Looked like these could be "forum" ABX tests to me, has Jakob2 passed judgement on these?
Yes & I presented them for those who believe in the "scientific validity" of forum (means Foobar) ABX results.
The funny thing is that for a test which is so skewed towards null results, it takes a great deal of effort to overcome this skew & get a statistically positive result - not the normal null that results with those who don't have this level of discipline, tenacity & experience this guy has.
Why don't you ask him?
My wishes to you. Please report back.
Till now, I haven't good impressions from any CD remastering I heard compared to the initial CD release in classical music and opera (Sony included)
George
A sighted test does not test what people hear, but also what they see. Therefore it is not a good test of what people hear. What is difficult about this?mmerrill99 said:
Please explain my "misunderstanding" and "misconceptions". Do you hear with some organ other than your ears?
So if hundreds of tests all find the same result then this should be taken as evidence that the tests were flawed, not that the result is true? I'm sorry, physicists would not make much progress in understanding the world if we followed such poor advice from statisticians. If you calculate a daft number then your initial assumptions must have been wrong.Jakob2 said:
Any test can only say that no difference was perceived in that test. It cannot say that no difference can ever be perceived in any test. However, if some weight of evidence builds up then we can say that a difference is unlikely to be perceived in other tests.
People claim to hear differences when they know what they are listening to, and in some cases claim that these are clear 'night and day' or 'wife in the kitchen' differences. They often fail to hear differences when they don't know what they are listening to, even though the other test parameters are similar. The obvious conclusion is that it is the 'knowing' which makes the difference, not the hearing. This does not preclude the possibility that someone somewhere has sufficiently good hearing that they may be able to discern a difference without 'knowing', but such people seem very rare.
It is not surprising that those who make a living from selling audio, and those who claim 'golden ears', and those who have spent lots of money buying equipment, will wish to denigrate all such tests.
If there is an obvious difference then ABX will show this. If there is a subtle difference then ABX will show this. If there is little or no difference then ABX will show this. What is this 'skew' that you claim, unless of course you have already decided beforehand what the result ought to be (presumably the same result as sighted tests)?mmerrill99 said:
Actually, its not what they see, its what they know or believe. 'Sighted Listening' is the just the term we have been using to refer to the effect. Psychologists might call it 'primed listening.'
Last edited:
No. Not in general, it would depend on whatever other information may be available. The tests might be inclusive. If hundreds of tests show that lead cannot be turned into gold, that might or might not means the tests were flawed. Is the conclusion supposed to be confined to results based on use of chemical means? If hundreds of test show that gravity waves can't be detected, does that prove they can't be?
DF96, in your rambling you manage to say some things that are true and some things that are false. You also conflate things that should not be, such as
'wife in the kitchen,' and 'sighted listening.' If you would stick to things that are true then you would get much more agreement from some people.
Last edited:
If it is obvious, they why do we need ABX? We already know if something is obvious.
There is more than one thing at issue. Sometimes people exaggerate claims of obviousness, and ABX might be used to help show something isn't all that obvious. Sometimes the issue is whether or not small effects are audible at all, in which case insistence on using ABX is probably a poor choice likely to produce inconclusive results due to its well known insensitivity. When we do tests or experiments, we need to be clear in advance on what the experiment is supposed to be designed to show.
Another possible complication: Earl Geddes said that new test methods would have to be developed for the 5%.
Is it just me, or is there a whole lot of repetition going on? C'est la vie.... still, no ones getting hurt....so far. If it hurts them in their wallets more fool them, sorry can't agree with spaceistheplace on that one, I'm not in the business of saving people from themselves, in fact people spending money on useless stuff is the bedrock of western civilization.
We need ABX to exclude self-delusion and bias. Some people do not like ABX because it does not bring the expected result (the result they expect and wish to get) and they are upset of this situation. You are one of them.
Bill: Who may or may not be in the 5% can only be determined by careful measurement. But, please let's not refer to 'mystical' abilities. My guess would be we were getting closer to the truth yesterday: We measure one thing, but sometimes may be hearing something other than what was measured. When people claim to hear something, chances are if we carefully investigated we would probably find most claims can be explained by some mix of sighted listening and hearing of effects, or side-effects, that weren't specifically measured.
Regarding the last point above, when I adjust harmonic distortion compensation in a dac by a very slight amount and hear a difference it is very likely I am not hearing what I was measuring, which was HD, but rather a change in the IMD profile at a much higher level. Why do I think that? Because I can hear it. If all the effects of an adjustment were confined to -120dB then I wouldn't hear it. In this case I think I can rule out sighted listening effect due to the methodology used during testing.
Regarding the last point above, when I adjust harmonic distortion compensation in a dac by a very slight amount and hear a difference it is very likely I am not hearing what I was measuring, which was HD, but rather a change in the IMD profile at a much higher level. Why do I think that? Because I can hear it. If all the effects of an adjustment were confined to -120dB then I wouldn't hear it. In this case I think I can rule out sighted listening effect due to the methodology used during testing.
- Status
- Not open for further replies.