I heard this funny joke years ago:
"One third of Americans suffer from psychiatric illness. So look to your left and to your right. If those people look normal, it is you!"
There is no question that there are critical listeners and non-critical listeners. The problem is that while many of us think we are critical listeners, we have never had that hypothesis checked in an objective way. We may very well be the one who has it, but without some unbiased confirmation, we don't know. Yet we share and treat our opinions of audio equally, industry person or not.
Fortunately this issue has been tested, specifically by Dr. Sean Olive, ex-president of Audio Engineering Society, WBF member, and all around expert in testing acoustic opinions. Two of his studies come to mind:
1. Trained listeners vs not. This is probably the most depressing thing to read for most people here
. That is, compared to people who are trained professionals, most of us don't even come close to what they can do in detecting coloration in speakers/rooms. Fortunately there is software that is free and available to become trained just as Harman people have taken:
http://www.whatsbestforum.com/showt...e-in-How-to-Critically-Evaluate-Sound-Quality. In there you can see how we all do when we are not trained:
Image removed for brevity.
And this:
Image removed for brevity.
Too small to see but a group of these listeners were indeed audiophiles. We can see that trained listeners are far, far harsher on ills of bad sound. As a personal example, I can't listen to US DBS Radio (XM and Sirius) yet they have 25 million subscribers. To me the compression artifacts are unbearable even though I love the content and would subscribe otherwise.
I have sat through a large group of dealers at Harman taking the same test. As I have told the story before, with just a bit of training, I went from level 2 to 3 to level 6 or so using the software above. All the dealers failed to tell equalization changes to the music at level 2 and 3. Sean Olive though sailed past me and I believe their trained listeners can reach up to level 12 before they are qualified for the task. Download the software and take the test. It is easy, quick and humbling.
2. What we use to test matters and matters a ton. People are often surprised at the selection of music used at Harman for double blind acoustic testing. James Taylor anyone? That selection just in other fields, comes from huge amount of testing that detects what the most revealing content is, so that even untrained people have a shot at hearing the difference. Here is a great summary of Dr. Olive's AES paper on this:
http://seanolive.blogspot.com/2010/03/method-for-training-listeners-and.html
And the most sensitive material for testing whether people could tell a +-3 db EQ at different frequencies:
Image removed for brevity.
Pink music followed female *pop* music!
Look at the many reference clips we have that rank way, way low. Piano at #10 for example! And Jazz trio at #20???
As a way of reference, human voice is very hard to compress and hence it is considered a "codec killer." As a result one of the test clips is Suzan Vega's solo voice. More revealing than a million classical tracks. Indeed classical music is not even used to test codecs because it is so easy to compress due to harmonic nature of it! Yet what is critical in codec testing, i.e. solo voices, is not in acoustic space with it ranking #15. I bet your first instinct for many of you is to use classical music to rate any system but that is not the most revealing material.
The danger here is that without using proper material and training oneself, the opinion becomes ad-hoc. So let's not put weight on one versus the other.
So whether it is Peter, or many outraged folks in this thread, the odds are against us, way against us, in having a valid/scientific opinion of what is bad sound. We have a preference and thank heavens, listening tests show that when it comes to acoustic domain, is similar to trained listeners. But most of us are not remotely critical listeners.