Interconnect and Mains Cable Comparison - Blind Test Results Are Here

[Anton90125]

[quote user="Cablearser"]While 80% could be achieved by chance I consider it to be unlikely.[/quote]

So 80% is an arbitrary figure not based on any scientific/statistical derivation only what you consider "unlikely". Is there a danger that you may be making a subjective assessment of the type you despise when others evaluate cables? If not why not?

 

[Anonymous]

[quote user="Anton90125"] So 80% is an arbitrary figure not based on any scientific/statistical derivation only what you consider "unlikely". [/quote]

We are talking about random chance here but yes, you're right, I was being far too easy on them. If the differences are really as marked as some on this forum assert I would expect a much higher percentage, close to 100% accuracy.

[quote user="Anton90125"]Is there a danger that you may be making a subjective assessment of the type you despise when others evaluate cables?[/quote]

None of them even came close so it's irrelevant. Choose any figure you like.

 

[Anonymous]

[quote user="Anton90125"][quote user="Cablearser"]If there was any difference at least one person should have been able to score above 80%.[/quote]

Why 80%? Statistical significance will be dependent on the the sample size, the number of tests and the nature of the test.
[/quote]

For an experiment to have real credibility, from a purely scientific perspective, it needs to stand up to the 90 percent, or even better 95 percent of probability (that is that the exact same experiment can be repeated 10 or 20 times and the same results are obtainable). However, that said, they made a good fist of it. Coming from a strong scientific background I'm sure there is a way of designing a credible experiment on sound improvements (or not) based on a pool of results from members from this forum. The first big question would be, whatever the results, is that will they make any differences to what people want to spend their money on?

 

[Anonymous]

It would be nice but at what point you say that you've enough examples to factor out the source mains quality I'm not sure, since we've no metric to go by for how variable this actually is! The results would also be very open to multiple interpretation because of this second variable. So while I'm sure it could be done, I think it would be quite hard to do well.
But yeah, they did really make rather a fist of it didn't they

 

[Anton90125]

[quote user="Cablearser"]We are talking about random chance here [/quote]

If we are talking 100% random chance of a bi state result then the expected results should tend towards 50%. The more samples, the closer to 50%. Try flipping a coin 10, 100, 1000 times you will see that it will tend towards 50%.

However if there are any other factors coming in to play (like a sonic differential between cables, some people more sensitive to them then others etc... ) then it would move from 50%.

Or to put it another way - if you were to flip a coin 1000 times and had a result of 70% ,you'd suspect that you were dealing with a doctored coin wouldn't you? Or would you still require 100% result?

[quote user="Cablearser"]
If the differences are really as marked as some on this forum assert I would expect a much higher percentage, close to 100% accuracy.

[/quote]

So would I if everyone had identical perfect hearing but off course they don't.

[quote user="Cablearser"]None of them even came close so it's irrelevant. Choose any figure you like[/quote]

The simple point I was making is that with some people getting 60% & 70% accuracy, it would be interesting to further investigate this. Since no ones hearing is perfect I would not expect 100% but if these people can maintain 60% & 70% accuracy levels after several weeks testing then this would indicate there was something there beyond pure guess work and placebo effects.

 

[Anonymous]

[quote user="Anton90125"] if these people can maintain 60% & 70% accuracy levels after several weeks testing then this would indicate there was something there beyond pure guess work and placebo effects.[/quote]

A bit of luck?

 

[Andrew Everard]

[quote user="Cablearser"]
None of them even came close so it's irrelevant. Choose any figure you like.

[/quote]

This repeated ban-dodging is becoming very annoying, and your username is offensive.

Please desist, and go play elsewhere...

 

[Anton90125]

[quote user="Mythbuster"]
[quote user="Anton90125"] if these people can maintain 60% & 70% accuracy levels after several weeks testing then this would indicate there was something there beyond pure guess work and placebo effects.[/quote]

A bit of luck?
[/quote]

Or we could admit to reality and accept that something was there.

Or as the great Douglas Adams said you could prove black is white but then you might get run over whilst crossing a zebra crossing !

 

[Andrew Everard]

[quote user="Mythbuster"]
A bit of luck?

[/quote]

And more annoying ban-dodging. Please stop now.

 

[Thaiman]

No surprise to me that why some hifi forum ban cables debates!

 

[Anonymous]

[quote user="Anton90125"] we could admit to reality and accept that something was there[/quote]

We should accept the result of all blind tests carried out this far. That it's all nonsense.

 

[Andrew Everard]

[quote user="Thaiman"]

No surprise to me that why some hifi forum ban cables debates!

[/quote]

Sorry, going to lock this one - thanks to the usual trouble, taking up too much of our time when we should be helping other readers, working the Bristol Show, etc...