I go with that up to a point. One can make a test of something so uninteresting as to make it pointless. My CD player sounds fantastic. For some reason I seldom use it for more than an hour. TV and LP are day long. I almost feel MP3 is better for me. Could be CD is giving me more than I need and that more makes me feel distracted. Now AB testing will not tell me why that is true for ME, it's a snapshot. BTW. If you do AB it should really be AAB. If you do AA the second A sounds louder, more open and more expensive. It's that we learn from the first listen and focus on the better qualities next. Also when a professional footballer misses an easy goal or whatever it is often due to fear. Testing introduces fear. It can be a critical mistake in this so called science not to accept that. In other words the test is now the science and not the object of the test ( could be I read a lot of that at DIY Audio ) . I think this prooves the whole concept is wrong whatever you do. One just has to understand that people believe things for very personal reasons. Often these are very well imformed reasons.
Ahem, please see Confirmation Bias: https://en.wikipedia.org/wiki/Confirmation_bias This is one that is pretty difficult for most people to overcome most of the time. Often nearly impossible, especially when combined with effects like Social Proof: https://en.wikipedia.org/wiki/Social_proof
But Jakob, if you can see a big new shiny piece of electronics there, then the eye DOES have something to work with and it takes a mighty ability not to be prejudiced by that.
Do you have a reference to someone for whom McGurk does not work?
Do you have a reference to someone for whom McGurk does not work?
From Wikipedia: If a person is getting poor quality auditory information but good quality visual information, they may be more likely to experience the McGurk effect. ... People who are better at sensory integration have been shown to be more susceptible to the effect. Many people are affected differently by the McGurk effect based on many factors, including brain damage and other disorders.
The BBC designed 13 bit encoding circa 1969 and used soon after. It was the technical limit at the time and is my favourite invention in audio ( 10 + 3 ). It was a technical test with musical advisors. In the BBC lab some pumping sounds were heard at - 75 dB or lower. These caused doubt as they were distracting to some. When broadcast these became lost in hiss. I call that the work of real experts. A bias towards how they wanted it to sound and going with something that looked a wrong choice on paper. I was told Japan pulled it's hair out when technical testing became the fashion. Before then the technical was called on to explain or confirn something or to challenge a claim. When I asked what Japan did I was told. Same look of product for Japan with better sound for Japan. My ears tell me that is correct.
Ok, I'll test on my webdeveloper when he is back in. He has only 10% vision in one eye. Which eye is variable. Damn good developer tho and having a dog in the pen is good for stress levels.
What if the two pieces under comparison share equal "shinyness" ?
Is "mighty" an abbreviation for "allmighty"? If so, that is a hypothesis but where is the data?
If not, what does it mean?
Any comment on my example wrt stereophonic illusion?
Otoh it sometimes can be really irritating if the stereophonic reproduction is really convincing (to a certain listener) carries a sort of approximation to reality but there is nothing to look at.
Proverbio, A. M. et al. Skilled musicians are not subject to the McGurk effect.
Sci. Rep. 6, 30423; doi: 10.1038/srep30423 (2016)
Last edited:
How we work is still slightly above our science. We get very close when it matters. It was said only 10% of the brain was used, now that is doubted. Exactly how hearing works is not really known. Something I helped with for Oxford University showed that hearing uses an analogue section of about 30% THD ( interesting curve , exponential ). A digital system of sorts and a supposed servo signal in the 0.5uS range brain to ear. Most of this via the Hamster as nearest useful similar animal. Mice being too different. As far as I know this is not general understanding. Two tested KEF T27 were used to ensure > 40 kHz range. The research asked the question did the 0.5uS signal hold the key to understanding the system. US military research found something similar which gave birth to some thinking that very fast signals matter whilst high frequency tones do not. I have no idea myself. The hypothisis was the analogue plus near digital and servo make a system that knows differences. Usually this is to escape danger. My feeling is danger is transients and music is bird song. This was the world we came from long ago.
'mighty' as in strong and powerful. Mighty Thor, mighty mouse, mighty ability that us mere mortals do not possess.
That's a very interesting (and recent) study. I hope somewhere it is being replicated to see if it holds.
The auditory illusion fascinates me not least becuase, if you have something wrong in setup you can hear two speakers as distinct sources. But when things are setup well you cannot easily determine the source, even when you can see the speakers. With the right material you also get depth as long as you stay fairly still which is even freakier.
BUT there is still conditioning. If you have a pair of minimonitors in front of you you might think 'these will image great' whereas a pair of 3-ways you might think ' not chance'. But if you can ignore that conditioning and heresay then I agree completely.
That's a very interesting (and recent) study. I hope somewhere it is being replicated to see if it holds.
The auditory illusion fascinates me not least becuase, if you have something wrong in setup you can hear two speakers as distinct sources. But when things are setup well you cannot easily determine the source, even when you can see the speakers. With the right material you also get depth as long as you stay fairly still which is even freakier.
BUT there is still conditioning. If you have a pair of minimonitors in front of you you might think 'these will image great' whereas a pair of 3-ways you might think ' not chance'. But if you can ignore that conditioning and heresay then I agree completely.
https://pdfs.semanticscholar.org/42a6/e1cf2cb6a418b5761e05c99fdc983ab43c5d.pdf
Note. My interest was not what was being paid for.
Note. My interest was not what was being paid for.
Hi Nige.
In my experiments I conclude that the playback system behaviour changes according to peak levels encountered during the first play.
The second play is different and stable, so AB testing really needs to be AABB.
Dan.
Hmm mulling over what Mark and Jakob have been saying. Now if you subscribe to the view that we all sit on a continuum of synesthesia rather than normal one side and Messian on the other then we will have another degree of freedom for how well you can decouple the ears from other senses.
Probably a red herring aganst the bigger question around training.
BTW and harping back to the discussion on 'party effect'. Does anyone ever try and tune into specific random conversations when in a crowd? I'm assuming this is similar to the conditioning we are talking about to be able to single out anomalies in playback?
Probably a red herring aganst the bigger question around training.
BTW and harping back to the discussion on 'party effect'. Does anyone ever try and tune into specific random conversations when in a crowd? I'm assuming this is similar to the conditioning we are talking about to be able to single out anomalies in playback?
A fast detector for transients to bypass the 'normal' detector sounds like it makes sense for survival value.
I do remember a study on audibility of 20kHz+ signals, where the test subjects could not reliably detect the presence of such signals. But brainscans showed that whenever the 20kHz+ signal was present, it caused significant brain activity that would stop as soon as the tone stopped. However, the test subjects were not at all aware of it.
Jan
We run a fast attack/slower release AGC.
Fast transients that get through the AGC combined with the spectral change is what the ear/brain detects and brings attention to.
Pretty much none of us hear airborne 20kHz directly.
Pretty much all of us should be able to 'sense' 20kHz in the air, by discriminating a fine change in hearing or self.
We also sense magnetic disturbances, perhaps a confounder in headphone experiments.
I have never been comfortable with Class D switching amplifiers because of a something 'in the air.'
Ultrasonic signal emitted by digital playback systems, acoustic and/or magnetic may be the cause of your Nigel's discomfort/fatigue.
Dan.
That was Oohashi et al. ´s paper on the socalled hypersonic brain effect. Several studies by him and other seem to confirm the first findings.
Btw, the participants could reliable detect the socalled full range sound (means contained the spectrum below and above 20 kHz) as the psychoacoustic tests revealed but could not detect the signal containing only the high frequency spectrum (means above 20 kHz) and the PET scans did not differ from baseline with high frequency content alone, but differ if full range sound was presented. (and differ from scans where only low frequency content was presented)
Oohashi et al., Inaudible high-frequency sounds affect brain activity: hypersonic effect.
J Neurophysiol 83: 3548–3558, 2000.
Thanks,I missed that page.
The rebuttal letter is especially of interest to me. Saves a lot of re-work, I subscribe to "standing on the shoulders of giants".
http://www.gedlee.com/downloads/comments on howard.pdf
On "what is wrong with op-amps?":
My audio designs include a lot of old and new IP but as to why some hardware 'sounds' so much better than others is still a largely mystery. I had a Pioneer SX3000 (similar to sx626 but different pcbs) purchased in 1974 at an army PX in Bavaria that is still memorable as best amplifier ever. 28W, low tech solid state.
I did listen to it through full labyrinth speaker enclosures so yes that too may be a factor as Dr Geddes thinks.
I know at this point it is not completely the B(AB) crossover issue, its not global feedback, its not the lowest THD/TIM/IM and its not solid or vacuum state. Right now the op-amps are still questionable simply because so much equipment using them sounds terrible. The fault is probably not in the op-amp design itself.
A lot of (experimental) evidence is currently leading to a type of so-called quantum En connection. SR/Lorentz/Maxwell physics says no-way but I'm assuming much more remains to be learned in the coupling of living and non-living matter.
-
Hi Dan, I was told the problem is well demostrated by the Golf pro who misses a shot he can not miss. Learning brain and the brain that retains. If fear takes over the brain switches to learning brain. AAB is when we might get a truer picture than AB. Often bordom takes over. I set up a Nakamichi Dragon for TDK D tape which was a reference for the machine ( an excellent cheap ferric tape ). It seemed the tapes using AB were almost perfect copies. In the morning the next day it was clear the dynamic range was poor. When doing AB somehow the dynamics problem was not so obvious. That is the reverse of exspectations I feel. However with a little thought it would be true if how the ear works is factored in. Headphones don't help. As I get older headphones don't wow me. I tried a phase correction circuit or whatever it should be called and found it worse. That is the processing made it less transparent so put me off.
A person can have poor tonal frequency response yet still enjoy a fast amplifier or op amp. Michael Gerzon touched on this. I miss my talks with him. Michael never put people down, yet made sure they understood the answer their question raised. He had an near infinite gearbox of mental ability if he thought he was being understood. If he didn't he just giggled. I asked a friend why Michael was so weird in his ways. He thought Michael had made a very big jump in mathamatics which he kept to himself as he thought it very dangerous. So as to keep himself level he pateneted many aspects of audio and perception. The friend said the maths never let Michael sleep. His witten stuff is very level. You would never guess it was the same man if you met him. He never had a bath so best met in the street.
I'm talking about some kind of deeper (quantum ?) effect.
I have used two different USB cables to transfer two copies of the same wave file to flash memory storage/player devices.
Playback A and playback B sound different, and further ABAB is different to AABB.
During first play of A the system changes and takes a 'set' coinciding with the first high momentary peak level somewhere in the passage.
Upon subsequent playing of B, the system retains A signature until the first high momentary peak level occurs, the system then takes B 'set' and subsequent plays of B reveal no more changes.
Replay of A file will reset the system back to the A 'set'.
Change of source equipment will change downstream system in same manner.
Change of interconnects will cause same nature of change, although less magnitude.
Extended 'burn in' is a nonsense, all that is required is a system momentary peak amplitude greater than normal operating level.
Dan.
PS - Is there anybody in the audio world you have not met ?.
- Status
- Not open for further replies.