Interesting headphone website for measurists

[Fatmarley]

Comparing all the Sennheiser headphones, and going by measurements alone (I've only heard and owned the HD650), I'd take the HD650 (lower midrange hump and all). It looks like the bass of the HD600 rolls off too early for my taste.

diyaudioheaven

Sorry if this has already been posted somewhere. It's the first time I've seen it, and being a measurist, I found it very interesting.

 

[davidsrsb]

I don't know what a flat headphone response looks like, your ear shape and ear canal variations are variables, which make absolute measurements suspect

 

[gustav_errata]

I don't know what a flat headphone response looks like, your ear shape and ear canal variations are variables, which make absolute measurements suspect

Indeed, however they would still be useful for comparative purposes.

 

[Mike Reed]

My headphones achieved a very flat response when I accidentally dropped a heavy box on them some years ago. replaced by HD650s which are lovely all-round cans. Subsequently bought Focal Elears after extensive research and am certainly not disappointed. No culture shock (as Senn 800 series prob. would be); just 650s with more of everything, esp, transparency. As with many people who've made this kind of change, I've still got the 650s; use them too, which says something about these old warhorses.

 

[gustav_errata]

I've still got the 650s; use them too, which says something about these old warhorses.

I just ordered the Massdrop x Sennheiser HD 6XX (functionally equivalent to the HD 650 according to frequency response curves etc). This will be my first step beyond my trusty AKG K240 so I'm looking forward to hearing what all the fuss is about.

 

[Mike Reed]

This will be my first step beyond my trusty AKG K240 so I'm looking forward to hearing what all the fuss is about.

'Fuss' is one adjective I'd not associate with HD650s. 300 ohms, so need a reasonably powered can amp. Light, comfy and rugged. I can have my Focals and 650s plugged in simultaneously for quick A-B comparisons The 650s need a bit more volume than my (around 60 ohms) Focals. My friend has recently bought 650s (his first cans?) to go with nearly top Naim kit and expensive can amp on my recommendation, but he's not a headphone man.

 

[Fatmarley]

I don't know what a flat headphone response looks like, your ear shape and ear canal variations are variables, which make absolute measurements suspect

I know, but probably what's considered the most neutral Sennheiser headphones on this forum (HD600) do seem to have about the flattest measured frequency response.

 

[gustav_errata]

'Fuss' is one adjective I'd not associate with HD650s. 300 ohms, so need a reasonably powered can amp. Light, comfy and rugged. I can have my Focals and 650s plugged in simultaneously for quick A-B comparisons The 650s need a bit more volume than my (around 60 ohms) Focals. My friend has recently bought 650s (his first cans?) to go with nearly top Naim kit and expensive can amp on my recommendation, but he's not a headphone man.

By "see what all the fuss is about", I was just facetiously referring to the fact that they have been highly regarded for nigh on two decades.

Although I plan to buy a new headphone amp at some point (thinking the NJC Audio Reference Headphone Amplifier), for now I'll be using the amp on a pro audio interface that I have (Native Instruments Audio Kontrol 1):
Load Impedance: 8 ... 600 Ohms
Maximum output level: 4.26 V rms, 100 mW @ 100 Ohms
SNR (weighted): 102.5 dB
THD+N (60 Ohms): 0.006%
Frequency Response: 20 - 20000 Hz (+0 / -0.5 dB)

With my 55Ohm AKG K240's, I rarely have the volume above 9 o'clock, and practically never above 10 o'clock. So I think it should have enough juice to drive the HD650s (happy to take recommendations otherwise though). As an aside, I'm not sure exactly what DAC chip it has in it, other than that it's a Cirrus Logic, but I'm generally quite satisfied with the performance.

It was kind of a "chicken and egg" problem for me: upgrade my headphones first or my source first? I opted for the former, especially given the excellent deal that is the Drop/Sennheiser HD6XX. Now to start saving for the NJC amp/dac combo...

 

[Tony L]

Just had a quick look at two headphones I own, the Sennheiser HD-600 and Momentum 2. Responses are about what I was expecting (I’d read HeadFi which does similar measurements) and it is very obvious the latter has a bass-lift subjectively, but that is useful on public transport which is my context for these. The thing that surprised me is the amount of channel imbalance they measured on both, significant enough that I’d actually question their technique (headphones are notoriously hard to measure as a mm or so misalignment of the mic will impact the curve significantly). I’d expect the 2-4db at points of the treble to be audible. I haven’t checked my other brands to see is this is atypical though.

 

[gustav_errata]

Yeah I think the site would be better off doing 10+ measurements and showing the mean + standard error to account for obvious technical variability/noise.

 

[booja30]

I don't know what a flat headphone response looks like, your ear shape and ear canal variations are variables, which make absolute measurements suspect

Wouldn't you have to say the same thing for speaker measurements? Speaker sounds travel through your ear canals too.

Probably bone conducting headphones (and ESP) are the only sources where the ear canals don't factor in!

 

[Nero]

Indeed, however they would still be useful for comparative purposes.

....and the measurements here can only be indicative. The method used is similar to that used in a production environment, i.e. no allowance for ear impedance or pinna effects, so are valid for comparing frequency balance of differing headphones or for doing statistics on sample variation.
But at the end of the day, there's no substitute for proper measurements (doesn't have to be a HATS) using an IEC60318-4 ear simulator. This makes a lot of difference to measured distortion for example, and you can make proper HRTF corrections to simulate the 'free field' or 'diffuse' performance. These days, hi-res simulators are also available for measurements up to 40kHz, if that floats your boat.

 

[Nero]

I don't know what a flat headphone response looks like, your ear shape and ear canal variations are variables, which make absolute measurements suspect

They are variables, but like a lot of measurements, this is standardised. Ear simulators use impedance simulators and cylindrical ear canals, which is obviously an approximation, but if everyone uses the same......
New simulators use anthropometric pinnae and ear canals to get closer to the 'truth', but it's pretty well accepted what a flat headphone response looks like.
There's a nice AES paper (I'll find it later) which also shows how you can compensate for differences between headphones (frequency response and distortion) to make a true subjective comparison. The reference in that case was a pair of Stax SR-009s I seem to remember.

Edit: Download here. You'll need to register but it's non-invasive. Unless you are an AES member in which case go here

 

[gavreid]

Of course you have to convolute the measured response function with your own hearing response...

 

[Nero]

Convolve? Of course, but we do that every day, all day

 

[gavreid]

Convolve?

Either these days...

 

[ToTo Man]

diyaudioheaven is one of the more useful sites for headphone measurements IMO but I still wouldn't place 100% reliance on it, especially after experiencing first hand how finicky making accurate and repeatable headphone measurements is (albeit I'm using a cheap miniDSP EARS, but even guys like Tyll Hertsens with access to >$10k rigs still experienced difficulty and always posted measurements taken in many positions as well as an average). You then have the HIGHLY CONTENTIOUS issue of what compensation curve to apply to "remove" the colouration from the measurement rig (I'm not convinced by either of the comps supplied with the miniDSP rig and therefore just stick to the raw uncompensated measurement (I use the raw measurement of my HD600 as a mental baseline of how I think a flattish headphone should sound). All approaches are imperfect, but the most important thing IMO is to choose and stick with the one method and not evaluate headphones that have been measured on different rigs with different comp curves as such comparison is utterly meaningless.

 

[guey]

I just ordered the Massdrop x Sennheiser HD 6XX (functionally equivalent to the HD 650 according to frequency response curves etc). This will be my first step beyond my trusty AKG K240 so I'm looking forward to hearing what all the fuss is about.

I bought those a couple of months back, and am largely pleased with them.
They're certainly not hyped sounding - in fact I'd say a little dark. Perhaps that's no bad thing, as they're really good for long listening sessions.

I'd recommend people try the Sonneteer Bards Headspace gizmo. It seems to take the sound a little more out of your head, and a little more like listening to speakers. It's quite subtle, but at about 15 quid, worth a punt.

 

[ToTo Man]

Comparing all the Sennheiser headphones, and going by measurements alone (I've only heard and owned the HD650), I'd take the HD650 (lower midrange hump and all). It looks like the bass of the HD600 rolls off too early for my taste.

diyaudioheaven

Sorry if this has already been posted somewhere. It's the first time I've seen it, and being a measurist, I found it very interesting.

I own and have measured the HD600, HD650, HD660S and HD58X. It's important to note that intersample variations between the same headphone model can be significant, so basing decisions on measurement of a single sample can potentially be misleading. I tested this theory by buying two pairs of HD600 from Amazon within a fortnight of each other. I sent the second pair back because it had a very strong upper midrange that sounded coarse.

I've heard a total of three HD600, but only one HD650, one HD660S, and one HD58X, so I cannot say whether the samples of the last three models are representative, but you can see my Measurements here. The measurements mostly corroborate what I hear: Compared to the HD600, the HD650 sound a bit more forward in the upper midrange, a bit softer/veiled in the treble, and a bit thicker/woolier in the upper bass. Interestingly though, with a bit of EQ I can easily make the HD650 sound just like an HD600 and vice versa. The HD58X has the most extended bass response, and has less upper-midrange than both the 600 and 650.

PS - I suspect I might have a golden pair of HD600 as I do not hear or measure the steep roll-off in the bass that other folk report. My dad's HD600 roll-off more sharply below 50Hz according to my measurements. They also have significantly more HF energy (hence my point about intersample variations!).

 

[Mike Reed]

They're certainly not hyped sounding - in fact I'd say a little dark.

Yup, I'd agree. I liken them to Koetsu cart's and they're pretty popular. Measurements mean nothing to me as if the sound gets through all the detritus in my lugholes and still waxes (!) lyrical, everything's fine.