 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.69.128.19
In Reply to: It's too bad... posted by mkuller on January 21, 2006 at 11:08:08:
I feel the exact opposite. Blind tests are the only way to get true objectivity. Otherwise set prejudices can creep in.Do you think the blind wine tasting back it the 70's where the French finally admitted that California wines could be as good or better than theirs would have happened if the judges had been aloud to see the labels?
If you can hear the difference with a bit of time, knowing what the equipment is and is supposed to sound like, then why would listening over a period of time, NOT KNOWING, make them sound any different?
If it all sounds the same under blind testing conditions, maybe it does all sound the same.
Wine tastes different so blind tastings only take away any preconceived ideas about a certain wine. Speakers sound different in a blind test because they have audible differences. The blind test just might takes away the same prejudices.
Or do you feel blind tastings of wine are a waste of time? If not, why?
Follow Ups:
> I feel the exact opposite. Blind tests are the only way to get true objectivity. Otherwise set prejudices can creep in.>Yes, it's possible in sighted listening for prejudices to creep in. But forced-choice blind listening *always* removes small differences.
> Do you think the blind wine tasting back it the 70's where the French finally admitted that California wines could be as good or better than theirs would have happened if the judges had been aloud to see the labels?>
Of course not - the French judges had a lot to lose. However an audio reviewer who finds the latest megabuck component favorite to sound awful would be considered a hero.
> If you can hear the difference with a bit of time, knowing what the equipment is and is supposed to sound like, then why would listening over a period of time, NOT KNOWING, make them sound any different?>
Ah, the old simplistic ABX argument. 'Not knowing' isn't the issue - listen to A and B without knowing what they are and you'll hear plenty of differences. It's only when you're forced to chose one as X that things fall apart. Blind wine tastings are not forced-choice. They're just AB, like most audio listening.
> If it all sounds the same under blind testing conditions, maybe it does all sound the same.>
Right, and your Japanese reciever sounds just as good as those Krell or VTL separates.
> Wine tastes different so blind tastings only take away any preconceived ideas about a certain wine. Speakers sound different in a blind test because they have audible differences. The blind test just might takes away the same prejudices.>
Fine, as long as you limit the blind testing to ABAB, like you do with wine. You'll hear plenty of differences.
> Or do you feel blind tastings of wine are a waste of time? If not, why?>
No, blind tasting, or any other type of tasting can be informative. Just like almost any type of audio listenting...except blind forced-choice.
When you say:But forced-choice blind listening *always* removes small differences.you take upon yourself the entire burden of proof. As many people, myself included, have run listening tests in which very small differences were in fact distinguished, your claim is, in fact, falsified.
If you want to rehabilitate your claim, you will need to show the error in many tests, from Fletcher's to the present day.
To the reader: There is no evidence whatsoever, in any way, shape, or form, that "But forced-choice blind listening *always* removes small differences." In fact, the opposite appears to be true from all of the serious refereed work.
I can see where the ABX blind test could be a bit antagonistic. With wine blind tastings, you don't have to pick the best and then see if you can tell it against itself, but you still might go back over the wines to make sure you feel the same. But a blind audio testing could still be viable because you hear which you like best and then you might try again to make sure. But if you can't tell the difference then you should reconsider your stance.As for audio reviewers, they aren't doing blind testings either. They and their mags are paid to find differences. If they can't tell audible differences then they cease to exist. There is also the advertising revenues to be considered.
"Right, and your Japanese receiver sounds as good as those Krell or VTL separates." This is a perfect example of the preconceived prejudices. With wine blind tastings you are trying to eliminate cost, history, previous reviews, reputation, etc. and just taste the wine for what it offers.
Blind testings free your mind and let you judge something on it's own merits.
Aloha, amigos!I think I'm in both camps.
I like blind tasting and listening.
Blind tasting and blind listening don't analogize well, though.
The differences between wines are/can be so vast, that we no longer need to debate whether or not their differences CAN be discerned, it's that blind tasting is designed to take away pre-conceived notions about the qualities we do discern.
I can't think of any wine afficionado worth a hill of beans who claims wines can't be told apart in blind tastings! Well, maybe Consumer Reports...:)
With blind listening, however, this issue has not been fully sussed out.
In wine tasting, we have many many instances of positive controls, but we lack this in blind audio testing. I have yet to see a DBT afficionado demonstrate a test outcome that actually shows differences. Whenever it does, the afficionado blames differences in levels or some other testing artifact. I'd love it if you guys knew of some.
We we really need are blind listening tests that demonstrate peoples' abilities at discerning known/pre-programmed differences before we can rely on any null results in blind listening tests.
Seriously, I've never seen a blind test result that could tell the difference between a coat hanger wrapped in electric tape vs. any high end cable. Blind testing afficonados don't make the claim that those things can't sound different, but they never demonstrate it. They only seem interested in tests with null outcomes. We really need to get this blind testing stuff into the realm of showing what people CAN hear instead of starting at what we claim they can't.
At some point, the discrepancies between two components must become great enough for blind testing to delineate between the two, but that boundary is totally unexplored. We should be going there.
Once that's done, I'd like to see near-field white-noise blind testing to see if people can pick up on any differences. Blind testing may be tough in the musical setting where the signal is changing too quickly for differences to be as easily identified, especially with tests that include instantaneous switching while music is playing. That's a set-up for failed tests.
Then...we also need to validate blind testing "with training" or with "sensitization."
I've had many a wine in blind tastings that a beginner couldn't tell from another, but with some descriptive assistance, the light bulb went off and suddenly he/she could tell the difference easily!
Same with hi fi. When I've taken novices out shopping, two speakers (usually, it's speakers) that were "identical" sounding one minute, became obviously different with a little help at telling them what to listen for. So, if we find a listener in our positive control phase who can hear smaller differences in blind tests than the rest, we would become better listeners if we were to allow him/her to describe those differences. See, back I go to positive controls.
I think there's much fun to be had with blind listening tests, but we need to get off the horse that only allows negative results to ride.
Don't you ever wonder just how much difference it would take to demonstrate positive results? Answering that could make you famous!
Sorry to ramble, cheers, mon amis!
> Then...we also need to validate blind testing "with training" or with "sensitization."...Don't you ever wonder just how much difference it would take to demonstrate positive results? Answering that could make you famous!>You got it. The real issue with forced-choice blind listening tests using music, IMO and I've studied them, is that they've never been scientifically validated and no one knows how sensitive (or insensitive) the tests are.
They appear to be more of a test for the listener than for differences between components, because unlike the medical trials DBTs were designed for, listeners' results improve with training. Medical DBT results don't change with experience.
Tasting wine blind is just like listening to two components blind (with no X to be forced to chose)- you may be surprised by the qualities you identify - and what you like.
| ||||||||||||
| ||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
