Originally Posted by 67jason
sorry i just dont see it your way. you are building to what amounts to a straw man argument.
how can something (outside of clipping and the like) not be audible in a dbt but be audible in a non test evaluation?
that question is drawn from a statement you made. please explain and provide reliable non subjective evidence in your answer.
and is still like to know what leg you have to stand on to criticize scientific method testing protocols when you stated that is not how you ever evaluate gear. (for what its worth i have put forth my best effort with the resources i have available in order to conduct my own personal tests as well as studied on the topic extensively....and yes i have gear that i chose subjectively, but at least i know why i choose it rather then believing in magic or audio voodoo)
if the currently best available method for testing audio gear with out bias is so wrong or flawed or what have you, please tell me what is better for controlling bias in a controlled test in order to determine real audible differences vs only biased reinforced perceived differences.
No, it's definately NOT a strawman argument. In the first part of my response, I have merely demonstrated truth. A difference is either audible or it's not. As long as there is no proof that a difference is audible, we don't know if it's audible. Similarly, as long as there is no proof that a difference is inaudible, we don't know if it's inaudible, and the only thing that we can do is take a wild guess. Guessing whether a claim is true or false does not magically, and again I quote, "reveal the true fact of the matter". Get it? Good. You have now learned something about how pure science works. Bravo!
Further, the reason why DBT is not necessarily always more reliable than sighted listening when trying to detect differences by hearing, is because DBT itself can create
a set of circumstances that prevents the listener from being able to tell certain differences, i.e. just like Bob Stuart has explained to Robert Harley in the interview. If you don't want to believe that's true, all you have to do to start questioning your firm belief is decide to study psychoacoustics just a little bit.
Now, let's take an unbiased look at how "reliable" the socalled "evidence" obtained from blind listening experiments can be if you know very little about psychoacoustics. Please watch the blind listening experiment that was conducted by psychoacoustician Poppy Crum at the AES Show in 2009 held in New York, it starts 8:10 into this video from Ethan Winer.
The conclusion is that we are programmed to hear speech. Most of us grew up with music, so it is only logical that most of us are also programmed to hear music. (Music that contains vocal lyrics would definitely fall in that category). Either way, this blind listening experiment very clearly demonstrates the fact we are programmed to hear certain things, including certain things that we are not supposed to hear because they have literally been erased from the recording before listening blindedly, as was demonstrated by this experiment shown in the video above. However, obviously, listening while sighted does not magically stop us from being programmed. Rather, the set of circumstances that is created by partaking in a DBT is what can create a certain bias that affects the final outcome of the way we are programmed. Bias can shape our perception of sound.
DBT eliminates expectation bias. This expectation bias is created as soon as we somehow get informed about the sound and / or about what generates this sound. The crux of the matter there is DBT requires human test subjects to listen, and, when we listen we get informed about the sound, obviously. So if we want to eliminate bias, we should never perform the same test more than once. As a result, the way simple ABX tests are conducted by hobbyists and enthusiasts at home using the ABX comparator component for foobar2000 or similar technique, these simple tests are a practical invitation to let bias creep back in through the back door. Because various proponents of these kinds of simple tests have been encouraging people to perform these tests while allowing these people to freely decide for themselves if they want to listen to each test sample more than once, we have yet another major clusterfeck of a problem on our hands here.
Now, also obviously, sighted listening does not magically stop us from getting informed about the sound. During DBT, generally, we just focus more on those characteristics of the sound that our brain is telling us to focus on, which causes our perception to be shaped, and the exact same thing also happens during sighted listening. Only difference, the characteristics our brain tells us to focus on during DBT, are not the same characteristics as the ones it tells us to focus on during a normal listening session the goal of which is not the same as that of DBT. You see, the specific goal that is (or that we think is) the reason why we perform a certain task, is what affects how efficiently we can perform this task. (Even, if we are programmed to perform this task). The fact there is a logical relationship between the goal of performing a task and the efficiency with which we perform said task, was first discovered some years ago by a number of field researchers, and has since been confirmed by psychologists. Neurologists have also managed to undisputedly visualize the resulting differences in brain activity, using modern-advanced brain wave scanning techniques. There is a fairly recent documentary by the BBC that covers this subject. (I forgot what the documentary was called).
Finally, if the most reliable food you had was poo, would you eat poo? Because that is basically what you are suggesting by consistently ignoring that we are not only interested in proving that you can hear a difference, but also that it is more musically satisfying.