- Now, lets address this one.
First, what is the goal of the test? Is it to prove that A is better than B? ... nope, that would require a properly administered test (which by the way is not hard, just requires methodology)
No, our goal is to validate or disprove the claims which include, but are of course not limited to:
- There was a huge and immediate change in the sound
- The bass was extended and tighter
- The highs took on an almost ethereal quality
- The change was not subtle, it was enormous.
- How anyone could not hear a difference, I have no idea
- This "product" will always make a noticeable difference
So, you see, proving or disproving these claims does not require a highly detailed scientific study to attempt to isolate one variable in a situation where many variables can change, often without notice. This is a single variable change, and we don't need a huge sample set of people, we only need 1 person -- the person making the claim, but sure, if you would like to extend it to all people making the claim to increase statistical sample size, then sure ... have at it. The thing is Cleeds, proving something true to a level of reasonable doubt (we are talking a subjective subject) does require a somewhat robust experiment. However, we are not trying to prove something is true, we are trying to prove something is false, and that only requires one case, but if you would like to prove it false with many people, I am all for it :-)
I cannot be expected to tell a difference on a different system. I know my system inside and out, and can tell even the slightest change:
- Great, let's do the test on your system.
You can't tell the difference in a fast switching test. I need more time to listen, X needs to settle, etc.
- Great, no problem. Take absolutely as much time as you want, as long as you don't know which change is made.
You can only tell on certain music.
- You can use any music you want and listen for absolutely as long as you want.
I don't even know why anyone argues this point any more. Every excuse has a solution, and for every solution, there will be another excuse ... including your unfounded difficult to do a scientific experiment excuse ... because as I stated above, it is not. If you or anyone else claims that:
- They can always and easily hear the difference on THEIR system
Then all one needs to do is replicate the test, double blind on their system.
If one claims:
- Anyone with good ears can easily hear the change on a highly resolving system.
Then all one needs to do is allow them to configure a suitably resolving system (of their choice) and then let them pick the people with "good ears" and run the test (double blind). If those people can't show any statistical ability to detect the change, and keep in mind these are people motivated to do that, then you have proved the statement false.
It is totally totally weird how easy it is to set up
something approaching a controlled situation for proving claims (and
increasing sales) in audio, and yet no one does. I don’t mean most don’t
do it, I don’t mean it is not done very often, I mean well never
It
isn’t really clear what you’re saying here, but you’re completely
mistaken if you believe that setting up a valid controlled scientific
listening test is easy. Those that I’ve seen conduct such tests go to
great lengths to ensure their validity.
But of course, that isn’t
quite what you said, you said it’s easy to set up a test that’s
"approaching a controlled situation." Such a test isn’t scientific at
all, of course, so its value is no better than the sighted tests that
you decry.
Once again, we have a contributor here who pleads with
others to conduct the sort of listening tests he himself refuses to do.
What’s with that? Beware the audio ignorant.
