[plonter]

Hi all. just interested in the next question:
why are headphones made with a frequency response range that's above and beyond the human hearing capabilities?

is there a difference between a 20-20 can and a 5-45 can (not talking about sound signatures diferences,capabilities only)?
thx for your kind answers!

 

[Rempert]

I think you should not pay too much attention to some of these ridiculous frequency response specs with headphones.

It gets tricky because different companies use different levels of tolerance to determine their frequency response specs. Sometimes this tolerance is listed, but often it is not. For example, my nearfield monitors are listed as 49 hz to 20 khz, +/-1.5 db. I have also some cheap old Logitech Z640 speakers, that were listed as 35 hz to 20 khz. Now there is no tolerance listed for those Z640, but I'd bet it is more like +/-10 db. Not a fair comparison, by any stretch.

With headphones though, it gets even trickier, because the ideal frequency response isn't necessarily flat. There has to be compensation for the filtering your head and outer ear normally do to sounds which won't happen when the sound is pumping directly into your ear from headphones.

 

[apatN]

They don't really matter, no.
I did hear a story though that even though you can't actually hear it, it is better. Oh yes, head-fi style.

 

[plonter]

so the denon's 5hz-45khz is not true? they are not really produce those frequencys?

 

[shubhangam]

from what I know,

larger ranges can be better in the sense that <20 hz can be 'felt' so it gives some extra twang to the music.

>20 khz is for something that the distortion shifts towards the right of the curve to make the sound better or something. don't quote me on that though.

 

[apatN]

Quote:

Originally Posted by plonter /img/forum/go_quote.gif so the denon's 5hz-45khz is not true? they are not really produce those frequencys?

They probably are but you are not able to actually hear them.

 

[ClieOS]

Rempert said it well, and I have a few points of my own to add. Here is a portion of my IEM guide regarding headphone frequency. It is meant for IEM user in portable setting, but same principles are true to all headphones.

Quote:

Originally Posted by ClieOS /img/forum/go_quote.gif In regard to IEMs' frequency response Wonder why so many people tell you that IEM are very reveling and full of detail, yet most IEM don't seem to go pass 16kHz when you look at their frequency response? Here is the reason: Although human hearing range are listed at 20 Hz to 20000 Hz, most people can't actually (or at least have difficulty) hear sound above 16kHz, especially once you pass your twenty. Don't worry about losing treble/detail over 16kHz, as human tend to interpret sound b/w 5kHz to 16kHz as 'high' and most detail are actually on the lower region. A headphone that is capable of producing sound over 16kHz, 20kHz or even 30kHz doesn't help if you can't hear it, unless you are planning to share your 'phone with your dog. So why headphone manufacturers like to highlight the frequency response of their 'phone when they know you can't hear it? Cause general consumer don't know the fact and often believe 'more is better' with out realizing it is just a marketing trick. Some of you might have read from else where that human, although can't hear sound beyond 20KHz, can still feel the present of ultrasound (and the harmonic distortion caused by ultrasound) and it does have a positive effect on SQ - This is true. However, giving that almost all CD, lossy and lossless music produced these days have a cap on 20KHz in recording / encoding, plus most sources only equip with a 16bits DAC that ain't capable of rendering sound beyond 20KHz efficiently, we can safely conclude that most music we listen these days doesn't come with any meaningful sound beyond 20KHz that can be reproduced by your headphone or IEM. If you want to listen to a recording that does include sound beyond 20KHz, you'll have to go back to analog record (LP) or get SACD or DVD-A, but mind you, you'll still need equipments that capable of playing those sound back. For a better understanding of which IEM is more detail, the best way is to read its review (or better yet, an audition). Also, Headroom has done a great job in measuring headphone freq. response, so you might want to have a look. Note that, although freq. resp. does tell us a bit about how headphone (including IEM) might sound, it doesn't tell us how good it will sound. When it comes to picking up a pair of headphone, your ears are much more trust worthy than your eyes. If you are interesting in learning more about frequency response, here is an article that might provide you with more information.

 

[plonter]

an interesting post clieos..so is it all a lie? i mean..how can they do that?
all the big companies like senn agk...it doesnt make sense to me.
in fact..i really tought that headphones actually can produce those frequencies(in optimal situations and gear ofcourse)and we just can't hear them .
and one more thing..if it is a marketing trick,than should they need to know better? i am talking of the headphones companies...i mean,they are obviously know that a long time audiophiles(and even not a long time) are aware of this "lie" (and yeah! it IS a lie if it is not true!)

and still keep going with that?? that's what i don't get!
and ofcourse there are some that actually measuring the headphones!
thx!

 

[Shoewreck]

I'd assume that 5-35K is generally better than 20-20K. The tolerances are usually big (6-10dB). If 10dB falloff happens only far outside of your hearing range you may hope for quite even response within it.
Just don't pay much attention to comparing different manufacturers top headphones ranges. They usually just add 500Hz to HF limit of a competitive product and draw it on their boxes. But if you compare ranges within HP's of a single manufacturer a wider range would honestly mean more extension.

 

[krmathis]

Its all about frequency fall-off (tolerance), I would say.
The tolerance are often 6-10db, and a 'phone with a frequency range of 20Hz-20KHz would be -10db within the hearing range (ex. 19KHz and/or 25Hz. While a 'phone with a frequency range of 5Hz-35KHz might have the -10db drop outside our hearing range (ex. 30KHz and/or 10Hz)..

 

[Rempert]

It's not a lie. There are legit measurements involved. It's just that there is no real, useful standard of how much the frequency response can vary before we should start saying this headphone can't reproduce it. And also that 20hz-20khz is more than enough, but if you can claim better in a competitive market you've got to do it.

I might draw a comparison to the "viewing angle" spec on flat screen monitors. You get the cheap monitor with a viewing angle of 170 degrees, and the super expensive one with a viewing angle of 178 degrees. There is some truth to these specs, but they are still totally absurd! Who is going to sit 89 degrees from perpendicular to a screen, and what could they possible hope to see at that angle?! And why does a screen with 170 degree viewing angle start to have the colors get darker if I move 40 degrees or so? It would be better if they had to use a much tighter, industry standard tolerance for how much the image can get lighter or darker, and give us some actually useful information about where the viewing angle is really accurate. But there is no such standard, and manufacturers understandably must use the highest number they can claim a legitimate measurement for in order to be competitive. Frequency response specs are the same kind of game.

 

[gregorio]

IEM Guide - Inaccuracies in the post quoted above:

"Some of you might have read from else where that human, although can't hear sound beyond 20KHz, can still feel the present of ultrasound (and the harmonic distortion caused by ultrasound) and it does have a positive effect on SQ - This is true."

No it's not true. There was some research carried out by a Japanese scientist many years ago which proved this. This research has been quoted many times in numerous publications and on the net. However, it was later shown that the methodolgy for the experiment was flawed and although the experiement has been recreated a number of times, no one has ever managed to obtain a significant result.

"However, giving that almost all CD, lossy and lossless music produced these days have a cap on 20KHz in recording / encoding".

All CD format audio requires a brickwall filter to remove all frequencies above 22,050Hz. It's a condition of digital audio that the audio frequency encoded can be no higher than half the sample rate (the Nyquist Point). So CD has a fixed sample rate of 44,100 samples per second which therefore equals an absolute maximum audio freq of 22,050Hz. Anything higher than this must be completely eradicated!

"plus most sources only equip with a 16bits DAC that ain't capable of rendering sound beyond 20KHz efficiently".

Not necessarily true. CD source material as mentioned above can theoretically go as high as 22.050 but other 16bit sources can encode frequencies more than double the CD limit. The maximum audio frequency which can be encoded is not related to bit depth but to sample rate. Bit depth encodes amplitude, sample rate encodes audio frequency.

From about 60Hz down, the ears become increasingly inefficient and the body's ability to 'feel' sound gradually takes over. However, for the body to feel these low frequencies it requires that these frequencies hit the body, not just the ears. So cans which proclaim 5Hz freq response are, IMO, pretty much pointless, except from the fact that hopefully if they can reproduce 5Hz maybe they are reasonably accurate when it comes to frequencies which are useful.

G

 

[plonter]

ok..thanks for the replies guys. so the greater the frequency range is (usually) than it's benefits the frequency that we can actually hear.
am i right..?
and one more question if i might...how do they measure this..? do they have certain equipment that give them the results? and do you think that the results are similar to those they publish?..!

 

[nick_charles]

Quote:

Originally Posted by plonter /img/forum/go_quote.gif and one more question if i might...how do they measure this..?(1) do they have certain equipment that give them the results? and (2) do you think that the results are similar to those they publish?..!

(1) If they do it seriously they use some kind of fake head with microphones in it to mimic a human listener .

(2) Who knows ?

 

[gregorio]

Plonter "ok..thanks for the replies guys. so the greater the frequency range is (usually) than it's benefits the frequency that we can actually hear.
am i right..? "

No, there is no point in cans that can go up to 35kHz, humans can't hear that high and CDs contain absolutely no freqs above 22kHz.

G