Yes, and one concern about the proposed protocol is it targets an unrealistic level of confidence. Has anyone one calculated it?No matter how this experiment will be done, there will always be some who will find faults. That doesn't mean that we should be careless about it.
Well, if Steve Eddy burns in the cables, I have no doubt in my mind that the key is safe. I would suggest that he share it with you or rdf before Andre turns his answers in.
Thank you for your vote of confidence, SY.
Though I see no reason why we can't double key like we were going to do last time.
se
Yes, and one concern about the proposed protocol is it targets an unrealistic level of confidence. Has anyone one calculated it?
0.05 is unrealistic? It's absolutely standard in all research.
AgreedBetter than 0.05. That's pretty standard.
You are correct, but there are certainly limits. I don't know the formula off the top of my head. I seem to remember 16 tries (?) was the limit for 10 correct...Disagree. They do NOT have to be in a row.
Ditto. as long as the gauge and length are sufficient to provide adequate current flow, and high damping, the rest is irrelevant. Conductor conducts. - End of story.I run a mobile disco using mains cables for speaker cables
All good points. And yeah, for ten bucks, why not.... there will always be some who will find faults. That doesn't mean that we should be careless about it. ...
... I know for a fact that some out there will think the organizers cheated. Human nature. ... We don't have the resources to run like that now.
... I know that even if everybody else will not believe the results, us who are involved, knowing we do it honestly, will do. Good use of $10 and a bit of spare time for me.
I think the least of our worries is collusion by the organizers...Well, if Steve Eddy burns in the cables, I have no doubt in my mind that the key is safe. I would suggest that he share it with you or rdf before Andre turns his answers in.
0.05 is unrealistic? It's absolutely standard in all research.
Is that what it calculates to in the proposed protocol?
$879????? Jumpin' Jeebus on a pogo stick!
Scott, we're keeping a list of the people involved; if you're not joking, or even if you are , you can add yourself to the wiki list (see my sig for link).
OK, I will. If people see this through something good will come of it, what I don't know.
I'm in for my $10. How can people read this stuff with a straight face? Cable Cooker, Frybaby.
Did you get the stuff that they cryo'd the cable cooker internal boards and electronics, and that THAT improved the sound of the cables burned in? Wow....
jd
To me bragging about better speaker cables is simple hifi snobbery and doesnt impress me in the slightest.
I run a mobile disco using mains cables for speaker cables and have never ever had a single complaint about the sound.
I'm laughing too.
Yes, but knowing the distribution skews the odds, analogous to Monty Hall knowing which door has the goat.
Fortunately, Andre understands the whole coin-flipping thing and is OK with it.
It is one of the basic questions in statistic; let me cite Wikipedia on that one:
"In probability theory and statistics, the hypergeometric distribution is a discrete probability distribution that describes the number of successes in a sequence of n draws from a finite population without replacement, just as the binomial distribution describes the number of successes for draws with replacement."
Hypergeometric distribution - Wikipedia, the free encyclopedia
An example using the urn model is included. And it doesn´t matter if the draws were consecutive or at once.
rdf is absolutely right- as long as the cable identities are hidden, no additional help is possible.
The listener will even not know about the number of successful draws in between.
Wishes
rdf: Yes, but we haven't specified the number correct for 95% significance. Assuming that we go with true randomization (so no constraint beyond n=10), if memory serves, 8 out of 10 correct meets this criterion.
It is still _9_ correct answers for p<0.05.
Due to the dicrete natur of possible answers and the small sample size the step size is quite large. So the hard 0.05 criterion favours the nil hypothesis, that´s why these tests were called conservative.
Maybe a bit more fuzzy criterion could help, the probability to be successful by chance is for 8 answers 0.055, for 9 answers 0.011 .
I could surely live with 0.055 .
Wishes
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- General Interest
- Everything Else
- Burn In speakercable