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

How do you tell when a headphone is producing a neutral sound?

Discussion in 'Sound Science' started by sniperwill0, Oct 23, 2012.
2 3
  1. sniperwill0
    I've heard that one should strive to achieve a neutral sound with their cans. However, what exactly does a neutral sound sound like? I know everyone's perception of sound is different, but can anyone give me a general description? Thanks
  2. autumnholy
    To me, no single frequency overpowers or loses behind the other means neutral. Every frequency sounds equally loud to my ears.
    I'd also love to learn what others think about Neutrality.
  3. xnor
    The easiest way to experience neutral sound is with calibrated speakers.
    Every frequency sounding equally loud is not neutral. I guess you meant it differently because equal loudness of pure tones means very boosted bass and treble.
  4. bigshot
    The ballpark of neutral presentation is easy to tell with well recorded acoustic music in a natural ambience. Chamber music and orchestral music show up imbalances.
  5. autumnholy
    Still, how about headphones? xnor what do you think about neutral in headphones? I still think this concept of neutrality is too subjective for me to grasp.
  6. arcticears
    Thanks for bringing that up Xnor, many people seem to mistake a flat frequency response with "every frequency sounding equally loud" which are not the same thing.
  7. chewy4
    Would ensuring that a sound level meter stays at the same db level throughout a frequency sweep be a good way to get a balanced setup?
    Never used one before so I don't know how responsive they are.
  8. xnor
    To measure the frequency response accurately you need a (measurement) microphone. It's not that simple however. You have to couple the mic to the headphone and different coupling methods change the measurement results. Another problem is that the sound of speakers that measure perfectly flat will not arrive flat at the ear drum or even at the ear canal entrance.
  9. streetdragon
    well the ear canal entrance will deflect the sound the same way for loud speakers and headphone, its only iem's that have to compensate for that since they bypass the ear canal entrance
  10. bigshot
    It's not a simple absolute. You get it close, then you make small refining adjustments over time.
  11. xnor
    Sure, but measurement microphones are typically omnidirectional / don't have a pinna, head or torso. So the frequency response of a flat measuring speaker looks different for every person at the ear canal entrance.
  12. yepimonfire
    One good way of doing it is to listen to voice, or orchestra if you're familiar with how they sound live. As others stated, no frequency is accentuated or recessed. I don't think this is a very accurate way of telling though, as brains will "learn" a specific headphones sound as accurate, then when you switch over to something that actually is accurate, it will sound very "off" until your mind "re-learns" how things should sound. Another way of telling of course, is by looking at frequency response graphs (provided you can find them for your particular headphones). Keep in mind however you will not see a straight line from 20hz-20khz because headphones have to compensate for your ears natural equalization of frequencies heard at a distance. Here are a few example of headphones that are reasonably accurate.
    None of these headphones are 100% completely accurate, but they do come pretty damned close. None of them have accurate bass either. Theoretically the line should remain flat at 0dB from 10hz on to around 1000hz
  13. bigshot
    I've got a few tracks that I use for fine tuning response. One has a stairstep rise in the bass of over an octave. Good for fixing bass bumps. Another has widely different orchestral arrangements that highlight specific frequencies. Another has a vocal that can get buried if the low mids are too high. Others have the limit of sub bass or upper mids... Good for making sure the sound isn't too shrill or too boomy. (I pefer that my walls don't rattle even if that isn't totally accurate.)

    I try to make 1 or 2dB corrections and then live with them a while. Otherwise I can chase down rabbit holes.
  14. ProtegeManiac Contributor
    By one's own ears : no sound should sound overdone and detail should be clear; prefer the bass be detailed (you can hear individual strokes on the bass guitar, fast hits on double bass drum pedals, etc) over plain loud, midrange and treble shouldn't be overly grating on all recordings (don't confuse "raw" vocalization of some singers with this, but it still shouldn't be overdone), etc.
    Simplest technical way to find out : buy a USB mic, set it up between the headphone drivers, then run a 20hz to 20khz sweep with all freqs at the same db level. EQ to as close to flat, then work your way back bit by bit if you don't like the result. Make sure the mic is up to spec - some mics are designed purely for vocals, for example, and can be more sensitive at midrange frequencies.

    Obsessive compulsive technical test : get a ballistic test dummy, cut in half, install the mic between both ears and carve an ear canal of roughly the same size plus a canal to run the mic cable down the torso, mount it back up and run the same test. Just as a room can introduce room modes, a mic squarely in between headphone drivers without (simulated) meat, skin and bone between them putting pressure on the pads and getting in the way of the sound they probably won't sound the same as when you actually listen to it.
  15. tumburu
    I found out the term "neutral" only in non professional audio communities. This is because the usual consumer and even hi end (read expensive) products are very "colored".
    In pro audio the appropriate term for the so-called neutrality is "linear" or "flat". And that's easy to be understood if you look at the frequency response of hi end studio monitors, which have a (sometimes scary) flat line representing their response.
    The thing with all these "options" one can have in building a listening system, each of them coloring the sound in a particular way is that one can be easily lost in the process. The advantage is that you can build a very linear system if you know exactly how to achieve that, but also you can end up into something completely different, the result being quite a departure from the intention of the artist/producer/engineer. Using tube amps, tone controls, colored speakers leads you directly there.
    Now, I'm an engineer and I know what tweaking a recording to reach its best means. Every tweak of -1dB here, +0.5dB there is completely gone in the living room of somebody who will shape the tone by their own (sometimes completely ignorant) will. I'll tell you the truth, I don't like equalizer presets built into any player. They are an insult to the effort people put into producing their music and they bypass the careful processing applied with exquisite hi end equipment during mastering.
    Now, headphones are even more of a problem and that's because the ear perceives very different the sound coming from a source placed in its proximity (the closer the bassier). That could be the reason almost nobody is focused on producing headphones who have a linear-perceived response. In short, all headphones, including high end ones are heavily colored. Throw in a "rock" preset or a 'bbe enhancement" on your player and you're done. 
    The only way to achieve linear-perceived response on your headphones is to use non colored, non flattering convertors (DACs), amplifiers and a (preferably linear phase) hi quality algorithm equalizer. But, to know how to tweak the eq one needs a reference, that reference being a linear pair of speakers in an acoustically controlled environment and that's the hardest thing to have for the usual headphone enthusiast. The second thing is having some serious knowledge about using equalizers which only comes by using them on a daily basis. Otherwise you end up with something that will resemble the "classical" or "dance" presets on your player.
    But there are things that can help, one can find lots of graphs showing the response of many headphone models. Measurements are made by different people, using different equipments and can be smoothed etc, but at a closer look some data will correspond (like in "a narrow peak at 6kHz"). Manually running a sine wave sweep while carefully listening to the changes in the perceived amplitude from the headphones will reveal if the data is correct and help immensely at correcting the non-linearities.
    I believe this is the way to do it and also getting familiar with equalization and understanding the bandwidth that needs to be set for a specific problem, and that's why a fully parametric eq is needed. If done right equalization can transform copper into gold, software like Dirac or Accudio (though their "curves" can be a subject of debate) can prove that.
2 3

Share This Page