Technically, it is more an issue of False or not. Please see Ioannidis: Why Most Published Research Findings Are False
From the arguments I keep hearing in forums, it was not insignificantly AES article writers. That's the kind of stuff the so-called objectivists seem to consider as rock-solid facts (even though they often do not read the papers or if they do they usually don't fully-understand/overstate the conclusions).
Also, you may recall I once expressed concern after watching a youtube video of an AES lecture by a presenter you described as highly regarded in the AES. At the end, in the question and answer section, he said, 'you know, people hear what they expect to hear.' I said I would like to sit down for a face to fact talk with that guy, think he needs to catch up on cognitive psychology research as it applies to perception. Its not the same thing as engineering know-how applied to perception research. IMHO, one needs to understand both (and more).
Last edited:
Buried in noise? RC ladder equivalent networks in caps, that may vary with time and voltage (a ladder only for fixed sine waves maybe)? Then measure them if you can't hear them, instead of insisting people are imagining them.
Last edited:
I don't believe that is possible as you are looking at mutually exclusive things there.
Bill,
A rather sappy example, but an example nonetheless.
Altered States of Consciousness | Psychology Today
A rather sappy example, but an example nonetheless.
Altered States of Consciousness | Psychology Today
I don't see that as an example. But when I look at how Mark describes his methods for detecting tiny differences (similar method used by PMA on his recent preamp test) this is diametrically opposed to what you are saying. You can't be concentrating on one tiny transient such as a cymbal strike over and over whilst letting go.
Yes, even buried in noise.
I'm afraid these no longer qualify as "linear distortions", not even close. And Fourier is still valid, whatever you may think about.
You keep making up arguments. Do yourself a favor and check against a physics manual before posting.
Last edited:
It was an example of/for possibility......not an explanation.
Spitballing is the proper term!
Last edited:
True, sometimes they are not linear. They can look that way on a 1kHz FFT if the ladder behavior reaches a stationary point. The remaining non-linearity around that point may be very, very small.
Why do you think KSTR performs measurements using music rather than fixed sine waves? He needs to read some physics books too?
Look, I know I made some speculation about whatever is behind capacitor sound. The fact remains some people hear it. In more extreme cases it can even sound about like HD or IMD sounds to the ear. With good caps, it is almost certainly very small linear distortion, but audible to some.
That doesn't make it easy to confirm audibility by perceptual research. Some people are reliable expert-level skilled listeners, most audiophiles are far from that (and roughly 1/3 have other issues as we know). It means if we want to do research about the most people can hear in a certain dimension (say, for example, distortion), we should start by identifying and using expert-level skilled listeners relative to that dimension. Using audiophiles, college students, AES attendees, etc., will only add noise and confusion to the effort.
P.S. I'm about done talking for now. Have fun going after my comments if nothing more entertaining happens to come along
Last edited:
I assume this would apply even more strongly to poorly controlled self administered listening tests.
Some people claim alien abduction. Doesn't mean it's real. You are quoting anecdote as fact. Exceptions are filters where the effects are readily measureable and even Douglas Self admits that.
Last edited:
Look, I know I made some speculation about whatever is behind capacitor sound. The fact remains some people hear it. In more extreme cases it can even sound about like HD or IMD sounds to the ear. With good caps, it is almost certainly very small linear distortion, but audible to some.
That doesn't make it easy to confirm audibility by perceptual research. Some people are reliable expert-level skilled listeners, most audiophiles are far from that (and roughly 1/3 have other issues as we know). It means if we want to do research about the most people can hear in a certain dimension (say, for example, distortion), we should start by identifying and using expert-level skilled listeners relative to that dimension. Using audiophiles, college students, AES attendees, etc., will only add noise and confusion to the effort.
P.S. I'm about done talking for now. Have fun going after my comments if nothing more entertaining happens to come along
Where is the proof? We can measure sub 1 per 100 million distortions. No need to use humans to determine distortions.
What do you imagine companies like Harman do? They do work with trained listeners. The M2's you have are a direct result of that effort.
If you are bothered by the BR sound, stuff the pipe as per JBL's recommendations. Your other comment on the sound was right on the mark, no pun intended.
The reason is the justification found in the phrase just above (“A 2 standard deviation from normal covers 95.44% of population”). The adoption of the two sigma has been proven a robust consensus since now. A three sigma limit requires a whole lot more resources, as well as the study of the “outliers”. Is there a good justification for that?
In the rest of your post, I think you got it twisted.
In a study designed and conducted properly, you don’t expect 1 out of 20 to be wrong simply because of random chance, you would accept 1 out of 20 to be outside the two sigma bell curve for various reasons.
The real meaning of the other phrase escapes me (“Its because 5% is estimated to be the chances of a study being wrong on statistical grounds only, nothing to do with being wrong for other reasons”)
George
Attachments
- Status
- Not open for further replies.