Head-Fi.org › Forums › Equipment Forums › Sound Science › Proposal to ban posts that question validity of DBT from Science Forum
New Posts  All Forums:Forum Nav:

Proposal to ban posts that question validity of DBT from Science Forum - Page 6

post #76 of 89
Quote:
Originally Posted by JohnFerrier View Post
How about renaming it the sub-forum at head-fi where it's okay to discuss blind evaluations of audio components?
How about "The Deaf Listening to the Blind"?
post #77 of 89
Quote:
Originally Posted by Lazarus Short View Post
How about "The Deaf Listening to the Blind"?
How about "not so sorry about my wallet"?
post #78 of 89
How about "Lies, Damned Lies, and Statistics"?
post #79 of 89
Quote:
Originally Posted by Ham Sandwich View Post
A CD player acting as a CD player and outputting an analog signal is a whole different beast.
True.

Quote:
That's because the article JxK linked to was testing CD players as CD transports. It would be more accurate to say all CD players acting as transports sound the same.
Yes – but even there I don't agree.
.
post #80 of 89
Quote:
Originally Posted by Lazarus Short View Post
How about "Lies, Damned Lies, and Statistics"?
Let's go with that.
post #81 of 89
Quote:
Originally Posted by Shark_Jump View Post
I actually think its on you JxK as you are the one making the claim. eg. all CD Players sound the same.
So, between what and what is he supposed to find the difference ?
post #82 of 89
Which phrase makes more sense: "Prove the non-existent doesn't exist, or I believe it exists" vs "Prove the existent exists, and I believe it exists" ?

The first request is unsolvable, philosophically. No human being can be omniscient, or test and observe everything in every situation for every possibility. Science just can't do that. Science is about solving tiny parts of the world... "We know this works here at this time under these conditions." The second request, which is what the consumer should be asking, is much more reasonable. Anyone can set up a single experiment with specific equipment and variables and give you an answer, ie: what they observed.
post #83 of 89
Quote:
Originally Posted by Lazarus Short View Post
How about "The Deaf Listening to the Blind"?
In fact it's precisely those who say they hear differences -using their golden ears and whatever other advantage they have- to show that they can do it consistently. It's not a high bar for a result to be statistically significant and be accepted by a universal scientific standard. Just show that the choice responses ('x' is better than 'y') are not random.
post #84 of 89
Quote:
Originally Posted by eucariote View Post
Just show that the choice responses ('x' is better than 'y') are not random.
Define "better."
post #85 of 89
I think if you can't define "better" you're still wasting your money In any case, with audio it's usually whether or not one hears distortion, ie what has the most fidelity. It's simpler to start with just trying to hear a difference though, before you get into complicated issues like that. Baby steps :P
post #86 of 89
Quote:
Originally Posted by boomana View Post
Define "better."
No need to The only important measure in hypthesis testing is the distribution of results (here, choice responses).

Hypothesis testing begins with the assumption that a drug, cable, etc has no effect (the null hypothesis). The positive hypothesis is a statement about how the results will not be random "drug x will lead to reduced mortality rates" or "I can reliably discern between cheap and audiophile cables".

The criterion for rejecting the null hypothesis and accepting the positive hypothesis is just a demonstation of nonrandom results. In the case of cables, the subject can use any criteria, quality, etc. that may be available to their conscious (and as some on this forum stress) nonconscious experience to make choices that demonstrate a real effect.

If the result is nonrandom (p<.05), then the tester can design the next test to elaborate on its reasons "drug x reduces blood pressure", etc.
post #87 of 89
Quote:
Originally Posted by boomana View Post
Define "better."
"Better" means something that costs more.

"Better" can also be applied to something that is difficult to obtain or is out of production.

post #88 of 89
Quote:
Originally Posted by Chef View Post
What you seem to be preaching is not doing any kind of testing at all, but still spending money...
I'm not sure how it's even possible that what I wrote could be interpreted in this way. This is the exact opposite of what I've been saying.


Quote:
Originally Posted by eucariote View Post
No need to The only important measure in hypthesis testing is the distribution of results (here, choice responses).
Hypothesis testing is not where the discussion begins and ends. You're only considering the statistical analysis of the experimental data, not the rest of the experimental protocol!
post #89 of 89
Define Better?
1. If A is of the same quality as B, but substantially cheaper, is it better?
2. If A is substantially of greater quality than B, but also substantially more expensive, is it better?
3. If A is of the same quality as B and the same price as B, but more practical, is it better?

It all depends on the reference as to what the "better" refers to.
Does it refer to sonic fidelity, price ratio to cost ratio, usability, or any other criterion.

I think the problem here is that we tend to take our "spin" and tout it as the most important characteristic/viewpoint/concept/observation/experience/etc, etc in the discussion.
............ and then the flame wars begin..................
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Proposal to ban posts that question validity of DBT from Science Forum