Head-Fi.org › Forums › Equipment Forums › Sound Science › A couple of questions about frequency response measurements.
New Posts  All Forums:Forum Nav:

A couple of questions about frequency response measurements. - Page 2

post #16 of 27
Thread Starter 

Thanks. It's a shame that these measurements are so rare, they seem pretty important.

post #17 of 27
Quote:
Originally Posted by chewy4 View Post

Thanks for the examples.

 

Do you know of any websites that actually do these IMD measurements for headphones?

 

Not the measurements, but the test tones :

 

http://www.audiocheck.net/audiofrequencysignalgenerator_dual.php (online generator)

http://www.audiocheck.net/testtones_imd.php (subjective IMD test)

post #18 of 27

Boston Audio Society tests - a good while ago - so none of the recent $k can's existed then

 

http://www.bostonaudiosociety.org/pdf/bass/BASS-25-4.pdf  page 18+

post #19 of 27

I have a question which isn't entirely topical, but I was reluctant to start a new thread for it and this seems to be the best place for it.

 

How relevant is a frequency response graph to listening experience?  Or rather, how do I pair frequency response with harmonic distortion?

 

Take for example a comparison between two relatively inexpensive (or even cheap, depending on who you're asking) headphones. The Sennheiser HD 25 II's, and the AKG K81DJ:

 

 

 

Despite all this the K81DJ's bass kind of over-runs the rest of the music, whereas in the HD25's, the bass is more controlled and the highs better defined and instrument separation is leagues better. Yes yes, the AKG's are by far the cheaper headphones but this seems to negate the graphs, which shows a lower dB bass response. Shouldn't this, in theory, translate into a weaker bass?

post #20 of 27

Frequency response differences are the most audible of all, well above distortion unless THD or IMD gets very high.  

 

We hear response differences based on the total area affected.  Very small changes, like little narrow peaks, have to be fairly significant in magnitude to be audible, where changes that affect large, broad areas of the curve are audible with relatively small changes in gain.

 

Distortion audibility is harder to grasp.  I depends on the magnitude of the harmonics generated, how many there are and how far removed from the fundamental they are, and very significantly, the type of signal being used. Distortion is most audible using pure tones, but music isn't made up of pure tones, and many musical signals tend to mask distortion.  Audibility of distortion using music is therefore not a fixed number, but generally much higher than that of sine waves.  People can hear simple THD comprised mostly of third harmonic at 1% using a sine wave, but that's pretty hard to hear with music, and sometimes THD as high as 3% is barely audible.  But, if the harmonic content is extremely rich with products out to the 10th harmonic, that's much harder for complex music to mask.  IMD can be more audible because the products may not be harmonically related.

 

In your examples, though, there's probably more going on, like a seal issue.  Can't really say from way over here. 

post #21 of 27
Quote:
Originally Posted by MrLazyAnt View Post

I have a question which isn't entirely topical, but I was reluctant to start a new thread for it and this seems to be the best place for it.

 

How relevant is a frequency response graph to listening experience?  Or rather, how do I pair frequency response with harmonic distortion?

 

Take for example a comparison between two relatively inexpensive (or even cheap, depending on who you're asking) headphones. The Sennheiser HD 25 II's, and the AKG K81DJ:

 

 

 

Despite all this the K81DJ's bass kind of over-runs the rest of the music, whereas in the HD25's, the bass is more controlled and the highs better defined and instrument separation is leagues better. Yes yes, the AKG's are by far the cheaper headphones but this seems to negate the graphs, which shows a lower dB bass response. Shouldn't this, in theory, translate into a weaker bass?

try using innerfidelity graphs or goldenears, they have a little bit better graphs imo

 

http://www.innerfidelity.com/images/AKGK81DJ.pdf

http://www.innerfidelity.com/images/SennheiserHD251IIB2012.pdf

 

the senns seems to be more neutral, while the akg exhibits the usual dip in the mids typical of basshead cans

 

frequency response is quite relevant, you can get a semi feel of what to expect, this is partly how I based my purchase decisions. I purchased the Shure 1840 for its straight line freq response from 200 Hz to 4kHz to ensure the most neutral response in terms of speech frequencies but it steeply rolls off in the bass. I did a comparison of Shure 1840 and HE-500 which has a more straight line from 1kHz and down but has roughly a 5 dB difference in 2-4 kHz. The difference is quite noticeable imo but not a deal breaker. HE-500 has a better bass response but not overpowering but doesn't has a good of mids imo.

post #22 of 27
Quote:
Originally Posted by KamijoIsMyHero View Post

the senns seems to be more neutral, while the akg exhibits the usual dip in the mids typical of basshead cans

 

My query, the issue at hand for me, is that the AKG's exhibit a much lower bass volume on the graph, but in practice, their bass is very dominant, whereas the HD25's while more neutral, show a higher dB bass response, yet the bass does not dominate the sound like in the AKG's. This defies the graphic readings. I am seeking an explanation for this phenomenon if anybody should venture to offer one

post #23 of 27
You don't really need to look at the numbers in a frequency response, just the general shape

Also curious what is lower bass for you? (Hz)
post #24 of 27

Quote:

Originally Posted by MrLazyAnt View Post

My query, the issue at hand for me, is that the AKG's exhibit a much lower bass volume on the graph, but in practice, their bass is very dominant, whereas the HD25's while more neutral, show a higher dB bass response, yet the bass does not dominate the sound like in the AKG's. This defies the graphic readings. I am seeking an explanation for this phenomenon if anybody should venture to offer one

 

The AKGs seem to be quite sensitive to sealing, this can be seen on the "raw" graphs which show the frequency response with several different positions of the headphone on the dummy head. The averaged graph may therefore underestimate their bass response because of this. But it also means that the bass will be different for each person because of different head sizes and shapes, wearing glasses or not, etc.

post #25 of 27

e

Quote:
Originally Posted by stv014 View Post

 

The AKGs seem to be quite sensitive to sealing, this can be seen on the "raw" graphs which show the frequency response with several different positions of the headphone on the dummy head. The averaged graph may therefore underestimate their bass response because of this. But it also means that the bass will be different for each person because of different head sizes and shapes, wearing glasses or not, etc.

I'm not sure I follow my good sir (or ma'am), shouldn't the measurements take the sealing into account? Surely the driver membranes are not tested separately from the ensemble of the whole headphone?

post #26 of 27
Quote:
Originally Posted by MrLazyAnt View Post

I'm not sure I follow my good sir (or ma'am), shouldn't the measurements take the sealing into account? Surely the driver membranes are not tested separately from the ensemble of the whole headphone?

 

Yes, the measurements are of the headphone, so the seal has an affect on the measured response.

 

At InnerFidelity, there are FR results for multiple measurements taken at different positions.  The point is that if the FR is significantly different at different positions, then the positioning of the headphones has a strong effect on the response you get.  This is the case for the K550.

 

 

 

The colored lines at top are averages.  If the actual response on your head is most similar to the raw measurement at a certain position with the highest bass (or perhaps you get an even better seal than the dummy head in any of the tested positions), then you'd get stronger bass than what's implied by looking at the average there.

post #27 of 27

I know this is an old thread, but i couldn't help myself

 

https://www.youtube.com/watch?v=BYTlN6wjcvQ

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › A couple of questions about frequency response measurements.