advertisement


Art Dudley on blind 'tests'

Status
Not open for further replies.
P

paskinn

pfm Member
About time. An intelligent, passionate, all-out attack on the whole 'blind test' ideology. Contained in Dudley's column in the latest 'Stereophile.' (Nov 2014).
Lot's of real examples of the 'results' such tests achieve (including the infamous Pepsi '' Blind taste'' test.)

A flavour of Dudley's broadside can be gathered from the following: ''time and time again, the use of blind tests in an attempt to quantify human perception has been discredited.......only useful to assuage the ungifted and the insecure, and to validate shoddy or cynical design..''
Real examples are given, from a wide variety of fields. Whatever side of the fence you sit on, it's a great read. Intelligent and passionate. But then Dudley's perhaps the best writer in the whole audio sphere.
You can get 'Stereophile' , through the post, cheaper than UK magazines...and it contains proper measurements of equipment too!.
 
Here we go. No matter what anyone says, I think blind tests are very useful to the listener to put their own sighted impressions into some perspective. Otherwise, it's easy to be led along by others' opinion or expectations and exaggerate every real or potential difference.

But apparently, they often don't prove anything to other people on audiophile forums!
 
I hope the article will appear on the net before too long, then people can judge for themselves.
The point is not to change minds, which is unlikely, but to get some awareness of the issues involved. And Dudley does that well, I think.
 
paskinn said:
I hope the article will appear on the net before too long, then people can judge for themselves.
The point is not to change minds, which is unlikely, but to get some awareness of the issues involved. And Dudley does that well, I think.
Click to expand...

But would you say that if you hadn't known that Dudley had written the article?
 
Julf said:
That might well be. But does he know anything about science?
Click to expand...

I can reliably and predictably change the sound of my system by tweaking this or that and yet apparently there are no measurements that can be made that would differentiate these differences, why is science lagging behind in this field ? Surely there's a measurement that can tell if the frequency range has shifted, that seems pretty basic.
 
sbgk said:
I can reliably and predictably change the sound of my system by tweaking this or that and yet apparently there are no measurements that can be made that would differentiate these differences, why is science lagging behind in this field ? Surely there's a measurement that can tell if the frequency range has shifted, that seems pretty basic.
Click to expand...

If you mean a change in frequency response between one system and another for a given input, then there is, and yes, it is fairly basic.
 
sbgk, science isn't lagging behind, but if you were a manufacturer who figured out how to make an amplifier sound 'nice' without unduly affecting all of the normally published measurements too much would you tell anyone how you did it? The biggest problem with this hobby are those that actually have no clue but adopt an intractable line on what they believe to be right or wrong.
 
No, Art's articles are always entirely subjective. However, many articles conclude with tech test run in the lab by another Stereophile staff member. In many cases the tests vindicate the opinions, in others they do not.

I haven't read the article yet so can't comment further, beyond saying I enjoy Art Dudley's articles, even where disagreeing with his conclusions.
 
I think Art Dudley is a reasonably entertaining writer and will appeal to like minded enthusiasts.
I have been subscribing to Stereophile for decades and IMO he writes bollox. Entertaining bollox, but bollox.
 
Blind and double-blind testing have their place, but I'm starting to think that level matching is the more important variable to control when testing, whether blind or double blind.

Actually, I don't really care. Buy whatever makes you happy, but if you're testing something this implies that proper controls are in place.

Joe
 
Joe P said:
Blind and double-blind testing have their place, but I'm starting to think that level matching is the more important variable to control when testing, whether blind or double blind.
Click to expand...

That was my finding in previous tests, and a good part of the reason I'm planning DBO IV.

Actually, I don't really care. Buy whatever makes you happy, but if you're testing something this implies that proper controls are in place.

Joe
Click to expand...

I agree with the underlying sentiment. The problem I've pointed out in the past is that personal subjective opinion is often presented, or even perceived, as fact.
 
Status
Not open for further replies.


advertisement


Back
Top