Right, same thing with audio.
But I was just talking about perceived dynamic.
Which is why I simply cringe at the "trust your ears" mantra. That's perfectly fine when it comes to purely subjective preference, bug it does nothing to inform the objective reality. But most of those who parrot that phrase are working from the premise that it does and are in complete denial of our humanity.
se
If the argument is that you can't hear sine waves above the fundamental in a square wave, then they should not be level matched! Only the level of the fundamental in the square wave should match the level of the sine.
But the limits of hearing to a 20,000 hertz bandwidth has been mostly show to be incorrect by researchers with more modern equipment. There is a difference between testing with a sine wave and doing a bandpass limiting test.
A simple demo we do here is to play a 30,000 hertz tone. Folks don't "hear" anything but can tell the instant you turn it off. (Easy to measure that there are no subharmonics with the gear around lying around.) Now level that is required is a bit of a different tale.
Most small headphone transducers will do 30,000 hertz for those who wish to actually try things for themselves.
I'm happy to not be a totally crazy audiophile who believe in magic and deserves to be burned for sacrilege against the laws of psycho-acoustics.
I am told that the stake was quite disagreeable.
BTW: I remember to had, long time ago, my hearing tested by a doctor of the occupational medicine. He told-me that, at this time, I was limited at 15KHz. "There is a problem, doctor", I said to him. "You are using to diagnose the bandwidth of the people an headphone that is unable to reproduce such high frequencies at an acceptable level !". I knew this headphone very well and had made measurements of it, between some others. ;-)
Last edited:
What middle ear parameters tell about impedance matching and high frequency hearing
Is one article that covers the limits to hearing in mammals. With bone conduction frequencies as high as 150,000 hertz can be detected by children. This drops to 50,000 hertz for adults. It is presumed that mass damping of the ossicle that is thought to be the limiter for high frequency response.
Of course we all know that all new research stopped once we left college....
So the bone conduction research was done in 1950 and should apply to most here.
Is one article that covers the limits to hearing in mammals. With bone conduction frequencies as high as 150,000 hertz can be detected by children. This drops to 50,000 hertz for adults. It is presumed that mass damping of the ossicle that is thought to be the limiter for high frequency response.
Of course we all know that all new research stopped once we left college....
So the bone conduction research was done in 1950 and should apply to most here.
Oh, found a measurement that shows why i was unable to hear the sinus tone at 15Khz and pretty low level with this headphone 
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
None of the available couplers for headphones can make meaningful measurements above about 8 KHz. The cavity resonances and other aspects of the coupler prevent it. You can try free space measurements (Audeze does this) and get some idea of what happens at the driver. The plot shown has significant response at 20 KHz which is pretty good.
Relating that to what happens at the eardrum is a problem. Even with a probe mike in the ear canal you still get that type of response. If you crank up the HF so you can see it in the plots you get "Beats" sound, something only tenuously connected to reality.
The ear has a transient response that does not match its frequency response. Its much faster I have been told.
Hearing test headphones are more about consistency with measuring a population. Switching headphones can completely invalidate the measurements compared to the thousands of others who have been measured for hearing loss. I would not use headphones as a means for quantitative testing of ultrasonic sensitivity on humans. Just too many variables.
Relating that to what happens at the eardrum is a problem. Even with a probe mike in the ear canal you still get that type of response. If you crank up the HF so you can see it in the plots you get "Beats" sound, something only tenuously connected to reality.
The ear has a transient response that does not match its frequency response. Its much faster I have been told.
Hearing test headphones are more about consistency with measuring a population. Switching headphones can completely invalidate the measurements compared to the thousands of others who have been measured for hearing loss. I would not use headphones as a means for quantitative testing of ultrasonic sensitivity on humans. Just too many variables.
Current technique is not to use a headphone for testing. Instead a transducer drives a small tube. So even though the wave is created by a piston with a small tube it can present as a point source. This is then spot frequency calibrated. A very small tube has a resonance well above the frequencies of interest.
One can however demonstrate limits with a small light weight transducer typical of ones that are used in some earphones.
One can however demonstrate limits with a small light weight transducer typical of ones that are used in some earphones.
I tend to agree with John. The musical content does come first, but I'd be a liar to say I don't really care how well it's been reproduced. At the cost of being an elitist, a boom box is all right if I have no choice, something is better than nothing, but if I do have a choice, I'll gladly skip the boom box and go for something which sounds better to me.
Just because someone splashes like $100,000 on a Mercedes-Benz does not necessarily mean he's a car freak, perhaps he simply enjoys riding in relative silence and comfort.
Just because someone splashes like $100,000 on a Mercedes-Benz does not necessarily mean he's a car freak, perhaps he simply enjoys riding in relative silence and comfort.
Just because someone does not splash $100k on a Mercedes-Benz does not necessarily mean he's not a car freak, perhaps he simply enjoys sleeping in relative safety and comfort.
(A relative installed a 50k armored door in his bedroom, after he had to hand over his Benz keys in his PJ's. Neither is a problem any longer, after he unlawfully blew every pre-tax dollar in his company on the stock exchange, the IRS is not likely to stand next to his bed with a cocked Glock)
(A relative installed a 50k armored door in his bedroom, after he had to hand over his Benz keys in his PJ's. Neither is a problem any longer, after he unlawfully blew every pre-tax dollar in his company on the stock exchange, the IRS is not likely to stand next to his bed with a cocked Glock)
I Agree with each of your words.
I had comparative measurements of a lot of headphones to make for a revue.
The method we used was to chose several "best sounding ones" for their tonal balance, and apply a crorrection in the measuring mic according to the accidents that were shared by headphones. Average way. And even with this, it can change a lot with any difference of positionning, air sealing for the closed air caps etc...
Comparing with free air response in high frequencies too...
A nightmare.
Nb: the accident shown in this curve with this headphone had, helas, some consistancy. Even in free air. Probably due to the back enclosure behind the membrane.
Last edited:
Exactly.
To them, they say they can hear real differences in all sighted tests regardless of what is being tested. At which point, I ignore them -- that claim is simply not credible. eg. copper vs OCC copper to name one.
Last edited:
But they will distort in a non-ideal way and introduce sub-harmonics that we can hear, that is why we might 'perceive' such things. The obsession with being able to perceive super hf in audio makes no sense to me in the context of what real people can physically hear in natural sound. Attached is an extract which shows the effect of age on tone perception threshold, up to age 55, by percentile of population. Be afraid, be very afraid. Above age 55 is restricted viewing only, too gruesome for anyone over age 55 to view. 30kHz, forget it.
Attachments
Considering the general falling spectral response of real free field music, I turn to the simple tests with/without low-pass filters on real music. I have not found compelling evidence to look there for any answers. As Dave Griesinger has published one first needs artifact free reproduction at these frequencies.
Simple experiment play 1/3 octave noise around 25kHz from a really good 24/96 DAC directly into your analog signal chain at varying levels.
Last edited:
And your financial interest in replacement tweeters is exactly what?
On large scale systems due to air losses the HF limit is usually 8,000 hertz.
Last edited:
As I read the graphs 25% of the subjects hearing improved below 5,000 hertz!
For me the scariest thing is how bad the hearing is for professionals in audio. I don't know how many of you read Rolling Stone, but the feature on Jimmy Iovine is most illuminating. In one section he talks about being able to hear immediately what is wrong with something, I think saying "from a block and a half away" if memory serves. Then in another place he talks about his hearing loss and how his kids make fun of him by mumbling.
And of course the characterization by him and Dre of audio engineers as wearing white lab coats and carrying clipboards...
And of course the characterization by him and Dre of audio engineers as wearing white lab coats and carrying clipboards...
- Status
- Not open for further replies.