you cannot trust your eyes, so why trust your ears? - Page 2

No. In fact there are several good reasons that tests, whether to evaluate a component or to determine if two components sound the sound, should be done blind.

I've edited in a correction to my story. Got my Greens and Blues mixed up in the sentence where the spectrum analyser measures the the headphone output. Completely changed the meaning of that sentence - sorry and just shows how easy it is to get in a muddle when comparing colours (or sounds).

My further point is that many types of illusions are subconscious and therefore can still affect your judgement whether or not you can see the audio equipment. So, although there is no reason to need to see the audio equipment, you shouldn't assume that blind testing will magically cause all those subconscious illusions to disappear. As I've said before, placebo is the only illusion that blind testing will eliminate, but the illusion demonstrated in this thread is not placebo.

You need a very expensive graphic design monitor with expensive power cable and expensive power conditioner and have the monitor calibrated. That or print it out. The "optical illusion" is compounded by a lot of color fading and dirty power, printing out the optical illusion onto paper makes it much easier to see through. Here's a hint: the best way to see through it is to keep shifting up and down up and down constantly on an apparently "blue" line, keep it up and it will turn green again. The reason why this optical illusion works is because it causes people to stare.

When I listen to music, I like to hear something that is pleasing to my senses. Isn't that the whole point of listening to music? I trust my senses because they are what matters to me when I'm trying to enjoy something, whether it's music, food, or whatever.

Of course. But should you pay $500 for a cable when a $3 one gives you the same pleasure?

That extra $497 can then be spent on another item that you will enjoy.

But if you're into cables it doesn't, that's the whole point. The expensive will give some people extra enjoyment which the cheap one simply can't deliver. If that extra enjoyment is worth $497 to the person in question then the $500 cable will be a good investment - by definition.

Of course not. But the question is what method to you use to determine whether it gives you the same pleasure? Do you (1) trust your own ears, (2) trust someone else you don't know who tells you he has listened and it sounds the same to him, or (3) trust someone else you don't know who hasn't listened but tells you he has researched the issue on the internet and the two cables (or amps, DAC's whatever) must sound the same?

I like to give some consideration to (2) and (3), but in the end, when determining what pleases my senses the most, I think it makes more sense to go with (1).

The method would be to do a proper comparison using an ABX switch and a double blind test in order to avoid any bias like aesthetics or to justify any expensive piece of equipment, and to understand the scientific data that backs up that there is no audible difference between those two cables.

(To explain it in a concentrated sentence)

I agree exactly. To say that "we can't trust our senses" is odd when the whole reason audio exists is to please our senses.

I agree that I shouldn't spend more money than necessary on a cable, and I don't get much pleasure from the appearance of my cables so I really want any expenditure to matter to the sound.

What's odd, though, is that so often we hear the statement "We can't trust our senses." Your very signature, Uncle Erik, contains a statement implying that only measurements, not senses, can be used to improve something (which, if it were true, would put instrument builders in the 17th century in a difficult position).

Now, if the "objectivists" mean to say something like "You need a double-blind test to distinguish accurately between components," then why not say that? Why say, as the OP did, "You can't trust your ears?" Because the latter statement goes much further than the first, and implies a kind of epistemological stance that I disagree with. In fact I find it absurd.

That's your preferred method. Not everyone accepts the assumptions inherent in your decision to favor that method, and not everyone wants to spend the time to conduct a blind test every time they must decide between component A and B (and we're not just talking about cables, the typical "whipping boy"). Feel free to conduct all the blind tests you want. Me, I will generally (but not exclusively) trust my ears, and will continue to suggest that others do the same.

tbh Phil, that's how all science is done. Isolate the variable and keep all other variables the same. One way to keep all variables the same except for the one you are measuring is to eliminate those variables and that's what happens in a double blind/ABX test.

I understand how it's done. The issue is whether you have to do one, or should do one, every time you are trying to determine whether two things are really different to your senses.

I'll agree with the double-blind part, but I don't think using a switch is a proper method to test differences in cables.

You say that you've kept all variables the same except for the one you are measuring. This assumes that human brains are devices which behave in a reliable and repeatable fashion--- as though they were measuring instruments.

In reality:

• Context affects perception. Change the context from "normal listening for enjoyment" to "quick-switched small samples" and you've failed to control a variable.
• Music is not meant to be heard repeatedly. Use a test which repeats the same snippet many times, and you've failed to control a variable.
• Use a test signal that has little resemblance to music and you've failed to control a variable.
  

ABX itself is fine, but sense can very from person to person by a lot. The logical way to minimize the impact of variation is to carry out the ABX with a very large sampling population, preferable all different classes of the society from young to old and from casual listener to audiophiles alike.

Now, even if such an experiment can be carried out, the inherent nature of any scientific experiment is not to provide an absolute answer, but merely a explanation of what is closest to (or 'best fit') the truth as we know it. That explanation doesn't need to apply to every subject in the sampling population, it only needs to hold its water for the majority - that said, it doesn't mean those who are just right outside of the 95% or 99% significance must have invalid data, opinion, or sense. It only means the current set of explanation is not perfect enough to cover their part of the story.

There lies the biggest of all problem about ABX on small or individual scale - How do a person know which side of the bell curve he/she fits in? When a small group of subjects are measured, the standard of deviation can be way off the chart thus making the result / conclusion insignificant to the public. However, this doesn't in any way suggest that the obtained data is invalid (though the conclusion that drew from it might be). The data is always valid as long as the design of the experimental procedure is valid (that is, you are measuring what you want to measure and not something else).

Here we come full circle: Assuming everything else is the same, if a person is sensing 'A' from an object while the rest of the 9 person are sensing 'B', it doesn't mean the first person is wrong - it just means the first person's data doesn't fit into the group and doesn't support the theory you might have explaining why most of the group are getting a particular result. Do you think by telling the first person that his sense is not the same as the rest of the group, he must have sensed the wrong thing? That is definitely not the case since we already make sure they were all sensing the same object to begin with. 'A' is as real to the first person as is 'B' to the rest of the group. So if you tell the first person not to trust his / her sense because he / she is different, what is the point of the experiment? You could have just fixed up the answer at the first place.

Listening to music is not about finding the truth of the whole world, but only the truth about the listener. If you feel sad after listening to a piece of music, do you really need to make sure the rest of the world won't call you a fool because you get affected by some sound? Do you really need justification?

As Kurt Cobain so eloquently put it, rape me.