2. I don't buy it. No one goes "up" from AES to J. Neurophys. after rejection. What did Sy have to say? Yes, please back up any of this.
3. Pretty much all studies contain flaws. It's not a zero sum game, rather one decides which flaws are minor and which destroy credibility. And doing so in an honest way, not depending on whether you "like" the results.
Makes sense.
Wait.... oh no..... SAM2? Am I mmerrill100? Are we both part of the Jakob(x) collective? Do me a favor and go to the mirror... do you see me?
lol...
You will be assimilated into the mmerrill-jakob2 collective. Resistance is futile!
And here's a reminder, for Jakob and everybody who would like to follow on this:
Oohashi reported in Jakob's much beloved 2000 paper the results of experiments that he claimed are proving people can perceive ultrasonic sounds, thus confirming for both audiophiles and sellers of “high definition” products that CD quality audio is crap. Oohashi did an (unintentional, I guess, although the convenient (for the "high end audio" aficionados) conclusion could make someone believe otherwise) fatal mistake: he used only one loudspeaker to play multiple ultrasounds at once. The inherent intermodulation distortion in the tweeter created products that are well in the audible range. The Oohashi experiment was repeated years later by Kiryu and Ashihara using separate loudspeakers, and as a result none of the test subjects were able to distinguish the ultrasounds. A random quote from their work:
I believe it is now clear why the Oohashi study was rejected for publication in all audio technical journals of the time, until it got hosted in an rather obscure "Journal of Neurophysiology" where I can bet none of the reviewers and readers ever heard about the "tweeter" and "intermodulation" concepts.
P.S.
- AES 110th Convention: Session P possible there was a full paper in the AES Journal. Can't look for that, since it is behing a pay wall.
- Full paper in Acta Acustica united with Acustica 89(3):540-546 · May 2003
- Related: “Hearing thresholds for pure tones above 16 kHz”, The Journal of the Acoustical Society of America 2007, Volume 122, Issue 3, pp. EL52-EL57
Last edited:
I wrote the same reasoning to Jakob2 once or twice, without an effect.
Attachments
Last edited:
That's your flaw, since you were expecting one
I've also posted Dave Greisinger's presentation on this topic several times. It it also worth asking anyone testing sources with extended high frequency content if their speakers are up for it.
Last edited:
They often are small, sometimes very, very small.
Still, not hyperbole. For people who hear the small differences, attention tends to be pulled towards the small things by parts of the brain operating outside of conscious awareness. The brain presents the small differences as outsized defects to conscious awareness, and so they seem large as they are mentally experienced.
It might be something like a native speaker of some language being able to identify what city someone else originally came from due to sometimes very small differences in accents. A non-native speaker of the language might never become expert enough at it to notice such small differences. Of course, all analogies fail at some level if picked apart, they are only useful as very simplified models.
Don't give up that easily
No, the individual instruments/sounds may have been panned using a time difference between channels to emulate ITD along with normal level panning (or not)
You can measure phase through the system but the chances are the phase relationships between different elements of the recording would be all over the place anyway due to miking, mixing, EQing etc. Simple old recordings maybe better in that regard.
Measurement tells you the real thing in real time. Simulation anyhow, can show you more regarding how things work. When you make changes during tweaking (adding wool into enclosure, adding weight to speaker cone, adding capacitance to notch filter, etc.), you should know what parameters are affected. Simulation can help you here.
There are misunderstandings about phase. But absolute phase imo is not (very) important.
@ syn08,
The facts are:
-) Ashihara & Kiryu did _not_ repeat Oohashi et al's experiment, instead they carried out a substantial different experiment
-) Nobody knows if the tweeter in Oohashi's experiment produced any audible intermodulation products below 20 kHz, because nobody replicated their experiment and did measurements that the original authors did not execute.
They considered the possible intermodulation effect and _therefore_ used already a seperated "super-tweeter" that solely reproduced frequencies above ~21 kHz.
-) the measurements published in Ohashi's showed no intermodulation artifacts below 20 kHz (admitted that the resolution was quite crude and that there might nevertheless have been some)
Edit: when the high-frequency content was reproduced alone none of the listeners could detect it, which is imo a fact opposing the assumption that audible intermodulation distortion was present below 20 kHz
-) the level of the "ultrasonic content" reproduced by the super-tweeter was well below the levels that A & K used in their experiment
-) A & K did not try the same configuration like Oohashi (not the same kind of super-tweeter and not the same kind with baseband reproduction via one loudspeaker and ultrasonic content over another one with comparable levels). Did I already mentioned that they carried out a totally different experiment?
Recently I listed some standard excuses for not accepting published evidence, "the reviewers weren't qualified" line was at the fourth line iirc...
Don't get me wrong, the review system has its own problems, but given the list of facts that you somehow did not report, it is obviously not justified to attack the reviewers in this case.
@ PMA,
That's what you've posted as argument:
My gentle pointer that Oohashi et al _did_ _not_ use a speaker with "one" driver, had obviously no effect.....
The links to the last posts:
John Curl's Blowtorch preamplifier part III
John Curl's Blowtorch preamplifier part III
As said before, friends of surrealism must love these forums discussions......
The facts are:
-) Ashihara & Kiryu did _not_ repeat Oohashi et al's experiment, instead they carried out a substantial different experiment
-) Nobody knows if the tweeter in Oohashi's experiment produced any audible intermodulation products below 20 kHz, because nobody replicated their experiment and did measurements that the original authors did not execute.
They considered the possible intermodulation effect and _therefore_ used already a seperated "super-tweeter" that solely reproduced frequencies above ~21 kHz.
-) the measurements published in Ohashi's showed no intermodulation artifacts below 20 kHz (admitted that the resolution was quite crude and that there might nevertheless have been some)
Edit: when the high-frequency content was reproduced alone none of the listeners could detect it, which is imo a fact opposing the assumption that audible intermodulation distortion was present below 20 kHz
-) the level of the "ultrasonic content" reproduced by the super-tweeter was well below the levels that A & K used in their experiment
-) A & K did not try the same configuration like Oohashi (not the same kind of super-tweeter and not the same kind with baseband reproduction via one loudspeaker and ultrasonic content over another one with comparable levels). Did I already mentioned that they carried out a totally different experiment?
Recently I listed some standard excuses for not accepting published evidence, "the reviewers weren't qualified" line was at the fourth line iirc...
Don't get me wrong, the review system has its own problems, but given the list of facts that you somehow did not report, it is obviously not justified to attack the reviewers in this case.
@ PMA,
That's what you've posted as argument:
My gentle pointer that Oohashi et al _did_ _not_ use a speaker with "one" driver, had obviously no effect.....
The links to the last posts:
John Curl's Blowtorch preamplifier part III
John Curl's Blowtorch preamplifier part III
As said before, friends of surrealism must love these forums discussions......
Last edited:
Psst... don't tell anyone but "mmerrill99" is the Borg spelling of Jakob.... shhh!
I hope MM and Jakob don't take offense at my bit of fun...
They make me think of Batman and Robin, no prizes for guessing which is which Yes, thank you for your effort. I still have a couple problems:
-There is a difference between perception (Ashihara) and physiological effect (fMRI) (Oohashi). Oohashi says the subjects can't perceive the sound.
-Oohashi (2000) is cited 190 times. I didn't read them all just now and many are in Japanese. But a quick look at the titles seems (very risky speculation here) to show none have disputed it and some have verified similar results.
-Ashihara and Kiryu only cite Oohashi once, in this paper: Ashihara, K., Kiryu, S., Koubori, M., Kyoso, M., & Omata, M. (2010, November). Psychoacoustic Measurement and Auditory Brainstem Response in the Frequency Range Between 10 kHz and 30 kHz. In Audio Engineering Society Convention 129. Audio Engineering Society. It is not a replication but they do find "some subjects could perceive the high-frequency sounds above 20 kHz and the auditory brainstem response could be measured for one subject at the frequency of 22 kHz"
-Ashihara, K., & Kiryu, S. (2003). Audibility of components above 22 kHz in a harmonic complex tone. Acta Acustica united with Acustica, 89(3), 540-546.
is also not a replication, but it does say what you said, just nothing about Oohashi.
I think that was his first paper, right? His subsequent work delves deeper, including the one syn08 cites: Ashihara, K. (2007). Hearing thresholds for pure tones above 16 kHz. The Journal of the Acoustical Society of America, 122(3), EL52-EL57.
Therein he shows, after addressing the intermodulation problems (with measurements), that most of his 32 subjects hear above 20kHz and 2 hear 28 kHz.
Isn't Google great?
Edit: Damn! Too slow.
Last edited:
Don't give up that easily
No, the individual instruments/sounds may have been panned using a time difference between channels to emulate ITD along with normal level panning (or not)
You can measure phase through the system but the chances are the phase relationships between different elements of the recording would be all over the place anyway due to miking, mixing, EQing etc. Simple old recordings maybe better in that regard.
Does making a fool of oneself in public look like giving up to you?
So emulating ITD in the recording is a different approach to binaural....maybe more speaker friendly?
I would be looking for a pattern or commonality in the measurements in/out of this ASMR focus zone.
Probably not quite as you perceive it. I would strongly suggest reading Johnathan Haidt's, The Righteous Mind. He did some very interesting research into what you are talking about. In short, he found there are reasons the human population have tribal tendencies, there are reasons that there are liberal and conservative, etc. Each of those things has been needed at different times in human evolution, and natural selection has worked to preserve them. Haidt also found that several personality dimensions (a concept used in psychological testing and research) work well to characterize left vs right, liberal vs conservative, personality types across many very different types of cultures. He stated that he was once liberal, but after his research had to change his position to middle-of-the-road as he could no longer justify strong liberal leanings. Home | moralfoundations.org
Do you know what journal impact factors are? Check out Journal of Neurophysiology and JAES or any other audio journals. What besides JASA and JAES do you know of: I'll look them up too. J. Neurophys. is by no means obscure to those familiar with that type of study
- Status
- Not open for further replies.