Interconnect Directionality

The trouble I have with blind tests or any tests really is their inherent fallibility. So many things can go wrong. If the results are negative it could very well be for any of the following reasons. This is not intended to be a complete list.

The system has one or more errors in it.
The system is not revealing enough.
The subject is not capable of hearing the difference.
The test software is out of polarity or just plain not good enough to reveal differences.
Cabling and or electronics are either brand new or not broken in or not earned up properly.
Weather issues mask the audible differences.
Tests can be easily rigged to produce negative results.

On the other hand, if results of a test are positive, I might be more inclined to think something might be going on since positive results were obtained IN SPITE of all the pitfalls. But, generally to be convincing tests should be repeatable and transferable. One test has very little significance especially if the results are negative. If no tests are done at all then someone making bold claims that controlled blind tests will prove such as such is just plain absurd. 

- I am willing to test on any system you wish to configure. Based on this assumption every person who makes a claim has errors in their system
- I am willing to test on any system you wish to configure, hence if it is not revealing enough, that falls on you. 

However, since you are making the claim (as have others), then their systems would make the most sense to test on, and/or a manufacturers system who is also making the claim.

- The subject is not capable of hearing the difference: .. since it is the subjects ability to hear the difference claimed that is being rejected, I would say that is often a moot point. In the context of general testing, you normally do larger sample sizes and most who would participate in these define themselves as audiophiles.

- Cabling and or electronics are either brand new or not broken in or not earned up properly.  See point one. You (or the vendor) has complete flexibility in the system used.

- Test s/w is out of polarity or not good enough to reveal the differences. This speaks back to my point about the person making the claim to hearing differences being the ones who is being challenged. They have already made that claim on a range of S/W, so do the test with the same S/W.

- Weather issues mask the audible differences. If weather issues mask the audible differences, then how can you be sure you even heard a difference due to the component change in the first place and not due to a weather change. However, since this is a somewhat quick A/B test, both scenarios experience the same weather. We could also assume indoor and somewhat climate controlled.




geoffkait17,697 posts10-24-2019 12:15pm"The trouble I have with blind tests or any tests really is their inherent fallibility. So many things can go wrong. If the results are negative it could very well be for any of the following reasons, which is not intended to be a complete list.

The system has one or more errors in it. 
The system is not revealing enough.
The subject is not capable of hearing the difference.
The test software is out of polarity or just plain not good enough to reveal differences.
Cabling and or electronics are either brand new or not broken in or not earned up properly.
Weather issues mask the audible differences.

On the other hand, if results of a test are positive, I might be more inclined to think something might be going on since positive results were obtained IN SPITE of all the pitfalls. But, generally to be convincing tests should be repeatable and transferable. One test has very little significance especially if the results are negative."

geoffkait, you have again advanced the henny penny argument that "the sky is falling".  That all the small chances of something not being correct in the double blind test outweigh the much larger chance that nothing will go wrong.  In order for all the things to happen that you say could go wrong there is the unwritten inference that one of the test sponsors has something to gain from a certain choice and will sabotage the test to achieve his advantage.

jetterI have no ideal why what I believe is a small group of persons on A'gon would think that blind testing won't yield fair and impartial results. Please do not repeat the GK arguments against it that the testers are going to rig the test.

Nothwithstanding the fact that the conclusions are sometimes dubious, there have been many perfectly valid listening tests. Because it's such a time consuming task from the onset, organizers who actually go to the trouble of conducting such tests typically put extra effort into making them valid. That's pretty well established.