try the wikki.

Of course, it's a hard-line objective site, perhaps the very home of objectivist audio. I would never abandon my subjectivist friends, because, sure I'm interested in their subjective experiences and opinions of various sounds. I certainly don't feel, outside of a couple of areas, the need to shout "prove it!" at everybody over everything. But hey, I find that hydrogen audio provides a nice balance in audio life

 

[Ambio]

@Ambio: I'm not arguing with your belief in DBT - that's entirely your prerogative. My quarrel is only with your number (0.1 dB). It's simply not a realistic change in loudness level for humans to discern, whether he/she is a recording engineer or a regular citizen.

PS: I would love to understand more about DBT and I had asked a set of questions in an earlier post but no one has bothered to answer them. May be some educating (on the part of those who have been following this discipline in theory and in practice to help discern the grain from the chaff) is called for? To my untrained mind, the prime reason why the uninitiated find DBT hard to accept is because the practitioners do not bother to explain the nuts and bolts of the procedure, while also adopting a holier-than-attitude to those who doubt or disagree. Not exactly conducive to convincing others, I'd say I say do educate us. Let others believe and follow (or otherwise) on the basis of its merits/demerits.

If I am not mistaken you have already found the official site when you posted a link to show the difference in old Philips Cd Player and a Sony. The site belongs to Ary Krueger the inventor of ABX machine. In a way, I am indebted to him for his thrashing of my ignorance belief in the early 2000s. I conducted many DBT of my own just to prove him wrong and still hoping I could do so.

Anyway, I don't really agree to the way the official DBT is conducted as I prefer to do my listening at my leisure in my system. I don't want to level match. I let some one else to change the players or cables over a few weeks or months and I just compare the notes at the end of the DBT. please see my post on Critical Listening and also my response to Coaltrain.

 

[captrajesh]

Rajesh, I thought it was about double-blind tests. That has nothing to do with specifications, or even measurements, and it is certainly not double-blind tests versus hearing. As I understand it, it is completely the opposite: it is very much about using one's hearing, and allowing it to function unencumbered by other senses or mental processes.

Blind testing is about listening: pure listening. Using the ears, trusting the ears. Ears can be trusted: it's other stuff that gets in the way.

What's not to like about that?

(It's hard work and a pain to set up. Anything involving hardware means planning, preparation and team work. It is not something one can do on a whim at 3.00am. That's what I don't like about it! Thing is, though, most "audiophiles" are really a great deal more dedicated than I am.)

Thad, Im getting what you are trying to say. I tend to agree to your point of removing personal biases if one is required to judge and analyse SQ but am not sure if DBT is the Holy Grail in achieving the objective.

Anyways, my point was not about double blind tests at all. Its about futility of relying on data for judging the sound rather than ones own ears.

You are still confused and your tertiary education is no help here. Let me explain to you in a more simplistic way.

Take two glasses of water. Add few granules of salt in one. Let's say your taste buds too are as sensitive as your ears. Under DBT you may able to tell the difference. Just like the 0.1dB. Now ask your wife, mom or anyone to make a nice meal. Split the meal in to two and add few extra granules of salt in some of the dishes. I mean a few granules. Do you honestly could tell the difference?

When you listen to music, it is a mixture of many sound. You listen to the rhythm not a single fraction of a second. When it comes to the so called difference you hear that cannot be identified when it forms part of the music unless you have a broken equipment that changes make such a profound difference.

I brought up my science background because you are hung up on the so called science; only to clarify where Im coming from.

Now leaving my tertiary education aside, I did fully understand your analogies comparing human judgment of sound with buying of Gold, Onion and Petrol and your argument that human hearing is so poor, that it cannot differentiate a 0.1/ 1 dB difference in loudness and thus cannot be a reliable means of judging the sound quality.

Im sure you are aware that for every sensor, be it biological or electronic, there is a threshold limit. Unless the stimulus exceeds the threshold limit, it would not be detected. Hence, I agree with your argument that humans may not be able to perceive below a certain difference in loudness in music being played (or a certain quantity of salt in the food)though Im not sure of what that certain level is as I do not have personal experience in such experimentation.

While it may be comparatively straight forward to perceive a certain difference in concentration of the salt in water, I do not agree that it would be undetectable in food though it would be comparatively difficult.

My point is simple. What is this inability to perceive loudness below a threshold limit has got to do with ones ability to perceive what is pleasing to ones ears. Why does one need to know the science behind it to appreciate what is good or what is bad? Id be glad to have your views on this aspect.

If they say they could hear the difference then why can't they prove it? Either they are hallucinating or not being truthful to cover their guilt for spending money that only elevates you in your social standing.

I don't think you can generalise and stereotype all people in one single category. Each individual is different and has different capabilities. Lets accept the fact that the sensitivity and concentration levels of individuals vary. When you also add keenness and interest into the equation, quality of perceptions would vary quite significantly.

BTW, with regarding hearing, I was browsing and discovered this article which makes for an interesting reading: Pitch, Loudness, Timbre

 

[Thad E Ginathom]

Thad, Im getting what you are trying to say. I tend to agree to your point of removing personal biases if one is required to judge and analyse SQ but am not sure if DBT is the Holy Grail in achieving the objective.

No pun intended on "objective?" :lol:

It may not be the holy grail; but the idea is to avoid the pitfalls of sighted testing. The evidence seems to suggest that nobody can avoid those pitfalls, however practised, however trained. Closely related is the McGurk thing: researchers say that the most amazing thing is that, however much they study it, however much they conduct the test themselves, however much they know what is going on --- the illusion never fails even on them. As individuals, our brains and the way that they work are really not that different.

Yes, people have different skills and levels of ability. A pitch that is acceptable to me might be way out of tune to a musician ---or someone who just happens to have been born with perfect pitch--- but this doesn't mean that they, or I, would fare better identifying different samples of music unless pitch was the aspect that differed.

A closer-to-home example might be, say lossy-compression bit rates. People can, I hear, learn to look for and identify the exact artifacts that might give away one high MP3 rate from a higher one. To the codec developers, this is part of the job: no point in working on perceptual compression, if one can't perceive the results. There are specialities; there are speciality skills --- but there are still brains and bias and all that.

There is also a logical objection to bouble blind testing which sets out to show that there is no difference between A and B: of course, one cannot prove the negative: it may not conclude that there is no difference, that nobody would ever hear a difference. Ever. Ever!

But, in the controversial areas of audio, failure of the confident to prove the positive that they can hear certain differences seems common.

 

[keith_correa]

It's clear that the people debunking DBT's have never done this and have no wish to do it. That is their prerogative BUT then they lose the right to criticize and say that it is a load of rubbish [not in so many words, but...].

Now, if I were to have trusted my senses [not only my ears] and on that basis, would have bought some equipment, and if someone would come along and introduce a theory that the basis of my choice was wrong and then propose a test that may prove that he was right - do you think I would even participate in such a test? If I agreed, what would that make me?

 

[keith_correa]

OT, but the matrix-Hifi website has a nice article on speaker cables.

Years ago, I even wrote some formulae in MS-Excel to do the math that he details in the article to arrive at what size wire would be ideal for a given amplifier output and cable length [I think] from amplifier to speaker. I can't locate the damn sheet now. If I find it, I'll upload it here for you guys to poke holes in it.

 

[Thad E Ginathom]

Now, if I were to have trusted my senses [not only my ears] and on that basis, would have bought some equipment, and if someone would come along and introduce a theory that the basis of my choice was wrong and then propose a test that may prove that he was right - do you think I would even participate in such a test? If I agreed, what would that make me?

It would make you one of the most open-minded, adventurous, and willing to be proved wrong amongst is.

Of course, not every major expenditure leads to happiness. Whilst there are inbuilt biases that can be negative as well as positive, how many of us have never taken a precious purchase home, plugged it in, and with no need of blind tests, known that it was mistake. It happens.

 

[keith_correa]

how many of us have never taken a precious purchase home, plugged it in, and with no need of blind tests, known that it was mistake. It happens.

Happens all the time with me. And then comes that sinking feeling...

 

[keith_correa]

Re: Are CD Player still relevant?

The next significant sound level is +/- 3 dB (from a given reference level). Human hearing acuity is not blessed with the ability to distinguish 0.1 dB difference (or even 1 dB), but most people can make out a difference when difference is of the order of 3 dB.

Strange, but when I took that test, I could clearly detect a 1 dB difference and I scored 9/10. And me with my cloth ears!

 

[Thad E Ginathom]

Blind testing Level Differnece

1db reaches the point where it is very hard for me. No, I would not be able to answer any of the 0.5db tests.

This test is curious, though. Is it just me, or does anybody else find it easier to hear a rising-level example than a falling-level example? Is the some psychoacoustic reason for this?

But this is my theory...

A change in level which is too small to be described by the listener as "louder" or "quieter" might still lead them to hear a difference in the sound. This may be something to do with the hearing sensitivity curve. A slight increase in volume, too small to be "louder" might make a person say, "better bass," or "better details." This, of course, would not show with a single-frequency tone.

If that is true (and I think it is ) then it explains the importance of level matching to a greater accuracy than that which is simply perceived as a difference in volume when trying to make an objective comparison, sighted or unsighted.

Remember the maxim: Louder, if not already too loud, always sounds better. Hifi salesmen never forget that.

 

[Ambio]

.

My point is simple. What is this inability to perceive loudness below a threshold limit has got to do with ones ability to perceive what is pleasing to ones ears. Why does one need to know the science behind it to appreciate what is good or what is bad? Id be glad to have your views on this aspect.

It is simple. We hear sound only when it above of hearing threshold. Loudness is just the measurement of sound pressure level. What is pleasing or not depends on what is the pitch and how high and low and the rythmn. If I were to shine light to my ears I am not going hear it. If you hear sound at a constant level than you may hear the rythmn of the beats but a violin would make no sense as you wouldnt hear the difference in the loudness level.

This topic started with the audibility difference in cables. Measurements indicates only slight difference in the loudness drop at higher frequency. If you perceive something to be pleasing with a cable then it is not the cable or hearing but your false perception to hear things which are not there.

.
. Lets accept the fact that the sensitivity and concentration levels of individuals vary. When you also add keenness and interest into the equation, quality of perceptions would vary quite significantly.

Exactly, there are many superhuman in the world. You can watch them in the Ntional Georgraphy channel performing feats that science struggle to explain. One thing that is still misssing - superhuman would could tell difference in cables.

 

[sarith]

Blind testing Level Differnece

Fun test. Up to 1db it was a breeze, didn't even took one minute. But 0.5 took almost 2-3 minutes.

 

[sarith]

Exactly, there are many superhuman in the world. You can watch them in the Ntional Georgraphy channel performing feats that science struggle to explain. One thing that is still misssing - superhuman would could tell difference in cables.

I did try your link 'myhearingtest.net' and made my bro and a friend abroad try it too. Their hearing were around 10. Mine was better though I cannot prove my 'hearing' to them. Btw, I hate the sound of crackers.

 

[arj]

Re: Are CD Player still relevant?

Strange, but when I took that test, I could clearly detect a 1 dB difference and I scored 9/10. And me with my cloth ears!

if you can hear that then those are definitely not cloth ears !
I find it difficult these days

 

[Ambio]

I did try your link 'myhearingtest.net' and made my bro and a friend abroad try it too. Their hearing were around 10. Mine was better though I cannot prove my 'hearing' to them. Btw, I hate the sound of crackers.

Remarkable! but please make sure your soundcard is flat. The result depends on having a headphone and soundcard with flat frequency response at least up to 8 kHz.

 

[sarith]

Remarkable! but please make sure your soundcard is flat. The result depends on having a headphone and soundcard with flat frequency response at least up to 8 kHz.

Thank you. I uninstalled realtek from my pc some time back and been using default windows. I don't like eq. It's too noisy/coloured for my liking. I use my headphone plugged on to my mobo (I don't have a sound card).

I have a dac which is connected to the optical out of my mobo to my amp.