HydrogenAudio

I just came across this little hearing test.  It's fairly simple if you want to give it a try. I only started to hear the test tone when the voice-over said "15 kHz". I was wondering what is the max frequency that others hear?

I'm 48 y.o. BTW.

http://audiocheck.net/audiotests_frequencycheckhigh.php (http://audiocheck.net/audiotests_frequencycheckhigh.php)

I heard 19k (    I had no idea i could hear that high) in my left ear and 18k in my right one.

I heard 19k (    I had no idea i could hear that high) in my left ear and 18k in my right one.

Hi jm, If you don't mind revealing - what is your age?


Edit : No need, I just checked your birthdate in your profile. You're only 18 y.o.

17k for me (I'm 19).

I did a similar test several years ago and I could hear up to 22 KHz.

I prefer not to take this test. It will ruin my mood for a day at least.

If I generate a sine waves with Audacity I can hear 20 KHz. Using this test I was only able to hear 18 KHz.

I'm 21 BTW.

I don't know if it's because I'm only 19, or my laptop's crummy soundcard or my crappy 3 year old iPod stock headphones, but I could hear all the frequencies from right at the beginning.

And from what I can tell, the sound doesn't start until he starts to say the word "two" in twenty-two.




I did the low one as well, but I couldn't hear until about halfway between the 20Hz and 30Hz. But that could be the crappy 'phones fault.

The test file is at 44.1kHz.

Do you know if your soundcard is resampling at 48kHz?

(BTW, this is a general question to anyone who isn't already aware of this phenomenon)

Woohoo! 16kHz (17 maybe). Age = 43.

Cheers, Slipstreem. 

How could I find out if my card is resampling? It's an inbuilt thingy in a Toshiba A30-303 laptop.

don't blame your speakers too quickly. Unlike low frequencies, high frequencies are easily reproduced, even with cheap speakers. In this test, the weakest components in your audio system are your ears.

That's from the first paragraph of that page. Then, on the last paragraph

Beware: cheap sound cards may have trouble reproducing the highest frequencies of this test. At best, they won't reproduce anything. At worst, ghost frequencies will be generated in the audible range. These frequencies are not representative of the file's contents! If you hear strange ups and downs or anything else that does not sound like a continuously descending sweep, suspect this test to be corrupted by aliasing (the scientific name to this phenomenon).

This *is* important. I've heard sometimes some tests that caused troubles like the ones described in there.


Just for completeness, i heard from 17Khz down (i'm 28, so i'm going down already  )

Edit: I actually hear from somewhere between 18Khz and 17Khz. Did the mosquito test too, and can hear it. (although i have to put the volume up a bit)

That's a bad choice to offer the sample at 44 kHz speed.

Here is the same file resampled to 48 kHz for those that think kmixer or their soundcard might be filtering due to resampling
http://shup.com/Shup/57684/audiocheck.net_...ycheckhigh.flac (http://shup.com/Shup/57684/audiocheck.net_frequencycheckhigh.flac)

??? I can hear the sound right fom the beginning ???

1: I am 30 year old
2: my CPU fan is noisy (I thought it would mask it)
3: I can hear it right from the speaker I didn't even focus or used my headphones like I do when I ABX hard sample sometimes.

I never did such a test for myself so I am very surprised specially because of my age.

I recall I did a test like this when I did my military service ... the game was to guess if the sound was coming from the right, left, both or none ... guessing from the mood of the military guy behind the glass I thought I won the game at that time

Is it normal from a medical point of view to be able to listen to 22Khz noise at 30 year old ? personnaly I was expecting 19-20Khz as I knew I was not deaf ... but I right now am very surprised by my ears myself ... thks for the link 

Mmm.... there's also a test for decibels

http://audiocheck.net/testtones_dynamic.php (http://audiocheck.net/testtones_dynamic.php)

This one reproduces pink noise at full scale, followed by pink noise x dB's lower (so it's a good test to not fake it).

Playing at full volume (with this laptop) with the headphones i can hear down to the 72dB one (probably more), although at the usual volume i listen, i couldn't hear the 66dB one.

I don't know if it's because I'm only 19, or my laptop's crummy soundcard or my crappy 3 year old iPod stock headphones, but I could hear all the frequencies from right at the beginning.

And from what I can tell, the sound doesn't start until he starts to say the word "two" in twenty-two.

Yeah I looked at the waveform in an editor and that's exectly where it starts, it's just after the word twenty and exactly at the begining of the word two. You must be part bat AliL


Oh dear, so far at 15Khz I'm the lowest

One good thing though, I don't have to worry about the default low pass filter settings in my lame -V3 encodes

Actually that was one of the reasons I did the test. I took the the original wave file and then I encoded it with my normal settings (lame3.98 -V3) and then decoded it back to wav. Then I did the test, first on original wav file and then on the decoded mp3 file. In a wav editor I could clearly see that the first few seconds of the tone had completely vanished in the decoded mp3 file, it only started 18kHz for this waveform. But for me it made absolutely no difference, it was a good 2.5 to 3 seconds after the tone kicked in (at 18k) before I could hear it in either case. Phew, what a relief

One good thing though, I don't have to worry about the default low pass filter settings in my lame -V3 encodes

I had similar thoughts, hence why I'm so happy. I encode everything with LAME at -V3, and the lowpass filter imposed at this setting has very nearly no effect on what I hear. It's conveniently confirmed for me that, in that respect at least, I've chosen exactly the right setting for me personally for maximum compression efficiency. 

Cheers, Slipstreem. 

Is it normal from a medical point of view to be able to listen to 22Khz noise at 30 year old ?

Did you hear the frequency going up, or down, when playing  22, 21, 20, 19, 18?
I ask this, because as I said above, (and is briefly explained in the test page), The hardware can play a role here, and most often than not, listening with speakers and with AC-07 soundcards, what you hear are in fact aliasing sounds and ressonances, rather than the real tone.

You could start trying with the headphones and see if you can also hear it, but the soundcard could still have an effect.
To avoid it, play the sample that benski posted instead of the one in the website.

I downloaded the file and was clearly able to hear at 17kHz (using some basic earphones).
When I moused over the download link though I could hear it right from the start.  I suspect that all playback routes are not equal.

I'll have to test later with a quieter environment and better speakers.

blah ... it's only 17Khz with benski's file ... you ruined my day

I'll have to test later with a quieter environment and better speakers.

As was said earlier, speakers/headphones are not likely to make any difference, nor is the environment provided that you have you're playing back at normal levels for that environment.  The tones in this sample are extremely loud.

I've seen cheap integrated CODECs in older motherboards and laptops that can't even play 48000 samples-per-second material. The page describes what exactly you should hear: a clean sine wave slowly decreasing in frequency. If you hear multiple tones going up and down, the test has failed.

18.5 kHz @ 27 years.

somewhere between 21 and 22kHz and now I have a monster headache  (28 years old)

18.5 kHz @ 27 years.

gnypp45: 16.5 kHz @ 37 years

uart: 15 kHz @ 48 years

That's about 2 kHz every 10 years. 

I could hear 18kHz at 24 years old.

With benski's file I can hear the sound at 19 kHz. I'm 22 years old by the way.

17K @ 37 (I'm not old!), but that was with speakers (not headphones) in a less-than-ideal listening environment.

between 17-18 khz 26 years old

12 kHz, I think (age 62). I heard several quiet up and down sweeps first, but then a solid high frequency tone just at the end.

between 19-20 kHz in the right ear and 14-15 in the left, I'm 19.

I've been strictly wearing ear protection at concerts and just listened to low volume over my headphones... And then I got issues with blood circulation in my left ear. 

How do you get a 17kHz sine wave from a 48kHz sample?  ...You can't!  You can get a waveform with an average frequency of 17kHz.  And, that's good enough for regular audio reproduction...  I'm not trying to start-up another sample-rate debate. 

A 17kHz sine wave has alternating positive and negative peaks every 29.4 uS (a period of 58.8 uS).  At 48kHz, you get one sample every 20.8 uS.  Those original peaks are lost as soon as you sample.  There is just no way that the digital-to-analog converter can reproduce the original waveform!  The sample-points don't "line up" with where the positive/negative peaks need to be. 

I know Nyquist[/u] (http://en.wikipedia.org/wiki/Nyquist_frequency) says you can theoretically reconstruct the original 17kHz sine wave (given enough samples over a long-enough time).  But, DACs and soundcards don't do that kind of mathematical analysis or processing...  They just spit-out a series of data-points (analog voltages), one at a time, with low pass filtering to smooth-out the waveform.

I'm not trying to start-up another sample-rate debate.

Then why post such nonsense?

How do you get a 17kHz sine wave from a 48kHz sample?  ...You can't!  You can get a waveform with an average frequency of 17kHz.  And, that's good enough for regular audio reproduction...  I'm not trying to start-up another sample-rate debate. 

A 17kHz sine wave has alternating positive and negative peaks every 29.4 uS (a period of 58.8 uS).  At 48kHz, you get one sample every 20.8 uS.  Those original peaks are lost as soon as you sample.  There is just no way that the digital-to-analog converter can reproduce the original waveform!  The sample-points don't "line up" with where the positive/negative peaks need to be. 

I know Nyquist[/u] (http://en.wikipedia.org/wiki/Nyquist_frequency) says you can theoretically reconstruct the original 17kHz sine wave (given enough samples over a long-enough time).  But, DACs and soundcards don't do that kind of mathematical analysis or processing...  They just spit-out a series of data-points (analog voltages), one at a time, with low pass filtering to smooth-out the waveform.

This isn't an accurate description of how a DAC works at all.  PCM values are NOT voltage values of the waveform over time.  They are an impulse train that can be used to create a given waveform.  They are NOT values of that given waveform.  The fact that the waveform and the PCM values look nearly identical at values far less than Nyquist is what gives many people the impression that PCM values are supposed to mirror the analog waveform.

The sample-points don't "line up" with where the positive/negative peaks need to be.

Who said you'd have to do that?
(http://img247.imageshack.us/img247/6/17000jq4.png) (http://imageshack.us)

17K with speakers and background noise.  34 years old.

18k in the left ear, 17k in the right. 26.

19k, 23.

Quote from: AliL on 2008-08-20 18:22:20

I don't know if it's because I'm only 19, or my laptop's crummy soundcard or my crappy 3 year old iPod stock headphones, but I could hear all the frequencies from right at the beginning.

And from what I can tell, the sound doesn't start until he starts to say the word "two" in twenty-two.

Yeah I looked at the waveform in an editor and that's exectly where it starts, it's just after the word twenty and exactly at the begining of the word two. You must be part bat AliL

Thank you  *starts singing the batman theme*

However, Im confused. With the sample on the website I heard the noise right from the off, but using benski's file I can only hear it from just before the 18KHz, I guess about 18.2KHz. Which is the correct value?

I still couldn't hear anything at 12KHz.

64, using speakers. Tinnitus a little bit loud this morning.

Not too surprised, since on another test I started hearing the tone between 11 and 12 K.

Or, why I don't help with listening tests.

However, Im confused. With the sample on the website I heard the noise right from the off, but using benski's file I can only hear it from just before the 18KHz, I guess about 18.2KHz. Which is the correct value?

What kind of soundcard do you have?  There's definitely audio right from 22khz in the file I posted.

(http://shup.com/Shup/57764/108720184429-Frequency-Analysis.png)

That's the thing, I don't really know what soundcard i have. It's inbuilt into my 5 year old laptop. A Toshiba Satellite A30-303. When I bring up the sounds menu in control panel and the device manager, it says Realtek AC97 Audio. I don't know if that means anything to you, but it means nothing to me.

42 years old, 16kHz was the first one clearly audible (17kHz was more of a sense that there is a sound).

17khz at 24 years old.

About 16.5 to 17 khz, age 28.

@AliL
Unlike the one posted on the site, benski's file should work properly with all soundcards.

Hi,
Okay, here goes. I am the youngest member yet to have taken this test (I'm only 15).
I was able to hear a distinct shreek right when the voice started to say 21K, which doesn't surprise me because my similar personal tests are showing the same thing.
BTW, my stepdad, who is almost 47, and who used to listen to loud rock music with headphones, cannot hear 15KHZ tones, when my stepsister had her ears covered and her face buried in her shirt...

21khz @ 22 yo  I have tried with better headphones. Scary stuff

This is a neat test. Using speakers I heard 19k with the original file, and on benski's 48k file I heard 21k. I'm 36 and have played in loud bands since being teenage. I started using earplugs sometime in my 20s, but still I have slight tinnitus symptoms since about 4 years ago.

I heard a tone all the way through the original WAVE.

With Benski's I started hearing a tone after he said 20K and just before he said 19K.

I'll have to re-read this thread to see what this means, but from greynol's post above I guess I should only pay attention to Benski's file.

I am 38, listening on headphones.  I'm not sure that I've admitted my age on this board before. 

BTW, using foobar's tone:// feature (File > Add Location... > "tone://22000") I can hear tone://23000 but not tone://24000.  Is this relevant?

Edit: Hmm, just used "sweep://24000-10000,24" and that produces all sorts of weird and wonderful effects, so I guess this is still being affected by my soundcard (work machine, Aida32 reports it an ALi M5455 AC'97 Audio Controller).

I'd feel better if one of the R&D experts can vouch for my claim, though I've never heard of a soundcard that plays 44.1 natively but resamples 48.  Then I guess someone might inquire about the resampler Benski used.

For the record, just over 17kHz and I'm 36.  My soundcard sucks (sblive 5.1) as do my speakers but I would be surprised if they couldn't reproduce frequencies above what I can hear.  Maybe one of my neighbors will let me borrow one of their kids.

FWIW, I can hear aliasing in the 44.1kHz version but not in the 48kHz version so long as I'm not using an external resampler like what is in foobar2000.

16.5k. Couldn't resist.

My job for 2 yeas was to play back loud music over a smiley face EQ'd system. I wonder if this has anything to do with the poor result.

I've never heard of a soundcard that plays 44.1 natively but resamples 48.

Well, the soundcard could have stuck in 44.1 kHz mode, or doesn't support auto switching at all.

I guess it would be wrong to assume that improper playback would only result in people thinking they can hear over a greater range of frequencies and never result in a reduced range of frequencies?

17Khz on open back headphones in an fairly noisy office..
Oh, and 39 years old..

Benski,

Out of curiosity, how did you resample the file? It appears to have a very slight lowpass filtering applied (at about 21.8 kHz).

When you posted your version I already had created a 48 kHz version with Audition and its sweep starts about 500 ms earlier because the file is not lowpassed.

)

I'd feel better if one of the R&D experts can vouch for my claim, though I've never heard of a soundcard that plays 44.1 natively but resamples 48.

The original SB 16 and SB 32 AWE ISA sound cards supported only up to 44.1 kHz sample rates. I have not used my SB 32 AWE for ages, but I suppose a recent Windows version would resample 48 kHz to 44.1 kHz. I don't know if the Windows versions from early 90's would be able to play 48 kHz at all with these sound cards.

At some stage Creative's later PCI models changed to support 48 kHz instead of 44.1 kHz. I think this was a part of Microsoft's multimedia PC specifications.

Edit: As far as I know, only the "creation mode" in Creative's X-fi series supports 44.1 kHz again natively.

.. Creative's later PCI models changed to support 48 kHz instead of 44.1 kHz. I think this was a part of Microsoft's multimedia PC specifications.

I think it had to do with Intel's AC'97 specification.

Probably you are correct.

Microsoft's Multimedia PC specifications Level 1, 2 and 3 from the 90's specify sample rates only up to 44.1 kHz. I didn't find any newer "levels". (Isn't Google wonderful?)

Benski,

Out of curiosity, how did you resample the file? It appears to have a very slight lowpass filtering applied (at about 21.8 kHz).

When you posted your version I already had created a 48 kHz version with Audition and its sweep starts about 500 ms earlier because the file is not lowpassed.

Foobar2000 PPHS ultra mode. 

On my file I can hear* a slow fade-in when the tone starts.  Thought this might have been because I couldn't hear higher, but looking at the waveform indicates that this might be an effect of the low-pass filtering

* really, at this frequency, I feel it.  I don't recognize it as a musical tone until around 19.5khz

How do you get a 17kHz sine wave from a 48kHz sample?  ...You can't!  You can get a waveform with an average frequency of 17kHz.  And, that's good enough for regular audio reproduction...  I'm not trying to start-up another sample-rate debate. 

A 17kHz sine wave has alternating positive and negative peaks every 29.4 uS (a period of 58.8 uS).  At 48kHz, you get one sample every 20.8 uS.  Those original peaks are lost as soon as you sample.  There is just no way that the digital-to-analog converter can reproduce the original waveform!  The sample-points don't "line up" with where the positive/negative peaks need to be. 

I know Nyquist[/u] (http://en.wikipedia.org/wiki/Nyquist_frequency) says you can theoretically reconstruct the original 17kHz sine wave (given enough samples over a long-enough time).  But, DACs and soundcards don't do that kind of mathematical analysis or processing...  They just spit-out a series of data-points (analog voltages), one at a time, with low pass filtering to smooth-out the waveform.

Doug that is such a common misconception, I see it everywhere (and especially on pro-vinyl forums) but it's just not true. You are seriously underestimating how good is the analog re-construction of modern soundcard DAC's.

Let me simply post an oscilloscope photograph of the actual  output (line-out) voltage coming from the onboard sound (realtek AC97) of my computer. This sample was taken while playing the wave-file from this very listening test. I took the shot midway between the 21kHz and 20kHz voice over, as you can see the wave is about 20.5 kHz. Looks like a pretty good sine wave yes.

We have to try and dispel this myth.

see image here :
http://img213.imageshack.us/my.php?image=soundcardoutbl9.gif (http://img213.imageshack.us/my.php?image=soundcardoutbl9.gif)

though I've never heard of a soundcard that plays 44.1 natively but resamples 48.

I use the ESI Juli@ soundcard in my HTPC. It always plays with the correct sample rate and does not resample.
The M-Audio Delta Audiophile 192 and the RME Digi 96/8 Pad are also soundcards that do not resample. I've tested all three - the RME Digi 96/8 Pad I use in my main PC.
The ESI Juli@ I can recommend most because card+drivers are very good and stable and allow a use in all areas (WDM DirectX ASIO...).

Sure, though I never suggested there weren't cards that could do both natively.

If it's any consolation, I own a Delta 1010.

Ummm...you all realize that this test is horrible inacurate?
Can your speakers properly output 20kHz and are they calbrated (referenced to 1Khz @ 0dB HL)?
--This is considered the Audiological standard.

Tried this again with headphones at work.  Using Benski's file, I hear a sound as the speaker is finishing up saying 22, it kind of sounds like it's winding up through when the speaker says 21. Then I don't hear any tone again until 18, where it's then constant.

Sure, though I never suggested there weren't cards that could do both natively.

If it's any consolation, I own a Delta 1010.

Now I got it. Seems I did not pay attention enough when reding your post the first time...
English is not my native language.

Anyway, I really have to back away from my claim that Benski's file should work for everyone.  Besides the hardware, there are other factors to consider such as drivers, software and OS considerations.  I have no expertise on this subject.

Then I don't hear any tone again until 18, where it's then constant.

Then, that implies that you really can only hear up until 18Khz, and the other sounds are not the tone, but bad reproduction of the sample.

I can start to hear a tone right before the person says 18k.
I'm using a SB X-Fi XtremeGamer set in "audio creation mode" and bit-matched playback, I don't know if this is correctly playing back the file.

Quote from: greynol on 2008-08-21 09:26:27

though I've never heard of a soundcard that plays 44.1 natively but resamples 48.

I use the ESI Juli@ soundcard in my HTPC. It always plays with the correct sample rate and does not resample.
The M-Audio Delta Audiophile 192 and the RME Digi 96/8 Pad are also soundcards that do not resample. I've tested all three - the RME Digi 96/8 Pad I use in my main PC.
The ESI Juli@ I can recommend most because card+drivers are very good and stable and allow a use in all areas (WDM DirectX ASIO...).

Have you thought about what actually happens when two applications each simultaneously play sounds with different sample rates? The driver/OS/card then HAS TO resample for obvious reasons.

I was able to hear up to 17k, and I am 17.  But I was using an integrated sound card (I am using M-Audio Sudiophile AV 40 speakers), so I don't know if that would affect it

no for 22, yes for 21 to 18, then no for 17, then again yes. hmmm.

no for 22, yes for 21 to 18, then no for 17, then again yes. hmmm.

If that happened with the sample on the webpage, try with the one benski posted in this link http://www.hydrogenaudio.org/forums/index....65398&st=0# (http://www.hydrogenaudio.org/forums/index.php?showtopic=65398&st=0#)

But from your description, it probably is 17Khz for you too.
Of course, it also means that your equipment can't properly reproduce 44Khz content.

I can barely hear 15k I'm 30 years old
Edit; with my AKG 240 and SB x-fi i can hear ~17K

Have you thought about what actually happens when two applications each simultaneously play sounds with different sample rates? The driver/OS/card then HAS TO resample for obvious reasons.

That's true, in this case it has to (multiple sounds with different samplerates played through the same channels).

Between 16 and 17 Khz. I'm almost 28. I guess this is the result of too much loud metal.



I guess this test may be biased depending on the sound card, earphones etc...I had a hearing test with a doctor about 3 years ago and he told me my hearing was very healthy...well maybe it's no longer.

EDIT: When opening the file with winamp I was able to hear starting from 18Khz, it's a pretty annoying sound.

' date='Aug 21 2008, 22:49' post='583990']

Quote from: smok3 on 2008-08-21 21:37:06

no for 22, yes for 21 to 18, then no for 17, then again yes. hmmm.

If that happened with the sample on the webpage, try with the one benski posted in this link http://www.hydrogenaudio.org/forums/index....65398&st=0# (http://www.hydrogenaudio.org/forums/index.php?showtopic=65398&st=0#)

But from your description, it probably is 17Khz for you too.
Of course, it also means that your equipment can't properly reproduce 44Khz content.

got 16k with that sample and i'am not 17 anymore no.....

Ummm...you all realize that this test is horrible inacurate?
Can your speakers properly output 20kHz and are they calbrated (referenced to 1Khz @ 0dB HL)?
--This is considered the Audiological standard.

I think everyone is fairly clear that this test is, like the subject says, simple. But it's a useful quick and dirty guide to what can be heard on one's own equipment.

I see people beating up on themselves because they can only hear 15KHz. I, as the definite low scorer on this test so far, console myself with the thought that the diff between me and a super-sharp teen is just one octave. More realistically, two harmonics on the top note on the piano. Any idea how much difference this would make to the timbre of the sound? For obvious reasons, this is one I can't test for myself.

Somewhere between 18k and 19k on both ears, 27 years old.  I'm fairly happy about that, considering the hears of playing in orchestras with unprotected ears, shotgun blasts while hunting, and loud concerts.

These noises are WAY painful though.  I think I'd rather not hear them.

Doug that is such a common misconception....

...We have to try and dispel this myth.

see image here :

uart,

Wow!   That looks GOOD!!!!!  Thanks for posting that image.    Your output is much better than my soundchip...  (Also says Realtek AC'97).  I wish I could capture the output from my 'scopes, but I don't have the capability*...  I hope you don't think I'm lying about my results! 

When I posted before, I was "talking theory".  I had never hooked a 'scope to a soundcard.  But, I knew I was going to stur-up trouble,  So, I decided I should try it (which I did before I read your post).

I made a 17kHz file. Then I hooked up an analog 'scope, which happens to be on the same bench as the only computer in my "work area" with a soundcard.

There was quite a bit of "jitter", especially amplitude jitter, which made the analog 'scope almost useless.  But, although the waveform image was messed-up I could see that the waveform was fairly rectangular and stair-stepped.  I didn't expect that... I expected the low-pass filter to do a lot more smoothing.

So I brought-over the digital 'scope which can capture a single sweep, jitter or not.  When I captured single-sweeps, the waveform looked alot like the zoomed-in waveform on my audio editor, except th stair-steps were slightly rounded.    The digital 'scope has a frequency counter, but the jitter wouldn't allow a steady reading.  It was jumping between about 16kHz and 25kHz.  The period was about 62uS which works-out to about 16kHz...  Which is 3 samples at 48kHz.

Since the stair-steps were clearly present, and the filtering wasn't doing anything...  all of the cycles and half-cycles were exact multiples of the sample rate.


* My digital 'scope has a GPIB connecter (and I have GPIB on my computers, but it's not hooked-up and I'd have to find the software install it, and figure out how to use it.  I don't even know if we have the GPIB interface software.  We bought it used.

17k/high 16k here, not bad for the abuse my ears have been through over the years and dropping noticeably from 'excellent' during testing in my military entry days to 'downright horrible' now (things like catching half of what my wife says even when giving her my full attention and in close proximity)
Perhaps I should have done more than 'wear' my earplugs very loosely when working in those loud mainspaces and around whizbang pumps. 

interesting, those emoticons do not work when JS is disabled

18kHz at 22 years old.  I think I've been to a total of two live concerts in my life, and I really can't think of anywhere else I would've experienced prolonged exposure to unhealthily loud noises.  Interesting just how quickly you can lose that section of your hearing without even doing anything bad to them.

I think you're doing OK, Mickey.

I've selectively chosen the results from this thread that best "fit the curve" to make the simple table below and you seem to be about average.



PS I realise that this table represents the results from a non-scientific test, so please don't flame me.

i'm 24 and i can hear 22k when i crank the volume.

DVDdoug,

I think we need to get to the bottom of what you are seeing, because it doesn't make any sense. A 17 kHz signal is very close to the lowpass cutoff frequency of your soundcard, and as such it must necessarily be a near-perfect sine wave, because any distortion at all would imply the presence of frequencies well above the cutoff frequency.

Is it possible that your soundcard is resampling (poorly) your signal and introducing beat frequencies, which would explain fluctuations in amplitude and period? Unfortunately this would not explain stair-steps in the waveform, which, as I said above, should not be possible because of the lowpass.

Sorry for going so-far off-topic, but I'm still thinking about this....

BTW - When I did my single-frequency 17kHz experiment with my lousy soundcard and cheap speakers, I was hearing sub-harmonics... Maybe the tweeters rattling/resonating.    I don't know if I can hear 17kHz, but I was hearing something way below that! 

Quote from: DVDdoug on 2008-08-20 22:27:19

The sample-points don't "line up" with where the positive/negative peaks need to be.

Who said you'd have to do that?
[a href="http://imageshack.us" target="_blank"] And, when I did the experiment above, I did get a ship-load of amplitude jitter...  The peak-levels were jumping all over the place.

I just mentioned the "peaks", because it's easy to visualize what you're missing.  Once you've sampled, you don't know where (in time) the peak was, or its amplitude.  It's not just the peaks that are lost...  A digital audio file doesn't contain any information between the samples.  The ADC never even "looked at" the data between the samples!  If you "pretend you are a DAC" and "connect the dots", and add some smoothing (filtering), you don't get the original waveform.

Note that DACs, don't directly connect the dots...  The DAC will hold each value 'till the next sample.  So you get rectangular steps, with vertical & horizontal lines (before filtering), rather than triangular or angled lines.  I use GoldWave, and when you zoom-in it shows the waveform as rectangular steps.  But, this is choice made by the program developer.  The "real data" is just a series of sample points (i.e. "dots" on a graph).

Now, it turns-out that the only pure sine wave that will exactly fit those "dots" is the original sine wave, and higher alias frequencies.  But, my soundcard is't that smart, and most of my files aren't pure sine waves anyway.  So, who knows...  Maybe the original audio "looked" more like the "connect-the-dot" result, than a pure sine wave.    All we know is that the original waveform passed-through those data points, and we know that there is nothing "complicated" between the sample-points, because the original waveform was (hopefully) low-pass filtered before sampling.

If you "pretend you are a DAC" and "connect the dots", and add some smoothing (filtering), you don't get the original waveform.

Please re-read my reply to your post.  You do not understand how a DAC works nor how PCM works.  Please stop posting this nonsense.

See: http://www.dspguide.com/ch3/2.htm (http://www.dspguide.com/ch3/2.htm).  It actually has an example almost exactly like Raiden's screenshot. 
From the text:

[T]he situation is made more difficult by increasing the sine wave's frequency to 0.31 of the sampling rate. This results in only 3.2 samples per sine wave cycle. Here the samples are so sparse that they don't even appear to follow the general trend of the analog signal. Do these samples properly represent the analog waveform? Again, the answer is yes, and for exactly the same reason. The samples are a unique representation of the analog signal. All of the information needed to reconstruct the continuous waveform is contained in the digital data.  Obviously, it must be more sophisticated than just drawing straight lines between the data points. As strange as it seems, this is proper sampling according to our definition.

All we know is that the original waveform passed-through those data points

The PCM samples are not voltage values that the original waveform passed through.  They are impulses that allow the DAC to reproduce the original waveform.  The PCM samples are taken after the Sample-and-Hold circuit and an anti-alias circuit and are not necessarily identical to the original waveform.  They just happen to match closely when you're looking at waveforms much lower than Nyquist.

Just a word of warning, don't crank up the volume on freqs you can't hear.  Just because you can't hear it doesn't mean high energy waves aren't hitting your eardrums.

With Benski's I started hearing a tone after he said 20K and just before he said 19K.
...
I am 38, listening on headphones.

Just tested with different headphones on my laptop at home and this time it was 16K. 

I'm assuming that level is more likely (given that I used to play the drums and have listened to personal stereos way too loud for around 20 years), but this difference caused by hardware is frustrating.

It is frustrating.  I grabbed a microphone and recorded the output from my speakers.  Sure enough, they could reproduce all the frequencies.  The response was far from flat, but my setup was crude.  There was still plenty of energy present in the frequencies I could not hear, so I know that the equipment isn't influencing my results.

I remember doing similar tests in my late 20s and being able to hear closer to 18kHz.  I'm not concerned with what is normal high frequency loss and I feel pretty good that I haven't lost all that much.  What I do find annoying is that I have constant ringing in my ears at a very high frequency (~16+kHz?).  Just guessing, but I think I can hear it over a background level of 60 perhaps 70 dB.  Listening to these test tones doesn't help any either.

@DVDdoug
You should familiarize yourself with the impulse response of a low pass filter.  It would seem that you've had no formal training in signals and transforms.
http://en.wikipedia.org/wiki/Sinc_function (http://en.wikipedia.org/wiki/Sinc_function)

Quote from: Martel on 2008-08-21 21:04:07

Have you thought about what actually happens when two applications each simultaneously play sounds with different sample rates? The driver/OS/card then HAS TO resample for obvious reasons.

That's true, in this case it has to (multiple sounds with different samplerates played through the same channels).

And how can you be so sure that the waveform doesn't get resampled no matter what (even when a single waveform is playing)? You cannot - unless you know the exact workings of the API/OS mixer/device driver/soundcard.
Perhaps your soundcard/its driver sets the samplerate between the chip and DAC to that of the sound played first and adjusts every other sound playing simultaneously to this chosen rate. Or perhaps it just resamples everything to some fixed internal rate (and it could get worse if it resampled even the sounds that already match the internal rate!). Pair this with a lame (reckless) resampling algorithm and you get some nasty aliasing artifacts. At least some people can tell these from the actual tone.

Uhhh, just below 20k and I'm 20 years old. Sounding good. ^^
Surprised a bit 'cause I always listen my music loud as hell  but maybe it's just good practice.

I think you're doing OK, Mickey.

I've selectively chosen the results from this thread that best "fit the curve" to make the simple table below and you seem to be about average.

Yep I made a similar table and plot. I did this a few days ago when there ware only 45 replies, so this doesn't contain all the data. It does however contain all the data points (up to reply 45)  for which the respondent gave their age.

As you can see the data shows a pretty consistent pattern. Apart from just a few outliers the data is quite well clustered. My effort is also not particularly scientific, I put that approximate "line of best fit" in by hand. It seems to show typical frequency limit of about 19Khz by late teens, falling by about 1.5 kHz per decade down to about 12kHz at about 65 years old. So even though this test is crude, that data agrees pretty well with just about everything I’ve ever read about human hearing.

Take a look at the graph here :
(http://img153.imageshack.us/img153/5418/hearingkhzvsagejy3.th.gif) (http://img153.imageshack.us/my.php?image=hearingkhzvsagejy3.gif)

...but maybe it's just good practice.

Genetics play a big role too.

I had a hearing test with a doctor about 3 years ago and he told me my hearing was very healthy...well maybe it's no longer.

AFAIK they only test it from 125 (or was it 250?) Hz to 8kHz, as that's all you need to clearly understant speech

Between 16 and 17 Khz. I'm almost 28. I guess this is the result of too much loud metal.



I guess this test may be biased depending on the sound card, earphones etc...I had a hearing test with a doctor about 3 years ago and he told me my hearing was very healthy...well maybe it's no longer.

As jmartis said the Doctor possibly didn't even test the very high end. And even if he did you have to understand that the last thing a doctor wants to do is to unneccessarily alarm or stress-out a patient. As such they generally have fairly broad "normal ranges" for things like this. I think it very likely that if the doctor did test you right up to 20kHz that you'd still be very safely within your doctors "normal range".

only 16k @ 35 years  (with Benski's 48kHz-Resample - the 44,1k-sample was faulty)

It is frustrating.  I grabbed a microphone and recorded the output from my speakers.  Sure enough, they could reproduce all the frequencies.  The response was far from flat, but my setup was crude.  There was still plenty of energy present in the frequencies I could not hear, so I know that the equipment isn't influencing my results.

Yeah I definitely wouldn't trust the accuracy of this test in the over 20kHz frequency range. My sound cards response is pretty flat up to 20kHz but I observed a noticeable drop off above that (even when resampling to 48kHz). Personally I don't think that results above 20k are really that important to the average person.

PS. When I looked at the actual voltage out of my sound-card (as per my earlier post) I noticed the signal was down about 8dB at the start of the wave (22kHz), climbing to about -1dB at 20KHz. As for my speaker/headphone frequency response, that's anyone's guess.

As for my speaker/headphone frequency response, that's anyone's guess.

Yep. Gotta love those ambiguous specs the manufacturers hand out for headphones! My Sennheiser HD447s are specced from 25Hz to 21kHz. There's no clue as to how many dB down they are at those points though. I could assume those were the -3dB points, but I have no way of knowing.

Cheers, Slipstreem. 

18Khz for me, im 33.  I could somehow tell the difference a little earlier on the 44.1KHz sample than on the 48 one. Using speakers with quite a bit of ambient noise. Having worked about 10 years as a DJ and a lot of shooting in the army doesnt seem to have made its presence felt too much at least.

15kHz, right ear; 17kHz on the left. Sort of. The very high frequencies kind of break up in my left ear, like a failing tweeter. Very annoying.  I know this through professional hearing analysis.... Late 30s. Lifelong musician and former PA/HiFi level abuser. Now I'm a walking, talking low-pass filter. If this keeps up, soon I won't even have to bother looking at Hydrogen Audio, unless I feel like reading about what those still blessed with the gift of aural reception are up to.

Seriously, though, I hope you youngsters don't make the same stupid mistakes I made when I was younger. My hearing deficiency is entirely self-inflicted, unfortunately. Believe me, you don't want diminished hearing long before you're old and grey. This is easily preventable, so please do take care.

17khz here at aprox. an age of thirty. Though, HF cutoff isn't my main problem. I am too sensitive to sounds inside of my hearing range, espacially 10-13khz. I can notice sounds as low as 30hz, even at low levels.... partially because i do not just "hear" with my ears. In public transports for example, i can notice how the speech from the speakers makes stuff vibrate. People frequently ask me why i'm talking so quietly, when to me, it seems as if i'm just talking at an for me comfortable level - on the other hand, i frequently asks others to speak a bit quieter. No, i have no interest in training this ability further by participating in audio DBTs :)

I wonder how many replies I would have gotten by asking, "how old are you?"

42? 

Cheers, Slipstreem. 

32 years old and 22 khz 

Maybe I cheated using my iAudio 7 and Sennheiser MX550.... not too loud though

Using my Dell laptop and Sennheiser PX100 headphones, I could hear "something" starting between 19 and 18... I could hear the sound clearly between 18 and 17. I'm 26 years old... I guess that I either need a better pair of headphones or those concerts did more damage than I thought.

Gotta love those ambiguous specs the manufacturers hand out for headphones! My Sennheiser HD447s are specced from 25Hz to 21kHz. There's no clue as to how many dB down they are at those points though. I could assume those were the -3dB points, but I have no way of knowing.

Well, the problem is that there is no solid reference what it should be, unlike with speakers where the target is flat frequency response.  Outer ear changes quite a lot the signal that hits eardrums, most notably attenuating the high frequencies.  Since headphones feed the signal almost straight into the ear they can't have flat response.  But what should it be, ears are not all alike?

For example, Sennheiser HD-600 is generally considered to be flat and neutral sounding headphone, but measurement is not something you could call flat:
(http://www.freeimagehosting.net/uploads/4caf3795d6.png)

I see people beating up on themselves because they can only hear 15KHz. I, as the definite low scorer on this test so far, console myself with the thought that the diff between me and a super-sharp teen is just one octave. More realistically, two harmonics on the top note on the piano. Any idea how much difference this would make to the timbre of the sound? For obvious reasons, this is one I can't test for myself.

Found this link: http://www.visualizationsoftware.com/gram.html (http://www.visualizationsoftware.com/gram.html), it does a real time spectrum analysis.
Played a late string quartet by Beethoven, the bulk is below 10 kHz, hardly anything above the 15 kHz
My hearing drops somewhere between 12 - 13 kHz, so I must be between 50 - 60 I guess

Thanks very much for the reply, tot. I learn something new here almost every day.

Cheers, Slipstreem. 

33, and could hear a bit between 18 an 17 kHz.
I used resampled file, Soundblaster Audigy 2 ZX Platinum and Koss PortaPro.

- 16kHz @ 32 years old.
- normal listening volume levels (not loud).
- SBlaster 128 / Ensoniq chip
- Notice that there is a drop between 15kHz to 14kHz.
- currently dealing with an old eustachian tube blockage condition on both ears, so I may hear higher when 100% hearing is back.

17khz here at aprox. an age of thirty. Though, HF cutoff isn't my main problem. I am too sensitive to sounds inside of my hearing range, espacially 10-13khz. I can notice sounds as low as 30hz, even at low levels.... partially because i do not just "hear" with my ears. In public transports for example, i can notice how the speech from the speakers makes stuff vibrate. People frequently ask me why i'm talking so quietly, when to me, it seems as if i'm just talking at an for me comfortable level - on the other hand, i frequently asks others to speak a bit quieter. No, i have no interest in training this ability further by participating in audio DBTs

I'm with you on that. One can live without 18, 19 kHz sounds, but the increased sensitivity to lower level sounds drives me crazy as well. I too am frequently asked to "speak louder," and I'm constantly wondering why people always seem to be yelling, when in fact they are adamant that they are not doing any such thing.

There's a "silent" fan on my iMac that I find unbearable. Those with normal, more or less full frequency hearing tell me that they don't notice it at all, but to me it's an incessant, low level whine that can be maddening. I can literally step into the adjacent room and still hear it. The only way I'm able to work in here with the computer on is to either have music playing or keep an electric oscillating fan across the room going. The oscillating fan is just loud enough and emits the correct low level frequency to cancel out the computer fan.

For those who want to do some more testing (Equal loudness contours): http://www.phys.unsw.edu.au/jw/hearing.html (http://www.phys.unsw.edu.au/jw/hearing.html)

Using the extremely limited resources I have available (a chap no-name computer speaker system, and earbuds that came with a portable player), I can hear up to about 13-14kHz clearly, 15-16kHz tops.  I'm 27.

On the low frequency extention, the limit is around 50 Hz.  Something MUST be wrong with my equipment - normally with even a semi-cheap pair of circum-aural headphones (Panasonic RP-HT355 - not currently using cause they only sound in one ear), I can hear down to about 10 or 12 Hz or so.  Also, I've heard music with bass notes down to 4 Hz, and I think I'd be able to detect notes (assuming they're NOT sine waves - I'd still be able to detect the fundamental though) down to about 2 Hz or maybe a little lower.

Dynamic range - my practical limit is around 48-54dB, absolute limit about 66dB or so.  (I thought it was higher - maybe I don't crank up my portable music player as loud as I thought, or I didn't crank up the noise as loud as I normally crank up my music?)

Sound level difference: I could detect as close as 0.5dB.

Pitch difference:  This one unpleasantly surprised me.  Using their tests, I can hear a difference of somewhere between 10 and 20 cents.  However, when two not-perfectly-tuned strings on a note on a piano are played simultaneously, I can detect that they're out of tune even when they're within probably about 0.1 to 0.5 cents, if not closer.

On the low frequency extention, the limit is around 50 Hz.  Something MUST be wrong with my equipment

That can easily be, soundcards used to roll off from 80Hz down already, some may have -30db at 30Hz. Coupled that with some cheap earphones and you get it.  I could hear 20Hz fine with some sony Headphones, in that test, but the sound was clearly softer than the 100Hz one.

Also, I've heard music with bass notes down to 4 Hz, and I think I'd be able to detect notes (assuming they're NOT sine waves - I'd still be able to detect the fundamental though) down to about 2 Hz or maybe a little lower.

Sorry, but when talking about frequencies, you strictly talk about sine waves. It doesn't matter if you play a square wave at 2Hz and hear the harmonics. Hearing the fundamental is the same as saying you hear the sine wave. You cannot say you hear 2Hz.

On the low frequency extention, the limit is around 50 Hz.  Something MUST be wrong with my equipment - normally with even a semi-cheap pair of circum-aural headphones (Panasonic RP-HT355 - not currently using cause they only sound in one ear), I can hear down to about 10 or 12 Hz or so.  Also, I've heard music with bass notes down to 4 Hz, and I think I'd be able to detect notes (assuming they're NOT sine waves - I'd still be able to detect the fundamental though) down to about 2 Hz or maybe a little lower.

Unless you're an alien using alien technology, i doubt that. What you are hearing in those cases most probably is something else.

20k in left, it is faint but I can hear that eeeee noise like the others.

19k in right, faint eeee on down.  Right ear is less probably due to military abuse (Infantry for 2 years).

Male, Age 27.

Through Notebook speakers on my Lenovo Thinkpad T61p, outside on the patio with ear pointed in direction of speakers and exposed to the sounds twice with 10 second gap between exposure.

WARNING! WARNING! WARNING!

I had made such a test on my father's old 2-way Philips speakers. At about 15 kHz I coudn't hear anything and began to raise the volume. No way. After that I turned on some album to listen and . . . what the hell? The sound was terrible!

Since that test I don't know wether I can't hear anything above 15 kHz or those speakers can't (coudn't) emit anything above 15 kHz. I don't do such tests anymore - since I burned tweeters then.

BE CAREFUL!

Yeah, 21kHz on the left and 20kHz on the right!  I'm 22 by the way.

WARNING! WARNING! WARNING!

I had made such a test on my father's old 2-way Philips speakers. At about 15 kHz I coudn't hear anything and began to raise the volume. No way. After that I turned on some album to listen and . . . what the hell? The sound was terrible!

Since that test I don't know wether I can't hear anything above 15 kHz or those speakers can't (coudn't) emit anything above 15 kHz. I don't do such tests anymore - since I burned tweeters then.

BE CAREFUL!

Ouch!.

With this particular test there is a "voice over", so if you keep the level of the voice at normal listening volume then I don't really think it should be dangerous to your tweaters.

Actually I just used it to test the tweaters on my stereo (previously I did the test on my headphones). What do you know, with my ear next to the tweater I can hear 16kHz (previously I only reported 15kHz).

BTW. Both tweaters still working perfectly

22kHz

On Sennheiser HD650 Headphones connected to a headroom balanced max headphone amplifier plugged into a M-Audio Audiophile 2496 soundcard.

I'm still in my twenties though, so I guess it's not too suprising.

Quote from: de Mon on 2008-08-25 00:53:04

WARNING! WARNING! WARNING!

I had made such a test on my father's old 2-way Philips speakers. At about 15 kHz I coudn't hear anything and began to raise the volume. No way. After that I turned on some album to listen and . . . what the hell? The sound was terrible!

Since that test I don't know wether I can't hear anything above 15 kHz or those speakers can't (coudn't) emit anything above 15 kHz. I don't do such tests anymore - since I burned tweeters then.

BE CAREFUL!

Ouch!.

With this particular test there is a "voice over", so if you keep the level of the voice at normal listening volume then I don't really think it should be dangerous to your tweaters.

Actually I just used it to test the tweaters on my stereo (previously I did the test on my headphones). What do you know, with my ear next to the tweater I can hear 16kHz (previously I only reported 15kHz).

BTW. Both tweaters still working perfectly

I was listening to a 15 kHz sine - no voice.

WARNING! WARNING! WARNING!

I had made such a test on my father's old 2-way Philips speakers. At about 15 kHz I coudn't hear anything and began to raise the volume. No way. After that I turned on some album to listen and . . . what the hell? The sound was terrible!

Since that test I don't know wether I can't hear anything above 15 kHz or those speakers can't (coudn't) emit anything above 15 kHz. I don't do such tests anymore - since I burned tweeters then.

BE CAREFUL!

Ouch, that reminds me of what happened to grbmusic in this (http://www.hydrogenaudio.org/forums/index.php?showtopic=9772) topic.

12 KHz @ 17 Years. ;D And I'm listening to metal like the whole day. oO
Everything is fine.

12khz and you're only 17.

:S

12-13 khz and I'm only 59.

12 KHz @ 17 Years. ;D And I'm listening to metal like the whole day. oO
Everything is fine.

That's a matter of personal opinion. Personally, I'd be very worried indeed if my hearing was over three times as old as I was. Let's hope that it's at least partly down to your equipment and not entirely down to your ears. 

Cheers, Slipstreem. 

I was listening to a 15 kHz sine - no voice.

It is a bit strange that a 15Khz signal would damage the tweeters (i would expect a higher one do so), but it indeed can happen if you don't hear it and put it at high volume over a continued period of time.
This happens because tweeters both get hotter faster, and are not supposed to get as high amplitude signals as woofers.
The test pointed in this thread, though, does provide for a way to not go too far on the volume, since there's the voiceover. If the voice distorts, you know you've pulled the volume too high, so if you're not hearing the frequency, you won't.  It is some sort of reference level.

He may have simply pushed the amplifier into clipping. The high frequency energy generated by clipping never finds its way to the woofer or mid-range unit due to the crossover in the cabinet filtering it out. The tweeter generally has no such luxury and is likely to fry fairly quickly under those circumstances unless some form of power-related protection is fitted inside the cabinet by the manufacturer.

As you say, just because the guy didn't hear it, that doesn't automatically mean that the amplifier and/or tweeter weren't being driven into heavy distortion.

Cheers, Slipstreem. 

As you say, just because the guy didn't hear it, that doesn't automatically mean that the amplifier and/or tweeter weren't being driven into heavy distortion.

Why would distortion damage a speaker? All that should matter is resistive power dissipated in the speaker, as the voice coil gets too hot and warps or gets damaged in other way.

Distortion raises the high frequency content of a signal quite substantially.

Tweeters were really burned. I opened them and windings of both tweeters was all black at some parts. I was teen then and had lot of time and repaired windings of tweeters (just for fun - speakers were not worth of that effort).

P.S. It seemed tweeters were not overheated generally - windings were coiled over cylinder made of thick paper, and paper became black at some places. But black regions were just spotes under pieces of burned wire. So I think there were no total overheating - just the winding burned at some place and the paper under that place became black. (I stressed them only for 30-50 seconds at abourt 1.5-2x power of allowed)

Tweeters were really burned. I opened them and windings of both tweeters was all black at some parts. I was teen then and had lot of time and repaired windings of tweeters (just for fun - speakers were not worth of that effort).

P.S. It seemed tweeters were not overheated generally - windings were coiled over cylinder made of thick paper, and paper became black at some places. But black regions were just spotes under pieces of burned wire. So I think there were no total overheating - just the winding burned at some place and the paper under that place became black. (I stressed them only for 30-50 seconds at abourt 1.5-2x power of allowed)

It's possible that as the coils heated up, it melted through the insulation in spots and shorted some of the windings, increasing the power to some parts of the coil but completely bypassing other parts.

Great.  Now I'm hearing some kind of scratching/static-y noise coming out of my tweeters

on some of the higher frequencies, you can "feel" the tone but not actually hear it...
O_o

15 kHz, 27 years.

I wonder if it's my hearing or my equipment, the tone was a real pain (too loud) as soon as I heard it. There may be a low-pass filter @ 15-16k in my headphone amp (it's a cheap one from Thomann, between an ESI Juli@ and Sennheiser HD650). In my older tests, the tone appeared/disappeared gradually. What about yours ?

In my test the sound appeared quite abruptly at 12kHz, just before the test ended.

Here it's 17k on the left and 18k on the right ear. 38 years old.

19 kHz. I'm 31 and I used a FireWire 1814 connected to a set of Sennheiser HD 650 headphones.

edit: Is it suitable for this purpose to use fb2k's tone generation and the ABX tool between these different tones? Maybe it would result in a better test.

Is it suitable for this purpose to use fb2k's tone generation and the ABX tool between these different tones? Maybe it would result in a better test.

Sorry... ABX what? I don't follow your logic. (and using a tone generator has the previously mentioned problems)

ABX different tones and digital silence. I just think it's a better test than this, although not so imediate. I'll go the longer route and generate these tones in audacity.

ABX different tones and digital silence. I just think it's a better test than this, although not so imediate. I'll go the longer route and generate these tones in audacity.

Sincerely, i don't see where it is better to play a steady signal versus a frequency sliding down signal.

If the latter fails, the former does for the same reasons. Plus, steady signals make it easier to either break your speakers, or say you've heard something at an absurdly high volume.

I can see wanting a statistically valid test of being able to distinguish between a tone and no tone. Somebody could want to think that they are hearing something badly enough to fool themselves into believing it, but an ABX test could prove otherwise.


I think that the test tones would need to be combined with a reference (such as the voice in this test) to make sure the volume is not turned up to a ridiculous level.

This still wouldn't help determine whether people were just hearing aliasing.  I think for a down-n-dirty test, this is adequate.  For something better, consult an audiologist, though I don't know if they measure super high frequencies.

I feel bad for anyone blowing their speakers.  I did my best to warn people that these tones are extremely loud.

I don't think there's any reason for you to feel bad, Greynol. The voiceover starts at the same time as the tone and anybody running the test at what would be considered a normal volume for any other digital sound file can't possibly have damaged anything as a result, IMO.

Cheers, Slipstreem. 

Quote from: tuxWishful on 2008-08-25 12:30:41

12 KHz @ 17 Years. ;D And I'm listening to metal like the whole day. oO
Everything is fine.

That's a matter of personal opinion. Personally, I'd be very worried indeed if my hearing was over three times as old as I was. Let's hope that it's at least partly down to your equipment and not entirely down to your ears. 

Cheers, Slipstreem. 

I don't know what the problem is, I have downloaded the track and I can hear EVEYTHING until the end. What's the problem?

You said 12 kHz.  IIRC, the sample doesn't go this low, so I suspect what you wrote was just a typo.

Just tell me if it's ok that I hear everything til the end of the track oO (I have downloaded it, and saw the seconds running)

If you heard a steadily decreasing tone beginning at the "two" when the guy says, "twenty-two kilohertz," then your results are not only good, they are exceptional.

I can hear the decreasing tone beggining at the "twenty".

There is no tone when he says twenty.  Open up the track with a wave editor and see for yourself.

@greynol: I guess he means the "twenty" after the "twenty-one", not the "twenty" of the "twenty-two". (20Khz)

@tuxWishful: The test is not meant to tell *until* you hear, but *since where* you hear. I.e. the test is reversed, and the further it goes, the more common it is to hear the tone.

To make it more clear:

There is a tone that starts at 22Khz, and steadily* decreases until the 12Khz (or whatever is the last frequency).  What is important is to know when you *start* to hear the tone, not when you end.
The male voice is there to tell you which tone are you hearing as well as to have a volume reference.


*(i.e a single tone decreasing, not several tones going now up, now down)

I just checked.  It goes down to 11kHz.  Thanks for the clarification, JAZ.

A little above 15 kHz. I'll be 48 years old in 3 months.

I am very worried...when I first took this test upon it being posted I was somewhere between 18 and 17 for the right and 17 for the left ear. Now I am struggling to hear 16 in both ears...and I am 21 :S

I think its time to go back and see a hearing specialist :S :S

I am very worried...when I first took this test upon it being posted I was somewhere between 18 and 17 for the right and 17 for the left ear. Now I am struggling to hear 16 in both ears...and I am 21 :S

I think its time to go back and see a hearing specialist :S :S

If you are woried, by all means see a hearing specialist.
Just keep in mind, this test is in no way acurate.

17k today, im only 19   

ive tested my hearing on numorus occasions before, and its usually between 17 and 18

edit: took a shower and cleaned out my ears, heard 18 solidly

A little above 19 kHz. I'm  32 years old

I used a Sound Blaster X-FI Xtrem Music --> Amp. Denon PMA1500R --> Sennheiser HD 590 headphones.

Let's see on this test I percieved the sine tone between 17 and 18 kHz. I am 21 years old btw. 

A little above 19 kHz. I'm 32 years old

That's impressive. 

I'm 52 and start hearing something at about 14 kHz.

I, as the definite low scorer on this test so far, console myself with the thought that the diff between me and a super-sharp teen is just one octave. More realistically, two harmonics on the top note on the piano. Any idea how much difference this would make to the timbre of the sound? For obvious reasons, this is one I can't test for myself.

The top note of the piano has it's first (lowest) partial slightly over 5 kHz. The third partial of that note would be above 15 kHz. But the amplitude of the higher partials in the top octave is very low. The third partial of the top note of the piano just isn't there.

However. The overall timbre of the piano is built up of a lot of things other then the sound from the strings. Sounds from the action, hammers and keys as they are played, several different types of noise from the dampers, metallic sounding duplex strings (a kind of resonating strings), longitudinal frequencies is the bass strings etc etc. These sounds are very important to the piano timbre, as we perceive it.

Still I don't think there is much information above 12-15 kHz in the overall piano timbre to worry about it. Practically none, I would say without having anything else than an educated guess to base it on.

At least I'm telling myself that, since I work professionally as a pianotechnician. 

@soundberg
Thank you for the reassurance. 

As the harpsichord is notoriously difficult for some lossy codecs, do you happen to know if it has more energy in the high frequencies? Some of the few that I've heard live have sounded downright rattly.

I start hearing the sound at 17k, I'm 24 years old. Is this good or bad? At which frequency should a healthy, young person start hearing the tone?

ps - oh, okay, so 20Hz to 20KHz should be the ideal? Does anyone hear 20K?

What you hear in this test is the combined limit of your ears, your headphones and the equipment driving your headphones. It's not scientific proof that you have an upper frequency hearing limit as suggested by the test.

Cheers, Slipstreem. 

As the harpsichord is notoriously difficult for some lossy codecs, do you happen to know if it has more energy in the high frequencies? Some of the few that I've heard live have sounded downright rattly.

It does, Woodinville has summed up it nicely, zie the bottom of: http://thewelltemperedcomputer.com/SW/audio_formats.html (http://thewelltemperedcomputer.com/SW/audio_formats.html)

20k at 19yo, with average headphones and a RealTek onboard sound card (:|)

I can hear the tone at 19k, am 22 years old.

Quote from: MichaelW on 2008-11-02 03:37:32

As the harpsichord is notoriously difficult for some lossy codecs, do you happen to know if it has more energy in the high frequencies? Some of the few that I've heard live have sounded downright rattly.

It does, Woodinville has summed up it nicely, zie the bottom of: http://thewelltemperedcomputer.com/SW/audio_formats.html (http://thewelltemperedcomputer.com/SW/audio_formats.html)

I didn't know the that harpsichord was difficult to encode, interesting. Is there some samples?

As you may know the harpsichord is basically a plucked string instrument. The plucking (the attack of the note) generates a lot of high frequencies, I would say. I'm not sure that the same can be said of the decay part of the note, not generally. The bass strings in a piano also has a lot of high partials.

If this is going off topic I humbly accept appropriate correction.

Quote from: MichaelW on 2008-11-02 03:37:32

As the harpsichord is notoriously difficult for some lossy codecs, do you happen to know if it has more energy in the high frequencies? Some of the few that I've heard live have sounded downright rattly.

 

It does, Woodinville has summed up it nicely, zie the bottom of: http://thewelltemperedcomputer.com/SW/audio_formats.html (http://thewelltemperedcomputer.com/SW/audio_formats.html)

Thank you, @Roseval. I also found the following post of interest:
http://www.hydrogenaudio.org/forums/index....st&p=588119 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=64724&view=findpost&p=588119)

17khrtz, 20 years old.

I also work in the live sound industry so yah, my hearing isn't what it used to be.

Interesting post.

Between 19-18 khz. 22 years old. I can detect the change with my two ears.

Quote from: JRace on 2008-08-21 17:59:10

Ummm...you all realize that this test is horrible inacurate?
Can your speakers properly output 20kHz and are they calbrated (referenced to 1Khz @ 0dB HL)?
--This is considered the Audiological standard.

I think everyone is fairly clear that this test is, like the subject says, simple. But it's a useful quick and dirty guide to what can be heard on one's own equipment.

I see people beating up on themselves because they can only hear 15KHz. I, as the definite low scorer on this test so far, console myself with the thought that the diff between me and a super-sharp teen is just one octave. More realistically, two harmonics on the top note on the piano. Any idea how much difference this would make to the timbre of the sound? For obvious reasons, this is one I can't test for myself.

I know I've come to this a little late but I might have you beat in the low scorer category  . I'm 48 and can hear nothing - even at 12 khz- in any of the samples I downloaded. I used what are probably basic headphones(Sennh. HD 457s) & still get nothing but that erratic high-pitched humm I always hear(tinnitus-I assume).  I'll try for the same consolation you mention but I love listening to my music and will always be wondering "What am I not Hearing?"

20 Khz (not 22 Khz like I have posted hours before). 20 years old. These last years I cared a lot of my ears but when I are 13-14 old expose my ears to extreme loudness on my headphones (80-90 dB    [if not more]). I'm gone to two rock concerts this year, the last one of Scorpions on September.
My soundcard was set to 44 Hz and I made the mistaken of play the 48 Khz sample, after I noticied this make the test again with the 44.1 Khz sample and I can hear the 20 Khz sound but it's very quiet compared to the signal of 19 Khz. Anyway I don't have an equipament that can reproduce sounds above of 20 Khz (how I don't noticied this before?), so why I hear the sound at 22 Khz? It is not a question at all, I have read some posts above and  the answer to it.
Sorry by the mistake and by my bad english.

I can hear 17 kHz at normal listening levels and between 19 and 18 if I turn the volume on my headphone amp up a bit.

18khz. I'm 37. So pleased after abusing my ears for years

If you heard a steadily decreasing tone beginning at the "two" when the guy says, "twenty-two kilohertz," then your results are not only good, they are exceptional.

You just made my day     

22khz, both files, tested on 3 different computers with completely different audio setup.
I'm 17 though, so I guess it shouldn't be too odd, but still,      .


BTW, I personally take a lot of care when it comes to loud noises; so, I'll get a couple friends(who listen to extremely high volumes) on the morning and have them do the test, will post the results if anything interesting.

You just made my day     

22khz, both files, tested on 3 different computers with completely different audio setup.
I'm 17 though, so I guess it shouldn't be too odd, but still,      .

Huh - all these youthful posters make for depressing reading 
13kHz... @ 51
30 years of wind noise from riding bikes too fast I suppose!

Edit:
Just tried the low range test on in ear phones (V-moda Vibes) & get down to 20Hz, so it's not all bad...

48 yrs old:

hearing starts just before "13k" from the downloaded 44.1 .wav file , played by foobar2k (not using a  resampler), over an integrated Dell 'soundcard" (devices is listed as "SigmaTel High Definition Audio CODEC'), 'puter running XP Pro,  Koss TD61 headphones , in a room that gets considerable ambient street noise. I hear the tone first in my right ear, then my left.

I tried the 48 kHz  but the initial voice over broke up, suggesting it wasn't playing back correctly.

I'm going to retry on my MAudio 2496 at home, with better headphones and quieter room, but I'm not too surprised -- I seem to be in the 'normal' range for my age, also being a musician who played loud rock and listened to same, and noting too the voice over still sounds like it has lots of sibilants to me. 

Still, for my future I wonder about the state of hearing-aid technology -- is it good enough to 'restore' extended high-frequency hearing , or does it focus on just the 'vocal' midrange?


Oh, and BTW, the ABX part of the website doesn't seem to work -- it won't return my 'scores'.  Anyone else seeing that?  I'm using Firefox.

Oh, and BTW, the ABX part of the website doesn't seem to work -- it won't return my 'scores'.  Anyone else seeing that?  I'm using Firefox.

Same thing here. I was going to try IE but got distracted ...

My hearing cuts off between 17-17.5kHz. The modulated 17.4kHz mosquito tone was easy to hear.

Still, for my future I wonder about the state of hearing-aid technology -- is it good enough to 'restore' extended high-frequency hearing , or does it focus on just the 'vocal' midrange?

One of the absolute state-of-the-art hearing aids today only amplifies to 7kHz (Phonak Exelia CIC).
In Canada this retails for $3700.

Nothing well ever "restore' your hearing.

And if your hearing is bad enough (esp if you worked in noise/gunfire) no amount of gain will help as you develop "dead" spots in your cochlea. Amplifing these dead spots results in the neighboring hair cells distorting. Just imagine what this does to your perception of hearing.

Once its gone people, its gone for good.

18k, 37 years old.

18khz. I'm 37. So pleased after abusing my ears for years

OK, 18kHz and 31 (with Sennheiser HD 202). Not too bad considering that because of my work I have to spend quite a lot of time in stamping factory and near the deburring machines + plus I visit rather often rock-contserts, clubs and live-music pubs... 

But my girlfriend (2 years younger) can check in only at 14...15 kHz 
And she haven't experienced nearly that hostile working conditions.

Oh, and BTW, the ABX part of the website doesn't seem to work -- it won't return my 'scores'.  Anyone else seeing that?  I'm using Firefox.

I checked the ABX scoring script and found the bug    Now corrected...  Sorry for the inconvenience and thanks for the checking!

Our blind test page (is properly working again) : http://www.audiocheck.net/blindtests_index.php (http://www.audiocheck.net/blindtests_index.php)

S.

One of the absolute state-of-the-art hearing aids today only amplifies to 7kHz (Phonak Exelia CIC).
In Canada this retails for $3700.

AFAIK hearing aids are aimed at making human speech intelligible, so they only bother with mid range frequencies.

Recently had my hearing tested, and the audiologist was only interested in the average loss between 1-8kHz. I got -11dB & -17dB, which is apparently within reasonable bounds for my age.  -25dB is the hearing aid threshold, I'm told.

16 kHz out of my PowerBook's headphone jack on a pair of SR-60s.  I'm 25.

I tried with my Shure E2Cs and only got 14 kHz.