Search This Blog

Wednesday, March 28, 2012

personal concerns regarding other headphone-measuring websites

Please, I hope no one gets a chip on their shoulder- Yes, my DIY measurement setups are not as accurate! It's OK, it's just a DIY. LOL


As introduced previously, a correct diffuse-field compensation target is the key to an accurate representation of headphones' psychoacoustic equilibrium(aka the orthotelephonic gain) in reference to an AVERAGE human head-related-transfer-function(HRTF) when measured with a simulator. Of course, as outlined in IEC 60268-7, since the measurement system is just a mannequin, the representation would still be that of an average human, and it can never truly show the response in each listener's ear. IOW, it would be most useful for specifying industrial specifications, and should only represent the objective quality of headphones' performance.

Nonetheless, something must be fundamentally souped up with the frequency response data of Innerfidelity.com(IF) and en.goldenears.net(GE). They use very expensive, international-standard-grade conventional measurement systems, but what is up with all the fuss? It's all in their non-conventional compensation targets for headphone measurement data.




The case of Innerfidelity



Apparently, IF utilizes HEAD acoustics' patented-Independent-of-Direction(ID) equalization, which at first glance seems to be a technical compromise between FF compensation and DF compensation.

After lengthy discussion with the applications engineers at Head Acoustics, the Independent of Direction HRTF was picked as the most applicable for my purposes. http://www.innerfidelity.com/content/headphone-measurement-proceedures-frequency-response

According to the manufacturer's data sheet, ID, of which its feasibility lies in a mathematical equation, removes 'direction-independent resonances' of a concha (~5 kHz) & an ear canal (~3 kHz) only. Simply put, what this ID equalization does is, theoretically it turns an eardrum(DRP) measurement into a 6cc coupler measurement. However, this effectively leaves out a good amount of resonance in the mid-frequency range; Thus, recorded material would contain the directional-dependent residual resonance.
Ear canal gain + concha gain = Independent-of-Direction gain

Yet, as Theile and Ueda had proven previously, the characteristic of localization in headphones are diffuse-field referenced by its nature(circum-aural).

Leaving out something that must be compensated just because it's too ambiguous to deal with, does not guarantee accuracy of the result. Rather than yielding an accurate representation of the average human HRTF, the result of Independent-of-Direction equalization would be very statistical and of non-human.

...since all headphones have the same HRTF applied, measurements are legitimate for comparative purposes. http://www.innerfidelity.com/content/headphone-measurement-proceedures-frequency-response

In his AES presentation in 1990, Griesinger noted that "an equalization obtained from an eardrum measurement is clearly superior both in timbre and in localization", compared to that of an blocked ear canal.

Anyway, Hammershøi and Møller approached this equalization technique using blocked ear canal data in a similar fashion with practicality (less analytical), and came up with much more accurate method, which made into the ISO standard (which I adopted to my measurements).



The case of Goldenears




Goldenears have somewhat different issue. GE's data complies with all of the relevant international standards, yet utilizes their own diffuse-field compensation technique involving high frequency reduction in the upper frequency range, claiming the attenuation is supposed to be 'the X-curve' of the small room environment, aka the curve X of B(final) chain from ISO 2969, and it correctly represents the orthotelephonic gain of headphone listening due to recording studios' room characteristic.

...As master recordings are created in a studio with this characteristic, further calibration must be done in order to make the response more accurately resemble what we perceive. http://en.goldenears.net/388

Consequently, the frequency range above 2 kHz gets -3dB / octave. Fair enough. In 2007, Mead Killion, the founder of Etymotic Research, posted this article on his now-defunct blog:




Yes, this diffuse-field response shows the high frequency reduction in the 2 kHz ~ 20 kHz, and quite matches the response of one of their products, ER-4S, and also corresponds to the curve-X.

Don Wilson, the inventor of ER-4 series, once said:
Several engineers and myself did a study in a radio station recording studio in chicago for an AES meeting when the ER-4's first came out. We compared several different recordings through a pair of room equalized loudspeakers. The participants listened to the same recording through the loudspeakers and the ER-4 one ear at a time. We played the recordings through the ER-4B with several different equalizations. The participants voted as to which response sounded the closest to the loudspeakers. After we graphed the answers we discovered that the ER-4S had the closest response to the speakers .http://gilmore2.chem.northwestern.edu/ubb/showpost.php?fnum=1&tid=1718&pid=13330&fpage=3 

This high frequency attenuation definitely makes headphones sound more neutral, so most of headphone manufacturers utilize a gradual high frequency roll-off in their reference products, although its degree may vary slightly among each other. So far so good! It's a reasonable compromise in order to reproduce musical materials, which are meant to be played back in a free field condition, in a random-incidental condition. Then what's the problem?

Let us not forget: As noted in IEC 60268-7, a mannequin measurement shall be only considered beneficial for an industrial usage; a subjective tonal quality shall not be of a concern due to much deviation between simulator measurements & subjectively assessed results. IOW, what a simulator measurement must show is the orthotelephonic gain to the AVERAGE human HRTF(that of a mannequin), so that it can only show the objective data beneficial to the appropriate usage, not the subjectively comforting tonality. In order to make GE's claim true, instead of using a simulator, they must utilize a subjective assessment outlined in IEC 60268-7.




References

E.A.G. Shaw and M.M. Vaillancourt, "Transformation of Sound-Pressure Level from the Free Field to the Eardrum Presented in Numerical Form," Journal of the Acoustical Society of America, Vol.78 No.3 (September 1985).

D. Hammershøi and H. Møller, "Determination of Noise Emission from Sound Sources Close to the Ears," Acta Acustica, Vol.94 No.1 (January 2008).

D. Griesinger, “Binaural techniques for music reproduction,” in Proc. Audio. Engineering Society 8th Int. Conf., Washington, DC, (May 1990).

G. Theile, "On the Standardization of the Frequency Response of High-Quality Studio Headphones," Journal of the Audio Engineering Society, Vol.34 No.12 (December 1986).

K. Ueda, and T. Hirahara, "Frequency response of headphones measured in free field and diffuse field by loudness comparison," Journal of the Acoustical Society of Japan, Vol.12 No.3 (May 1991).

IEC 60268-7 – Sound System Equipment. Part 7: Headphones, International Electrotechnical Commission, Geneva, Switzerland.

ISO 2969-1987 – Cinematography -- B-chain electro-acoustic response of motion-picture control rooms and indoor theatres -- Specifications and measurements, International Organization for Standardization, Geneva, Switzerland.

1 comment: