Yes indeed, I can hear some store lighting and some switching regulators too - quite annoying. The reason I mentioned the CRT line frequency as being easily audible is that it has a known frequency of 15.625Khz (in PAL countries) so I know exactly what frequency it is when I hear it...and running out of retail stores with poor lighting (..that don't seem to bother anyone else).
-switching power supplies and their cap's screaming aren't a pick-nick either.
For lighting and switching power supplies who knows exactly what frequency they're whistling at unless you went to the trouble to measure it...
In one sentence you say it would be 1 in 1000 or 1 in 10,000 and in the next you say that nobody in audiology tests above 10Khz. So where do you get your numbers from then if nobody is doing the testing ?
Normal hearing tests only go up to 8Khz or so, they're not trying to test your ability to hear every last nuance of a musical performance, they're testing whether you have hearing problems that would cause you difficulty in day to day life.
Come now, sub harmonics ? Clutching at straws ? Aliasing perhaps, except that I've done the tests with entirely analogue systems (including signal generator) which don't suffer from aliasing.
I never said I had exceptional hearing at 17Khz, I said I could hear it, and without turning up the volume from a normal listening level to do so, that's all.
Why would you have reason to doubt what I say ? I say I can hear 17Khz, you say its unproven. A bit of a waste of time arguing over it isn't it ?
Sometimes it even changes sides.
There was a time I kept feeling both sides consistently sounded different, and I thought it was the room; later I just decided to get a hearing test, and discovered it was my own ears. But occasionally, for some reason, one side will open up. Really have not figured this one out.
I agree. I have experienced this countless times as my father worked for many years as a TV repair technician. His lab was always full of that stuff. When the horiziontal output vacuum tube is dead you won't hear that high pitch noise anymore. Any person with normal hearing can immediately spot it.
If our limit were really 10KHz I wonder why tweeters are still in use as there are so many small mids that boast perfect perfomance up to 12KHz at least, both domes and cones, and can be crossed over at relatively low frequency between 500 and 700 Hz wihtout troubles where dispersion starts beaming.
Simon,
I agree that some of us can hear a 17khz tone. I have done that with a Burr Brown analog tone generator and I could clearly hear that tone, I didn't look at what the spl level was but it was not at a very high level as I listened from low to high without turning the volume up. At the same time my brother could not hear it at all no matter the level, he probably was one of those who didn't hear much above about 12khz. I know by 20khz for me it was no dice, not level associated, just couldn't detect it. I'm 58 and have been around lots of loud music and also loud industrial equipment so it goes without saying.
I also must say that some low frequencies drive me nuts. If I am in a car with a sunroof open or a window open at just the right size the low frequency buffeting will drive me nuts and others in the car say they hear nothing! I'm not much of a fan of sub woofers getting down into the sub audio range, unless it is for effects in a movie I would just as much prefer that it didn't get down to those low sub frequencies.
I agree that some of us can hear a 17khz tone. I have done that with a Burr Brown analog tone generator and I could clearly hear that tone, I didn't look at what the spl level was but it was not at a very high level as I listened from low to high without turning the volume up. At the same time my brother could not hear it at all no matter the level, he probably was one of those who didn't hear much above about 12khz. I know by 20khz for me it was no dice, not level associated, just couldn't detect it. I'm 58 and have been around lots of loud music and also loud industrial equipment so it goes without saying.
I also must say that some low frequencies drive me nuts. If I am in a car with a sunroof open or a window open at just the right size the low frequency buffeting will drive me nuts and others in the car say they hear nothing! I'm not much of a fan of sub woofers getting down into the sub audio range, unless it is for effects in a movie I would just as much prefer that it didn't get down to those low sub frequencies.
Last edited:
Obviously it drops, but are you referring to threshold sensitivity (what's the quietest sound we can hear at a given frequency, eg the bottom rung on the fletcher munsen curve) or are you referring to perceived fall off in volume at higher frequencies at a moderate SPL level ? (Further up the fletcher munsen graph where it is flatter)
Two very different things. Just because your threshold for hearing is much higher above 10Khz doesn't mean you don't hear above 10Khz at more normal SPL's.
Where did you pluck 10dB from ? I've never seen this figure quoted for non harmonic masking. Perceptual codecs typically use a figure of closer to 30dB in high quality mode as the threshold below which nearby signals are masked and "don't matter" any more. Certainly not 10dB.
Your making quite sweeping claims here, do you have anything to back it up ? Perceptual codecs like mp3 and aac "don't do anything above 10Khz" ??
In fact mp3 typically brick wall filters the treble at between 17Khz and 19Khz (adjustable in the encoder) because mp3 with its fixed width buckets is inefficient at encoding high frequencies.
(The trade-off of brick-wall filtering the treble is that bits are more efficiently used to encode more important lower frequencies instead of being "wasted" at frequencies that many people can't hear)
AAC on the other hand does not have this fixed bucket width limitation and can encode high frequencies efficiently so typically AAC does NOT brick wall filter the original signal at 17-19Khz before encoding. AAC is a later, better codec than mp3 so why would they bother to remove the high frequency restriction if it didn't matter ?
Perceptual codecs have come along way in the last 15 years - I remember the first mp3 encoders running at 128kbit were dreadful, it was a number of years before anything close to acceptable quality was achieved, (I much prefer AAC over MP3) in the process some of the original design assumptions of what we can and can't hear have had to be revised...
So you say. But where's the proof.
Again, so you say. I remember reading an article not long ago where many musical instruments were recorded at very high sample rates with instrumentation grade microphones to see just how high in frequency the harmonics go.
And for a surprising number of instruments especially stringed instruments they had significant harmonics well up into the 30-40Khz range. The limit is not the source material - the limit is our hearing.
If nothing in nature exists above 10Khz why did dogs and other animals evolve to be able to hear as high as 40Khz ?
And speech bandwidth limits have absolutely nothing to do with how high and how low we can hear, nor how much bandwidth is required to faithfully reproduce music!
That's like saying black and white are the only colours we need to reproduce written text, therefore a television screen doesn't need to be colour because colour doesn't matter... If people "preferred" a cut in high frequencies above 8Khz in some test (made decades ago with what recording and playback equipment ?) that tells us two things - (1) People could hear above 8Khz well enough that they had the potential to be annoyed by it (otherwise they wouldn't have a preference) and (2) there was a problem with the sound quality of the recording or playback chain at high frequencies that was irritating the listeners.
Nobody applies an 8Khz low pass filter before their ears to limit the bandwidth of real life sounds we hear in every day life, and yet nobody is complaining about that. Aside from obnoxious noises like heavy machinery, jack hammers etc normal every day sounds around us sound just fine without rolling off the high frequencies.
Someone preferring to roll off the high frequencies on a recording is a very clear indication to me that there is a sound quality problem with the high frequency recording/reproduction system - most likely breakup resonances in the speakers, or peaking of the microphone. (Many 1/4" electret capsules have a large peak at 15Khz unless EQ'ed)
Last edited:
Perhaps Earl's statement that he doesn't think anything above 10khz is one reason that he says that all compression drivers sound the same? If you can't hear over 12Khz I would have to discount any instrumentation measurements that would come to this conclusion if you can't actually hear the sound they are making. I for one do not believe that all compression drivers sound the same and that you can use eq to make them all equivalent.
System like those have rather limited time resolution which under invariant conditions (i.e. measurement) also reflects into a limited bandwidth or the contrary if you like. Human time resolution is rather high, in the 1 us region, and it is not even demonstrated that is the human limit. It was just the instrumental limit when experiments were performed. In many cases having a better bandwidth is also an advantage in the time domain.
Most speakers are rather poor in the time domain and get even worse once placed in a room for playing music. Frequency response is not really the main thing to address although this doesn't mean it can be anything.
Last edited:
Earl,
Seems we go over this every year or so.
Your wife probably seldom checks above 8 kHz (the usual sonogram range), as we really don't need above that point for communication, and hearing aid amplification above that frequency will hinder speech recognition in most cases.
You can't hear above 10 kHz, so you assume content above that has no relevance to anyone.
You cite HF air absorption losses, when the preponderance of recording is done with mics inches from the source, and our loudspeakers within a few meters of our ears, rendering HF air absorption on the order of a few dB even here in the worst case dry hot desert conditions.
I have been checking my hearing for years using this:
Free hearing test on line – Equal loudness contours and audiometry
I have used Sony 7506 and GK Ultraphones (same transducers in Peltor muffs) for testing.
With the 29 dB noise reduction of the GK Ultraphones houshold noise is reduced to the point where I can detect the tone in the LF with far less gain, but as is easy to see I suffer from classic NIHL (noise induced hearing loss) and Presbycusis (old age HF hearing loss).
The measurements of the headphones were done with a BK 4007 in a flat baffle held to the phone transducer, without any pinna equivalent, so are not exactly representative of their actual frequency response, but show response extending past 20 kHz.
As you can see from my test results, my hearing "tops out" above 16 kHz
As I recall using a tone generator 15,750 Hz does not need as near as much gain to hear, but past 16 kHz I would need damaging levels to hear any tone.
Bonnie's hearing shows only a 3 dB decrease in sensitivity at 16 kHz, and sine wave tone tests indicated she had response to 20 kHz with little boost required, though above that response dropped rapidly, though the headphone response also dropped rapidly.
At any rate, my hearing is no where near as good as Simon, a few dB difference above 8 kHz would no longer concern me (though 40 years ago I could easily detect those kinds of detail) but loosing the 1/3 octave from 12500 to 16000 is still quite apparent to me.
Looking at my father's sonograms, it is probably only a matter of years when it will drop down another 1/3 octave, and I'll join you in the "10 kHz and above doesn't matter" club.
Until then, I'm still in the "flat to 16 kHz" club (Bonnie does not care about the 16 kHz to 20 kHz range, even though she can hear it clearly).
Art
Attachments
Earwax.
After cleaning my ears I get a much more extended freq. response that takes a few hours for my brain to compensate for. Until it is, everything sounds sibilant and "splashy" - with reflections being more noticeable.
It also happens when earwax moves around (like after sleeping or an air pressure change "pop") in one ear - though the overall effect isn't as noticeable as a cleaning.
Note this isn't around the outer ear, but further in the canal. (..in my case an increased adaptation due to most of my childhood spent swimming.)
Scott,
I'll have to remember that comment the next time someone says I'm not listening! I was a swimmer also, don't think the body would do that long after you stopped getting in the water though.
Soongsc,
One thing to remember besides wax buildup as Scott has just suggested is also the sinus cavity and the simple fact that changes throughout the day, sometimes one nostril is closed other the other side and if I remember correctly this also affects the ears?
I'll have to remember that comment the next time someone says I'm not listening! I was a swimmer also, don't think the body would do that long after you stopped getting in the water though.
Soongsc,
One thing to remember besides wax buildup as Scott has just suggested is also the sinus cavity and the simple fact that changes throughout the day, sometimes one nostril is closed other the other side and if I remember correctly this also affects the ears?
For the most part above 10kHz is musically irrelevant (triangles would be an exception). That doesn't mean though that your brain hasn't adpated to make use of noise above this, or that very high harmonics aren't audible above 10 kHz. Their absence may well be noticeable and detrimental to you because that's what you are accustomed too.
Note: brickwalling isn't a natural occurrence, and it almost certainly has negative consequences lower in freq..
As for your hearing.. the damage could have lowered your sensitivity in the right ear at *lower* freq.s.. (in the midrange), with a perceptual increase at higher freq.s. as a result. (Nerve damage.)
Was your left ear attenuated at higher freq.s before the "accident"?
Oh it does, definitely. I'd say I produce many time the average deeper in-ear even 20 years afterward. When I was a younger child before swimming, I didn't produce nearly the same amount.
The only thing that's really changed 20 years after being a frequent swimmer for me has been tinnitus (around 6 kHz left and a little lower in the right) and a gradual loss of hearing above 9 kHz or so. Still, certain lighting noise well above 9 kHz can give me headaches.
(..and the tinnitus I've learned to mostly ignore, except when trying to sleep - and then I use a "noise maker" .)
Scott,
I understand about the tinnitus as i have become aware of it myself lately but so far only in one ear. I only really notice when it is really quite and like you say when I am trying to sleep. I leave the tv on as low as it will go but I'm thinking about one of those white noise generators made to lull people to sleep. I don't know what frequency it is at as I haven't had a reference to compare it to yet but it seems fairly high up. Don't know that I can associate it with swimming though and I wasn't a diver so never hurt or punctured an eardrum.
I understand about the tinnitus as i have become aware of it myself lately but so far only in one ear. I only really notice when it is really quite and like you say when I am trying to sleep. I leave the tv on as low as it will go but I'm thinking about one of those white noise generators made to lull people to sleep. I don't know what frequency it is at as I haven't had a reference to compare it to yet but it seems fairly high up. Don't know that I can associate it with swimming though and I wasn't a diver so never hurt or punctured an eardrum.
The brickwalling was done via a digital filter, using the highest number of taps possible - this means 120dB attenuation over a tiny range of frequencies, visually, it's a vertical plunge in the EQ module - doing a spectrum before and after shows no impact to the response just prior to the cutoff.
Can't see that being the case, I've spent quite a bit of time with my ears jammed up close to speaker drivers while working pretty hard, and tonally the left and right ears sound the same.
Don't know about the left ear, never had the ears measured for response - fairly recently I started using a pure 18k sine wave to condition a tweeter, and that's where I picked up, and did some simple listening experiments on, the ear differences.
Last edited:
The most important thing in listening tests is the experience and understanding of real sound. Source of each spectrum content of real sound. How these change in real life is much more subtle than math modeling. This is one reason why we came to the conclusion that various auditors are necessary to help catch problems. Most of them may not have good hearing in terms of hearing tests, but the experience with having good understanding of real sonic characteristics is much more important.
I remember a movie, "The God's Must Be Crazy" from years ago. One of the points made in the movie, besides that the people of the Kalahari desert are very beautiful, is that they can hear and interpret sounds up to 20 kHz. We might want to investigate the reasons for this, but the movie claimed it came from hunting scorpions at night, by hearing alone, that children were better at this but that all members of the tribe hunted until quite old... no mention of what "quite old" might be either. Also, do you suppose that an observed irritating peak at 17.5kKz might be more noticeable than no noise at that point and that might have a bearing on the complaint?
I;ll see that and raise you a "we are affected by >20kHz" study/
Oohashi, Tsutomu, et al. "The role of biological system other than auditory air-conduction in the emergence of the hypersonic effect." Brain research 1073 (2006): 339-347.