The Windup... And The Pitch
Jun. 19th, 2006 06:36 pmRemember when we were talking about that high-pitched noise I can't hear? Well, some nice guy's got a page with MP3s of test tones at different frequencies, so you can find out where your hearing is on the high end. I've got to try these with headphones, but with my speakers I can just barely hear the 14,000 Khz one, and I mean barely. There's a link at the top of the page to a Flash version, but I can't hear the 14K one on that. On the other hand, it's got low frequencies as well. (Can't hear 15 Hz.)
So, where are you on these?
So, where are you on these?
(no subject)
Date: 2006-06-19 10:53 pm (UTC)(no subject)
Date: 2006-06-19 11:10 pm (UTC)Reason: Just like high frequency content in a JPEG image, high frequencies in audio files require more bits to encode properly. It's the MP3 equivalent of a low-quality JPEG: they're blurry. So, by rolling off frequencies above 15k, they allow for more bits to encode the stuff most people can hear the best, like vocals and bass.
Heck, it's the primary complaint that audiophiles have with the MP3 format in the first place: The lack of high freqency detail.
Additional note: Just cranking up the encoding rate of an MP3 doesn't mean it's getting the high frequencies. Most encoders roll off those frequencies no matter what, unless you tell them to do otherwise. This guy doesn't say. Maybe he doesn't know.
My point is: Maybe those frequencies aren't even in the MP3 files at all, and there's nothing wrong with your hearing.
Then there's the issue of what speakers or headphones are you previewing it on, and what's the frequency response. Maybe THEY don't go all the way up either.
(no subject)
Date: 2006-06-19 11:12 pm (UTC)(no subject)
Date: 2006-06-19 11:13 pm (UTC)(no subject)
Date: 2006-06-19 11:15 pm (UTC)My original statments still stand.
(no subject)
Date: 2006-06-19 11:20 pm (UTC)These are not certainties, but given this data, I think it is likely that he is both aware and capable of configuring his utility software.
(no subject)
Date: 2006-06-20 12:59 pm (UTC)Agreed. Viewing the files in an audio editor, I see that the 20k MP3 file actually does contain a 20k tone.
HOWEVER...
He still missed the fact that, even given the correct configuration, MP3 encoders really mangle high frequencies. He should have used WAV files for all of them, not just the highest ones. For instance, you reported not being sure whether what you were hearing was the actual tone, noise or overtones. Well, guess what? You were probably hearing a lot of noise. Here is a screen shot of the difference between his MP3 file and a 20k WAV file I generated myself:
See what I mean? At those kinds of frequencies, the slightest grit or innacuracy in the representation can color the tone and cause your ear to hear things it might otherwise not.
(no subject)
Date: 2006-06-20 01:07 pm (UTC)The second screen shot is a 22k tone. I typed the wrong value into the generator.
A 20k tone magnified in the editor doesn't look that much different than his MP3 file does. The MP3 file is still a bit gritty, so the point I'm trying to make doesn't change, but the screen shot wouldn't show it very well.
Anyway, I love MP3 files for their convenience, and I think that when they're encoded well, they're great for distributing music. I'm certainly not a purist in that respect. But for things like test audio tones, you shouldn't use them.
I'll shut up now. :-)
(no subject)
Date: 2006-06-19 11:11 pm (UTC)(no subject)
Date: 2006-06-19 11:12 pm (UTC)Up through 19Khz are very clear, after it's really faint. But it seemed to change on different levels so I think I was hearing something other than the sample start/sample stop noises. ^_^ I could hear both the primary tone (really, really faintly) and a lower sample harmonic (it was louder to me than the primary tone) at 21K. At 22K, I don't think I heard or felt much of anything. Results for me were pretty similar across the two pages.
(no subject)
Date: 2006-06-19 11:18 pm (UTC)By "change on different levels", do you mean different volume levels?
Human hearing actually has a different set of frequency sensitivity, depending on the volume level you're hearing the sound at. It's flattest at around 60-80db, which is pretty loud but not painfully so.
(no subject)
Date: 2006-06-19 11:28 pm (UTC)(Apple G4 powerbook 12".)
(no subject)
Date: 2006-06-19 11:15 pm (UTC)(no subject)
Date: 2006-06-19 11:23 pm (UTC)(no subject)
Date: 2006-06-19 11:37 pm (UTC)These tests will only give the worst-case limits of computer hardware and ears combined. People may well have hearing ranges beyond what their computers can reproduce, and unless there's someone else in the room who can clearly hear samples inaudible to others, the test would really need to be conducted with professional hearing-test equipment to get accurate results.
(no subject)
Date: 2006-06-19 11:38 pm (UTC)20 Hz, 8000 Hz, and 10000Hz sound normal. 11000-14000 get progressively softer in volume. 9000 is also notably soft, as is 15. 50Hz and 15000+ is inaudible to me.
This is from the flash version. Did anyone else notice volume changes using the flash version?
(no subject)
Date: 2006-06-20 12:09 am (UTC)(no subject)
Date: 2006-06-20 12:11 am (UTC)(no subject)
Date: 2006-06-20 12:19 am (UTC)On the Flash, I can't even hear the 16K, but I can hear the 15K, and on the low end I can only hear the 50, but not the 20 (25?) or 15.
(no subject)
Date: 2006-06-20 12:45 am (UTC)(no subject)
Date: 2006-06-20 12:47 am (UTC)(no subject)
Date: 2006-06-20 11:10 am (UTC)(no subject)
Date: 2006-06-20 08:38 pm (UTC)(no subject)
Date: 2006-06-20 09:33 pm (UTC)I've got sleep apnea myself, but that's a *far* more recent development than my asthma.
(no subject)
Date: 2006-06-20 01:54 am (UTC)I can easily hear all the lower tones though.
(no subject)
Date: 2006-06-20 02:55 am (UTC)(no subject)
Date: 2006-06-20 03:00 am (UTC)On the non-flash page, I can hear sounds up to 20,000 (though 16-19K sounded a bit quieter than 20, which is screeching loud).
On the flash page, I only hear sounds through 15,000; 16-18 are not a sound so much as a feeling.
I got the same results or worse with low-fidelity earbuds.
On the other hand Michael the semi-pro cellist had trouble hearing even the 15Ks. He is currently blaming it on my computer fan noise. I think this just proves that each one of us hears things the other doesn't. ;)
(no subject)
Date: 2006-06-20 03:24 am (UTC)"I can just barely hear the *14,000 Khz* one" !?
Date: 2006-06-20 03:44 am (UTC)All kidding aside I can definitely hear everything between 20 and 20,000 Hz, but nothing outside of those. However, like others 20-20,000 just happens to be a hardware limitation so nothing outside of that is going to be heard by ANYONE listening to my system :). At 37, I'm pretty happy about that - one thing I have always prided myself on is my well-above average hearing
(no subject)
Date: 2006-06-20 11:07 am (UTC)On the other hand, while the high tones start getting harder to hear aroun 21kHz, I *can* hear them all the way to 25kHz.
Then again, I used to be able to hear to 30kHz accord to a test in a science museum exhibit. And I learned the hard way that I can hear the emitters for ultrasonic motion detectors.
I say hard way because if you move too dfast relative to them and at closer range, the doppler shift will take the top of your head off. Ouch.
My high frequency range is, I am told, a common side effect of asthma.
(no subject)
Date: 2006-06-20 12:59 pm (UTC)(no subject)
Date: 2006-06-20 01:47 pm (UTC)(no subject)
Date: 2006-06-20 07:02 pm (UTC)I feel like quoting Lex Luthor's ultra-high freq message to Supes in the 1978 film.