headphone response curve question

Archived from groups: rec.audio.high-end (More info?)

Mark Hifi wrote:

> I am looking for some really flat headphones, but not for $500.
>
> I have taken a signal generator and done some frequency sweeps on various
> headphones and found considerable response peaks, just by listening. Really
> cheap headphones like walkman actually have no output at certain
> frequencies.
>
> One might claim is was my hearing except for the fact that every type of
> headphone is different. Some have peaks at the same frequency that some
> have low spots.
>
> The output of the generator remains flat on a scope.
>
> Anybody out there have any experience with headphone response?
>
> I haven't taken any apart, but it seems like the headphone element is more
> of a vibrator than a speaker, lots of mechanical resonance.
>
> Headphones with a lot of response peaks are uncomfortable to listen to and
> cause listener fatigue very quickly. Regards, Mark

One of the most educational articles on headphone/earphones:

http://www.linkwitzlab.com/reference_earphones.htm

Archived from groups: rec.audio.high-end (More info?)

chung <chunglau@covad.net> wrote:
> Mark Hifi wrote:

> > I am looking for some really flat headphones, but not for $500.
> >
> > I have taken a signal generator and done some frequency sweeps on various
> > headphones and found considerable response peaks, just by listening. Really
> > cheap headphones like walkman actually have no output at certain
> > frequencies.
> >
> > One might claim is was my hearing except for the fact that every type of
> > headphone is different. Some have peaks at the same frequency that some
> > have low spots.
> >
> > The output of the generator remains flat on a scope.
> >
> > Anybody out there have any experience with headphone response?
> >
> > I haven't taken any apart, but it seems like the headphone element is more
> > of a vibrator than a speaker, lots of mechanical resonance.
> >
> > Headphones with a lot of response peaks are uncomfortable to listen to and
> > cause listener fatigue very quickly. Regards, Mark

> One of the most educational articles on headphone/earphones:

> http://www.linkwitzlab.com/reference_earphones.htm

It wasn't clear to me which of these 'phones really had a *measurably*
flat frequency response, and which only 'sounded' flat to Mr. Linkwitz,
either out of box or after EQ. It appears to me that the first two were EQ'd to
sound flat to him.

--

-S.

"They've got God on their side. All we've got is science and reason."
-- Dawn Hulsey, Talent Director

Archived from groups: rec.audio.high-end (More info?)

"Mark Hifi" <hifiguy@sonic.net> wrote in message news:<c9ocbh0bbn@news1.newsguy.com>...
> I am looking for some really flat headphones, but not for $500.
>
> I have taken a signal generator and done some frequency sweeps on various
> headphones and found considerable response peaks, just by listening. Really
> cheap headphones like walkman actually have no output at certain
> frequencies.

The Sennheiser HD-580 is one of the best headphones made in terms of
flat frequency response and excellent phase and impulse response.
They're circumaural. I've owned these for over 5 years and still
listen to them every day. The HD-600 and HD-650 are sonically very
similar to the HD-580, but much more expensive.

The Etymotic ER4S are super clean and linear like the HD-580s, but in
a "sealed in your ear" setup which provides about 20 dB of exterior
noise reduction.

Here is a site from Headroom which has specs on these and many, many
other headphones. They do their own independent testing. And they sell
all kinds of headphone stuff too. I bought one of their headphone amps
and have found to be a superb piece of equipment: excellent sonics,
long lasting build quality and great factory support.

http://www.headphone.com/layout.php?topicID=10

Archived from groups: rec.audio.high-end (More info?)

"Mark Hifi" <hifiguy@sonic.net> wrote in message news:<c9ocbh0bbn@news1.newsguy.com>...
> I am looking for some really flat headphones, but not for $500.
>
> I have taken a signal generator and done some frequency sweeps on various
> headphones and found considerable response peaks, just by listening. Really
> cheap headphones like walkman actually have no output at certain
> frequencies.
>
> One might claim is was my hearing except for the fact that every type of
> headphone is different. Some have peaks at the same frequency that some
> have low spots.
>
> The output of the generator remains flat on a scope.
>
> Anybody out there have any experience with headphone response?
>
> I haven't taken any apart, but it seems like the headphone element is more
> of a vibrator than a speaker, lots of mechanical resonance.
>
> Headphones with a lot of response peaks are uncomfortable to listen to and
> cause listener fatigue very quickly. Regards, Mark

Headphone measurements are not easy to perform, and the response curve
is and should not measure absolutely flat to be subjectively regarded
as neutral. I was participating in a listening test with >30
headphones. They were measured with an artificial head model and all
headphones differed quite extensively. Measurements and listening
tests were published in "Musik & Ljudteknik nr 4, 2003, in Swedish. I
liked the sound of Sennheiser HD600 and AKG K271 studio, an opinion
which was shared by others. Although many of the STAX headphones
tested were highly regarded among most listeners, I found them to have
unnatural treble.

T

Archived from groups: rec.audio.high-end (More info?)

On 6/4/04 7:16 PM, in article So7wc.2950$HG.2729@attbi_s53, "Thomas A"
<Thomas_Akerlund@hotmail.com> wrote:

> "Mark Hifi" <hifiguy@sonic.net> wrote in message
> news:<c9ocbh0bbn@news1.newsguy.com>...
>> I am looking for some really flat headphones, but not for $500.
>>
>> I have taken a signal generator and done some frequency sweeps on various
>> headphones and found considerable response peaks, just by listening. Really
>> cheap headphones like walkman actually have no output at certain
>> frequencies.
>>
>> One might claim is was my hearing except for the fact that every type of
>> headphone is different. Some have peaks at the same frequency that some
>> have low spots.
>>
>> The output of the generator remains flat on a scope.
>>
>> Anybody out there have any experience with headphone response?
>>
>> I haven't taken any apart, but it seems like the headphone element is more
>> of a vibrator than a speaker, lots of mechanical resonance.
>>
>> Headphones with a lot of response peaks are uncomfortable to listen to and
>> cause listener fatigue very quickly. Regards, Mark
>
> Headphone measurements are not easy to perform, and the response curve
> is and should not measure absolutely flat to be subjectively regarded
> as neutral. I was participating in a listening test with >30
> headphones. They were measured with an artificial head model and all
> headphones differed quite extensively. Measurements and listening
> tests were published in "Musik & Ljudteknik nr 4, 2003, in Swedish. I
> liked the sound of Sennheiser HD600 and AKG K271 studio, an opinion
> which was shared by others. Although many of the STAX headphones
> tested were highly regarded among most listeners, I found them to have
> unnatural treble.

Most good headphones tend to have a bit of boost around 200Hz - and roll off
above 10kHz. It appears to have the most "euphonic" sound.

I have some studio phones that measure dead flat - and while the sound is
decent - it sounds a bit "dry" - probably the best way t put it, until I
EQ'ed them to have a bit of boost around 200Hz and rolled off the upper
response about 3-4dB by 20kHz. Go figure. I am sure there is a
psychoacoustic explanation or something.

Archived from groups: rec.audio.high-end (More info?)

Steven Sullivan wrote:
> chung <chunglau@covad.net> wrote:
>> Mark Hifi wrote:
>
>> > I am looking for some really flat headphones, but not for $500.
>> >
>> > I have taken a signal generator and done some frequency sweeps on various
>> > headphones and found considerable response peaks, just by listening. Really
>> > cheap headphones like walkman actually have no output at certain
>> > frequencies.
>> >
>> > One might claim is was my hearing except for the fact that every type of
>> > headphone is different. Some have peaks at the same frequency that some
>> > have low spots.
>> >
>> > The output of the generator remains flat on a scope.
>> >
>> > Anybody out there have any experience with headphone response?
>> >
>> > I haven't taken any apart, but it seems like the headphone element is more
>> > of a vibrator than a speaker, lots of mechanical resonance.
>> >
>> > Headphones with a lot of response peaks are uncomfortable to listen to and
>> > cause listener fatigue very quickly. Regards, Mark
>
>> One of the most educational articles on headphone/earphones:
>
>> http://www.linkwitzlab.com/reference_earphones.htm
>
> It wasn't clear to me which of these 'phones really had a *measurably*
> flat frequency response, and which only 'sounded' flat to Mr. Linkwitz,
> either out of box or after EQ. It appears to me that the first two were EQ'd to
> sound flat to him.
>

There is difficulty measuring the response of the earphones to judge
actual performance, since "due to the acoustic impedance mismatch
between transducer, ear canal and ear drum cause a half wavelength
resonance in the canal" according to the article. In other words, your
own ears are part of the system, and add a frequency response distortion
that is different than that from listening to loudspeakers.

Siegfried Linkwitz is saying that there is a resonance due to the above
effect, and there is also intentional equalization in some
earphones/headphones. He found the Shure E2C's to be the most accurate
perceptibly, followed by the Sony and the Etymotics.

Archived from groups: rec.audio.high-end (More info?)

Bromo <bromo@ix.netcom.com> wrote in message news:<qb9wc.4595$4S5.1622@attbi_s52>...
> On 6/4/04 7:16 PM, in article So7wc.2950$HG.2729@attbi_s53, "Thomas A"
> <Thomas_Akerlund@hotmail.com> wrote:
>
> > "Mark Hifi" <hifiguy@sonic.net> wrote in message
> > news:<c9ocbh0bbn@news1.newsguy.com>...
> >> I am looking for some really flat headphones, but not for $500.
> >>
> >> I have taken a signal generator and done some frequency sweeps on various
> >> headphones and found considerable response peaks, just by listening. Really
> >> cheap headphones like walkman actually have no output at certain
> >> frequencies.
> >>
> >> One might claim is was my hearing except for the fact that every type of
> >> headphone is different. Some have peaks at the same frequency that some
> >> have low spots.
> >>
> >> The output of the generator remains flat on a scope.
> >>
> >> Anybody out there have any experience with headphone response?
> >>
> >> I haven't taken any apart, but it seems like the headphone element is more
> >> of a vibrator than a speaker, lots of mechanical resonance.
> >>
> >> Headphones with a lot of response peaks are uncomfortable to listen to and
> >> cause listener fatigue very quickly. Regards, Mark
> >
> > Headphone measurements are not easy to perform, and the response curve
> > is and should not measure absolutely flat to be subjectively regarded
> > as neutral. I was participating in a listening test with >30
> > headphones. They were measured with an artificial head model and all
> > headphones differed quite extensively. Measurements and listening
> > tests were published in "Musik & Ljudteknik nr 4, 2003, in Swedish. I
> > liked the sound of Sennheiser HD600 and AKG K271 studio, an opinion
> > which was shared by others. Although many of the STAX headphones
> > tested were highly regarded among most listeners, I found them to have
> > unnatural treble.
>
> Most good headphones tend to have a bit of boost around 200Hz - and roll off
> above 10kHz. It appears to have the most "euphonic" sound.
>
> I have some studio phones that measure dead flat - and while the sound is
> decent - it sounds a bit "dry" - probably the best way t put it, until I
> EQ'ed them to have a bit of boost around 200Hz and rolled off the upper
> response about 3-4dB by 20kHz. Go figure. I am sure there is a
> psychoacoustic explanation or something.

Well, the response curve if one compares to normal stereo listening
via loudspeakers are different. With speaker listening, the impact of
the body, head and outer ear is affecting the response curve. So, a
freqeuncy response curve from a headphone must take into account the
effects of these to be judged as neutral. I have not seen any
headphones that measure flat using an artificial head model. The
headphones that were tested and measured in MoLT were:

AKG K240 studio, AKG K271 studio, AKG, K501, BD DT-100, BD, DT150, BD
DT-770, BD, DT-880, BD DT-990 pro, Bose Triport, Creative HQ-2000,
Etymotic Research ER-6, Grado SR325, Koss Porta-Pro, Koss UR-40, Koss
Pro 4AA titanium, Koss ESP 950, Phillips HP890, Sennheiser PX100,
Sennheiseer, 250 Linear II, Sennhesier HD590 fusion, Sennheiser HD590
Prestige, Sennheiser HD600 Avantgarde, Sony MDR-CD850, Sony MDR-CD780,
Sony MDR-V700, Sony MDR-CD2000, Stax SR-001 MkII +3-4 STAX systems.

The flattest one, as measured, was the Senn. HD600. The "best ones"
according to several listeners independently were AKG K271 studio, BD
DT-880, Koss ESP 950, Sennh. HD600 Avantgarde. The worst ones were BD
DT-100 and Et. Res. ER-6.

T