1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Why Is 1kHz So Important As a Measure of Equipment Performance...

Discussion in 'Sound Science' started by TheSonicTruth, Jun 14, 2018.
2 3 4
  1. TheSonicTruth
    When, according to decades of tests and measurements, the average human's hearing is most sensitive around 3kHz(+-0.5kHz)?
  2. pinnahertz
    Specs at 1k go way back into history when every audio device could pass 1kHz without difficulty, but even 3kHz was a problem sometimes. Early audio devices also had trouble with bass. It’s also a frequency where distortion products are audible.

    Today it’s as wrong as any of the traditional spec methods, like single figure distortion, frequency response without stated deviation, s/n without a stated reference, and so on. But they’re all still used anyway.
  3. castleofargh Contributor
    those who care, never stop at measuring 1khz anything. and the consumers would be given all specs in banana units if marketing was totally free to make their own facts. it's already often the case as a lot of typical specs at 1khz are given incomplete and only act as more eye candy with no objective meaning.
  4. 71 dB
    If I measure a device with one frequency only, it's 630 Hz, which is the logarithmic "halfway" of 20 Hz and 20 kHz.
  5. TheSonicTruth
    Only thing is, our hearing sensitivity is already down 15-20dB at that frequency.
  6. 71 dB
    So what? It's a technical measurement that doesn't involve ears. It involves eyes when watching the output signal on computer screen.
  7. gregorio
    Going back probably 80 years or so, a 1kHz signal (1 frame's worth, 2 secs before the FFOA) was routinely used for aligning the sound with the picture in film. Then in the early 40's the VU meter was standardised, calibrated with a rise time and voltage level referenced with a 1kHz signal. It had nothing to do with human hearing, other than that even a short 1kHz tone is easily audible. I've no idea why 1kHz specifically was originally chosen, pinnahertz's explanation seems entirely reasonable, although there may have been some other technical reason at the time.

  8. Steve999
    Wow, that's interesting. 3 khz is around the highest G on a piano--a pretty high frequency note as far as melodic musical instruments go. 1000 hertz is about an octave above middle C--something that's a lot more within my range of practical experience as far as musical instruments goes.

  9. TheSonicTruth
    See for yourself - I didn't make it up:

    Lowest point of curve(in dB SPL) is where we are most sensitive. When playing music at a social function, that is area I tend to reduce on the EQs, just a couple dB between 1 & 4kHz, plus a small notch anywhere from 150-250Hz, where many speakers tend to exhibit tubbiness.
    Last edited: Jun 18, 2018
  10. gregorio

  11. TheSonicTruth
    Because it sounds better, that's why! I experimented with cutting in areas corresponding to the lowest parts of the 226 curve, found something that sounded better. Is that alright with you? And for times when I had to crank the music, people felt it as much as they heard it, without any fatigue.
  12. gregorio
    But why would it sound better?

  13. TheSonicTruth
    Do I need to explain? Look at the graph 3-4 posts back: Our hearing is most sensitive between 2-4kHz. If the sound is left flat, or EQd to flat, our HEARING is not flat. So even on a larger more powerful system peoples' ears will still hear a disproportionate amount of the audio in the range I specified.

    HERE is how we hear - an average of our ears' actual 'frequency response', as if they were a pair of speakers:


    Notice it is approx. the inverse of the ISO 226:2003 contour, which indicates how loud sounds at all frequencies must be to sound equally loud to us.

    Why, Gregorio, do so many typical consumers crank the bass & treble in their car decks, or create smiley EQs, albeit not accurate, with graphics on their home hi fis? Because it SOUNDS BETTER to them. The Equal Loudness Contour and the above hearing response curves prove it!

    I tend to leave my car and home controls flat, but I have Audio Forge's Equalizer app on all my mobile devices, so I can dial in the cut portions of these curves quite accurately when listening back on them.
    Last edited: Jun 18, 2018
  14. gregorio
    Who/What is leaving the sound flat or EQ'ing it to flat?

  15. castleofargh Contributor
    it is true that we're more sensitive in the mid range(you'll notice that it doesn't matter as much at listening levels though). but we have never known anything else. we are literally calibrated on our equal loudness contour curve, it is how we feel flat no matter what that curve is. so I disagree with your idea that it is why some will want to boost bass and or trebles. IMO that specific aspect of our hearing shouldn't be compensated by gears. or it would need to be compensated at all time to justify your argument and let our brain learn that the new curve is now our perceived flat.
2 3 4

Share This Page