Golden Ears.

For those who wonder how good is your ears. I found this abstract from Journal of the AES. Merlin, I think it is relevent to your other thread. Would you like to comment? Does anyone subscribe to this journal? I don't think I have time to go to my local library to check whether they keep this. Those of you does I would appreciate your comment about the quality of the study.


AES Table of Contents
2003 September, Volume 51 Number 9.

Differences in Performance and Preference of Trained versus Untrained Listeners in Loudspeaker Tests: A Case Study
Sean E. Olive 806

The audio industry makes many assumptions about the appropriateness of various quality testing methods, but there have not been any significant studies to validate these assumptions. The choices are reduced to using trained listeners, who are efficient and discriminating, or untrained listeners, who are more representative of the user population. This 18-month study shows that trained listeners produce the same conclusion as 268 untrained listeners when evaluating loudspeakers.


AES link

 

You can download individual AES papers for a fee (I think it's either $5 or $10 a pop). I downloaded one on the subject of jitter once.

OTOH becoming a member of the AES only costs $60 per year ($95 if you want to get the hardcopy journal in addition to the online version that all members get). Anyone can become an "Associate" member without any special qualiifications and the only difference compared to full members (who must have audio engineering qualifications and/or relevant experience) is that you can't vote or hold office.

I'm quite tempted to sign up myself

Michael.

 

It's worth the $60 membership fee...

...just to be able to quote this priceless gem

From just before the conclusions section in the paper Wolfgang references:

It is the authors experience that most of the differences in opinion
about the sound quality of audio product(s) in our industry are
confounded by the influence of nuisance factors that have nothing to
do with the product itself. These include differences in listening
rooms, loudspeaker positions, and personal prejudices (such as price,
brand, and reputation) known to strongly influence a persons judgment
of sound quality [9]. This study has only reinforced this view. The
remarkable consensus in loudspeaker preference among these 268
listeners was only possible because the judgments were all made under
controlled double-blind listening conditions.

And [9] is another one worth reading:

F. E. Toole and S. E. Olive, Hearing Is Believing vs. Believing Is Hearing: Blind vs. Sighted Listening Tests, and Other Interesting Things, presented at 97th Convention of Audio Engineering Society, J. Audio Eng. Soc. (Abstracts), vol. 42, p. 1058 (1994 Dec.), preprint 3894.

It's also worth knowing, that for $500, you can have almost every paper published by the AES on 19 CDROMs... It's a hell of a lot better reading than Stereophile/HiFi+

 

Nice one dat19. LMFAO.

:ffrc:

 

I'd be delighted to comment Wolfgang, if you were to post a full test report that would allow me to judge it's relevence. As it stands, I would have to handover $60 per annum and $4 per paper - nice little earner they have there.

Again, what i asked for Datty Boy was some published evidence. With respect all I have is the author's own personal conclusion of an unknown test involving unknown numbers of people, that asserts that room positioning and acoustics play a significant part in one's apreciation of differing loudspeakers. Hardly newsworthy that one is it?

In addition, it is worth remembering that for most of us, listening to our hifi's is a sighted experience. If the loudspeaker we own looks apalling, is this not likely to influence our enjoyment of the music?

 

I have to say this debate has raged on since I was a nipper....at least over the last 25-30 years. I'm still waiting to see anything meaningful in the way of published material. I've been involved in a couple of efforts in this area but without an sucsess. Having a scientfic background I'd like to believe in the DBT methodology but I remain unconvinced so far.

Personally I can detect difference between kit be it amp through cables within a few minutes of listening, but I know what to listen for now.

 

So you do have golden ears? Congratulations!

 

Re: Re: Golden Ears.

It's just $60 per annum or if you don't want to join there's a fee per paper. As regards "nice little earner", I doubt it. Do you suppose that printing a scientific journal costs nothing? Name me another academic journal that doesn't cost money to subscribe to. I suspect that the AES is run as a non-profit organization.

Regarding the rest of your post, you really should talk to Tony Blair about a job. You make Alistair Campbell look like a pillar of honesty and straight talking.

Michael.

 

Re: Re: Golden Ears.

Am I the only one who has noticed the incongruity of the offensive term 'Datty Boy' followed by the term 'With respect'? Merlin, learn some manners!

 

With respect , I was unaware the term was in any way offensive, still I'm happy to not use it in future.

Why

No one has yet provided me with any documentation whatsoever despite my requests, nor answered any questions to my satisfaction, with the exception perhaps of PeteH.

I am asking simple questions, I am not getting answers other than suggestions that I actually contribute to the furtherence of this mythical and short sighted organisation by paying to read the flawed papers some of their members seem to be producing.

So again, someone show me a text and I will read it with an open mind. In addition, what about the yugly speakers scenario? Still no answers just further attempts at indoctrination!

 

merlin, if you can't see the grotesque way in which you spun the abstract of the paper out of all recognition then I really can't help you I'm afraid

If you're not willing to read AES papers then you've got no right to criticse them based on what you think they might contain. I'd be tempted to download the paper in question and reproduce it here but that would of course be a breach of copyright.

Michael.

 

Tell you what Michael, just email it to me I won't say anything But I do reserve the right to criticise it's content should I find it to be unsatisfactory.

 

Of course what the speakers look like affects how you perceive them, although not always in an obvious fashion - that's why we use blind tests in the first place, to isolate the audible differences as best we can. However, non-auditory effects are more-or-less beyond the remit of the AES, and they're certainly not what we have 20-page arguments about on this forum.

 

merlin - you have mail

I don't think you'll find anything to quibble with. Having skimmed over it it basically says that trained listeners come to the same conclusions about which loudspeakers they prefer but that trained listeners do it more quickly or with greater reliability.

I think the point Olive was trying to prove was that the "training" that trained listeners have doesn't bias them (whether intenionally or not) to prefer different things to untrained listeners.

An interesting detail though is that audio reviewers (who, Olive regards as untrained listeners ) only perform marginally better than Joe Public in being able to rank loudspeakers by performance :torkmada: .

As an aside the paper shows exactly how a properly run test of this kind is done and documented (as would any similar academic paper).

Michael.

 

How fascinating. (mind you not fascinating enough for me to pay for it hehehe)..

I was wondering (for those who shelled out)... what TYPE of speaker was prefered in DB tests?

I ask because my preference are for dynamic expressive speakers, even if that means they may be innacurate..

It would be funny if I picked something else blind!

It would be interesting to discover if DBT's show a preference for say..studio monitors, sensitive speakers, ribbons, horns etc etc

Does it say?

Cheers Chris

 

He doesn't say exactly what make and model of speakers was used "Each loudspeaker is coded with a letter, since the brand name and model were not relevant to the aims of this study". I think not wanting to upset anyone was the more likely reason

However, he does detail the 4 speakers and they MSRPs as follows:

Code:

Code  Speaker type          MSRP
-----------------------------------
B     Three-way dynamic     $8,000
P     Four-way dynamic      $10,000
M     Electrostatic/dynamic $11,000
I     Four-way dynamic      $5,000

He also adds: "Given the relatively high prices of the loudspeakers, they should in theory represent “state-of-art†designs in terms of technical and sonic performance. Indeed, all four models have received high accolades and recommendations from the audiophile press. In one magazine, two of the models (P and M) have received the highest performance category status possible (class A) for the past three years, and loudspeaker M was declared a “product of the year.â€Â"

It's surprising he didn't label the speakers A, B, C and D, there may well be some relationship of the letter to the brand and/or model. With the information given it might be possible to work out what the actual speakers were.

The rankings of the speakers were pretty conclusive:

He also tested without including speaker "I" and that yielded similar results. Speaker "M" (an electrostatic hybrid - Martin Logan maybe?) was consistenly ranked as by far the worst speaker in the test by both the experienced and inexperienced listeners.

Michael.

 

Martin-Logan hybrids are among the worst I've ever heard.

 

Infinity ?

Polk ?

B&W ? or perhaps Boston Acoustics ?

 

Well I must say it's all incredibly boring, again it is amazing how the proponents of DBT "lift" the text that reinforces their preconceptions, just as we do.

Why no mention of the following given the thread's subject?

Now seeing as the thread is titled "Golden Ears" surely this should be brought to light. You see, many peoples contention is that you need to know what to listen for, something that only comes with experience or training.

 

I didn't really hide that finding though did I? I summarized with "trained listeners do it more quickly or with greater reliability".

And to be fair to wolfgang (who I presume doesn't have access to the full paper) the abstract which he started the thread with has no mention of this finding.

That significant differences in performance were noted is not that surprising. However, what's important is that the trained listeners ranked the speakers in the same order with similar relative scores. There's absolutely no suggestion that trained listeners can here differences where untrained listeners can't which is the skill (IMO non-existent) most normally associated with "golden ears".

It means that if you use trained listeners for a DBT you can usually use fewer listeners with fewer trials to get the same results. If you use untrained listeners it won't make any difference to the outcome of the test but you'll have to use more of them for more trials before you get a statistically significant result.

I notice that you subtly edited conclusion 3 to skip the bit about audio reviewers having such relatively poor skills. Here's the full text of conclusion 3 (the bold text is mine):

And yes, it may not beat reading Harry Potter or Lord of the Rings for page turning excitement but what do you expect from an academic paper?

Michael.