This is not directed toward you mr.sukebe but ABX tests are doomed for failure at the start imo, way too many variables and inconsistencies to be considered reliable.
They're a great way for any audio company to prove that their product makes a difference. You bring out a product, get a group of people together and get them to demo the kit with an ABX comparator. Publish the results and you have proof that you can happily use in marketing material that meets ASA guidelines (note how audio products are advertised at the moment). We could wonder why audio companies don't do that already, but I think we can draw our own conclusions from that.
Anyway, what do you mean they have too many inconsistencies to be considered reliable? How are such tests inconsistent? Could you provide examples of some tests in which you thought the protocol was insufficient?
The best thing anyone can do is demo cables or buy from someone who will accept refunds if you're not happy, then you're under no pressure to hear anything that may or may not be there.
So, you think double blinded studies are inconsistent, so instead we should use no protocol whatsoever and just live with cognitive bias which will always be present and therefore make the results of such tests meaningless? I can't see your logic there.
Nps and to a certain extent, I agree with you.
For all that, it's usually the people who think they've got a point to prove who demand blind tests. I just find it interesting that when you agree to them, it all goes rather quiet.
I don't know why this seems to repeatedly get missed by yourself and others.
No one has ever proved they can hear a difference with magic cables and whatnot. That's why, when someone repeatedly says that they can, it is met with skepticism by "trolls"* like me until that person can prove otherwise. It's not down to the skeptic to prove that the person is incorrect for it is impossible to do so.
Now, with regards to a cable test - you're right in that i didn't get back with a suitable testing protocol to use instead of ABX, but this does go back to the point above. The best bet for doing such a test would be for some friendly audio nerd to build a comparator, otherwise it's v hard to do a fair test. Remember, this test is for you, so that you can then say "I
can hear a difference". If you failed to demonstrate that you can tell a difference then my position remains the same as it was before the test because i am not the one looking for evidence.
*9designs' label