First Crunched Data From Harman Head Measurement Session

These are two potential compensation curve that may be derived from this experiment: The overall sum of measurements from the stereo pair; and the combined left and right ear response when listening to the left speaker only. There may be others.

I don't think I'd make a very good researcher. I get too pissed when the numbers don't look like I'd like. Guess that's why so many engineers are less in touch with their feelings than the average Joe. Oh well, here we go.

First I'll show you all the data.

Test Description
In this test, a 256 point stepped sweep was performed from 20Hz to 20kHz. The head was positioned at the reference listening point and fifteen measurements were made at different angular positions—azimuth angles (left to right) were +/-20, +/-10, and 0 degrees; these measurements were repeated with the head level, and tilted up and down 10 degrees. Left and right ear data were taken as separate files.

The speakers were set up in the room to provide the Harman preferred listening EQ. Here's a plot from a measurement microphone used to set up the the room for my visit.

MeasuringHeadAcoustics_Measurements_Plot_RoomMeasurement

Data recorded from the head was stored in Excel spreadsheets, one for each angle. The raw data was quite noisy. In my initial previous post, I smoothed the data with a moving average filter that I can manage to pull of in Excel. Good for a rough idea of what's going on, but not good for developing a compensation curve that doesn't have a lot of weird artifacts in it.

Fortunately, I know Arnaud well enough that I knew he might give me a hand with a bit of complex best-fit curve fitting math that can't be done in Excel...which he agreed to do (bless his nerdly heart). On all the data shown here, Arnaud has used a "d-2/xhat-6OB" Tikhonov 1/6th Octave Band smoothing derivative 2 best-fit filter. Don't ask me what that means...hell, don't ask Arnuad for an exact meaning. Here's what he wrote me:

Started looking at smoothing functions:
- I used open source code Octave and some data-smoothing package that was made available for a kind soul.
- I have to be honest with you and I am not a math guy per say so I could not explain the method in detail, but description of the package here mentions about Tikhonov regularization.
- I also tried the "standard" interpolation function using spline method (interp1 mentioned before)
- After interpolation, the curves go through a second fit (cubic interpolation using interp1 function) in order to get back to the original frequency resolution (e.g. so that you can use these directly in Excel to normalize all your existing spreadsheets)
- All interpolations are made directly in dBU scale, which is the way to go I think because your data is quite jaggy (would have helped to perform longer averages but time was of essence I assume)

Have played a bit today with the various options using 3 files and have attached the results.
- "d-" value corresponds to smoothing derivative used to perform smoothing (Tikhonov method)
- "xhat-12OB" is a best fit to 1/12th Octave Band set of frequencies (Tikhonov method)
- "xhat-6OB" is a best fit to 1/6th Octave Band set of frequencies (Tikhonov method)
- "gcv" is magic to me, could not tell you what it does (Tikhonov method). Besides, it does not systematically converge (iterative method) and you'll notice some blue color curves that do not seem smoothed (failed at 50 iterations…)
- "spline-6OB" uses a spline interpolation method onto 1/6th Octave band set of frequencies (standard interp1 function)
From what I have seen so far, "d-2/xhat-6OB" interpolation is the one to go for (red color curve in …_SmoothingTest4.png).

Here's one of the plots he sent. He modified about 30 spreadsheets for this project. Thank you, mate.

MeasuringHeadAcoustics_BestFitCrunch_Plot_ArnaudPlot

I then took all Arnauds data and re-built and modified my pre-existing spreadsheets to be able to display an array of summary data. We'll start with the stereo data with both speakers active.

Stereo Response Tests
MeasuringHeadAcoustics_BestFitCrunch_Plot_StereoAzimuthFamily

The above plot shows the left (top) and right (bottom) ear family of plots for movements in azimuth (left to right). At each angle, measurements from the three elevation angles are averaged to provide some angular averaging to remove fine features from the HRTF (head related transfer function).

A few things to notice right off the bat:

I would have expected the bass to come down to the 0dB line by 150Hz. The room tuning clearly shows a shelf in the bass that ends at 200Hz. Looking back at the initial data, it seems the best fit smoothing algorithm is extending the bass boost curve upward in frequency and taking too much detail out of the bass profile.

The right ear (bottom traces) shows a significantly more apparent downward inflection as the curve rises between 1kHz and 3kHz. Keep an eye on this area as it seems pretty lively in the measurements. Something is going on here and I know not what.

I've really liked how the presence and coherence of vocals improve when I EQ to match the tonal profile of the Harman curve (below) in the 200Hz to 3Khz area. The recorded data shows the Harman curve might need a little tweak to raise the peak to 3.5kHz and re-shape the inflection at 2kHz a bit for application with my head.

MeasuringHeadAcoustics_FirstPass_Plot_HarmanTarget

It's very clear to me from these measurements that the prescribed ever-increasing fall in response above the 3.5kHz peak prescribed by the preliminary Harman target response (shown above) does not match my head's response.

While the fall from 3.5kHz to 7kHz is very similar as head angle changes, the the area above 7kHz seems quite effected by angular change. We'll see this more clearly in the left speaker only data.

MeasuringHeadAcoustics_BestFitCrunch_Plot_StereoElevationFamily

The above plot show the family of measurements of the left and right ear for up and down movement. Each plot is the average of all left to right movement of the head at each elevation.

Generally speaking, the plots don't show much change except for the area between roughly 600Hz and 3kHz. In that area there is significant change...especially at that 2kHz inflection. Maybe...just maybe...there's a null in the room at 2kHz near the right ear and as I tip the head up and down the ear moves nearer and farther from the null. Dunno.

MeasuringHeadAcoustics_BestFitCrunch_Plot_StereoSums

The above plot shows the sum of all measurements combined for the left and right ear, and the sum of both ears together. This, it seems to me, is a reasonable candidate for a headphone compensation curve for my head. I continue to dislike the bass bump extending to 400Hz, however. That's never sounded right to me...I'm going to have to do something about that.

Left Speaker Only Response Tests
MeasuringHeadAcoustics_BestFitCrunch_Plot_LeftAzimuthFamily

The above plot show the family of measurements for azimuth changes of the left and right ear with only the left speaker playing. Each plot is the average of all up and down movement of the head at each azimuth angle.

It's cool when things look as expected, and the above plot is cool in that way. First notice the peak at 3.5kHz: This peak occurs due to the sound focusing effect of the concha bowl (the small bowl around the entrance to the ear canal). The left ear is variously facing the left speaker, and you can see that it is quite stable at achieving the needed gain on that side of the head to hear sounds at this frequency—like a twig snap. (Evidence of evolution—or God—helping us hear the approaching tiger?)

But the right ear, on the far side of the head from the left speaker, is shadowed from the directly propagating sonic wavefront of the speaker, so is ever more unable to focus the sound and looses gain. Yet another psychoacoustic cue to know where that sound just came from.

Higher resolution localizing cues are to be found in the area above 7kHz. This is a noisy area, but if you look closely you can see the trends. The first thing to ignore is the dip and bump at 11kHz and 13kHz. These occur from a tertiary resonant mode in the ear canal. All the plots show evidence of it—the more direct the sound into the ear canal, the more pronounced the wiggle. Just smooth it out in your head.

Now look carefully at the area between 7kHz and 10kHz and compare it to the area above 10Khz. What I see is that as the face of the dummy head is ever more pointing directly toward the speaker, the stuff 10kHz to 15kHz rises, but between 7Khz and 10kHz it sort of falls off. Essentially, the tilt in response above 7kHz goes upward as you face the speaker and downward as you turn away. As you directly face a sound source, the top octave tips up; as you turn away, it rapidly falls off. If you want to hear very high frequency sounds, I guess you should face directly towards them.

It seems to me, when looking at the plots so far, the data shows that the fall after 3.5kHz should end at about 7kHz 3dB above the midrange baseline, then remain about flat 'til 15kHz at which point it may roll off. This is substantially different than the Harman target response.

Very interesting to me is that I know this response would be too bright for my personal tastes. The Audeze SINE has a fairly similar response, and while I knew it would be a lot of people's cup of tea, it was still too bright in those top octaves for me.

This exercise have given me new insight into where neutral is, and a more precise understanding of my preference for a more laid back sound. I will certainly be dialing that into my understanding of how to review headphones more fairly when my taste drags me away from neutral. John Atkinson brought up our differing appreciation for the treble range in his LCD-4 review, and he like this profile much more than I.

One interesting thing to keep in mind though is that if the area above 7kHz is tipped up, it may be that we begin to perceive the sound as coming from more directly in front of us. This may psychoacoustically narrow the stereo image as the sound is perceived as coming from more directly in front, rather than 30 degrees to the side. May!

MeasuringHeadAcoustics_BestFitCrunch_Plot_LeftElevationFamily

The above plot show the family of measurements of the left and right ear for up and down movement with the left speaker playing. Each plot is the average of all left to right movement of the head at each elevation. Nothing too terribly important here, but a couple things worthy of note.

The inflection at 2kHz seems to be more pronounced in the left channel where in other data it seems to be more pronounced on the right. No idea why this is so.

It's clear that the region above 7kHz is tipped up in the left ear, and downward in the right, reinforcing the idea the more your head is turned toward the speaker, the more this region tips up.

MeasuringHeadAcoustics_BestFitCrunch_Plot_LeftSums

The above plot shows the average of all measurements from the left and right ear with the left speaker playing. The right ear show a clear loss of concha gain relative to the left ear, which faces the speaker. Again, above 7kHz the left ear is tipped up and the right ear rolls off.

The top green plot is the sum of both curves. This is the second potential plot I will consider as a compensation curves for my system. With speakers, both ears hear both speakers. The left ear hears both the left and right speaker. It's (somewhat) reasonable to assume that the right ear hearing the left speaker will be quite similar in response to the left ear hearing the right speaker. Therefor, adding the left and right channels together on this plot may be like the left ear hearing two speakers and is therefore a reasonable model for a compensation curve.

MeasuringHeadAcoustics_BestFitCrunch_Plot_ComparingSums

The above plot shows the averaging of all measurements in the stereo test to those measured with the left speaker only. It may be reasonable to assume that the difference in these to curves comes mostly from having two speakers playing in the room vs. one. Bass increase due to the confinement of long wavelength sounds may account for the added bass of the stereo plot. I have no rational guess as to why the area between 500Hz and 3kHz looses response with the stereo signal—maybe acoustic cancelations at those frequencies? (Wild-ass guesses in the comments are welcome!)

In the coming week I will try to contact and get comments on these findings from some of the (much better educated in the field than I) researchers and engineers who have been following along and helping me through this process. I will summarize and report later.

For now, I've taken a first quick stab at compensating a few headphones with the two above empirically derived curves. The following plots show the Audeze LCD-4, and Sennheiser HD 800 and HD 600, compensated with both the stereo and left channel only total sum curves shown above. The left channel of each headphone (top blue trace) is compensated with the stereo data; the right channel (top red trace) is compensated with the left speaker data.

I'd love to hear your which you prefer and why in the comments. (Note that the HD 600 looks like there is more difference in the compensations; this is happening because the channel matching of that headphone is a bit off.)

Trial Headphone Compensations
Audeze LCD-4
MeasuringHeadAcoustics_BestFitCrunch_Plot_LCD4Compensation

Sennheiser HD 800
MeasuringHeadAcoustics_BestFitCrunch_Plot_HD800Compensation

Sennheiser HD 600
MeasuringHeadAcoustics_BestFitCrunch_Plot_HD600Compensation

Further Steps
As I mentioned, I'll be reporting back on comments from engineers, but I'd like to do a few more things.

I'll be asking Arnaud for a little more help. I want to see if he can play around with the smoothing algorithm coefficients to get some more detail back in the bass. Also, the gathered data sets aren't exactly the same size and length as the headphone data I gather. Data was gathered for this experiment from 20Hz to 20kHz in 256 log steps; my headphone measurements are 10Hz to 22kHz in 512 steps. We'll need to convert the curves above into ones I can just tip into my spreadsheets. (I did a crude scale conversion by hand for this article, but want a really clean one for further exploration.)

Once I have some usable compensation curves in hand, I will produce a series of graphs showing the raw response; responses compensated with the two above plots; a Harman target response compensation; and with diffuse field, free field, and independent of direction compensations. I'll do this for maybe a dozen headphones, and then I'll post each headphone plot as a poll in which readers can vote on which compensation curve they think best represents each headphone. This should give us some interesting feedback on these curves.

That's it for now.

COMMENTS
Rillion's picture

If the tones coming from the speakers for stereo measurements are in-phase, I would have expected to see some comb-filtering. Maybe room and torso reflections are enough to smooth these out? Maybe with the comb smoothed out, this could account for the slight decrease you noted in the stereo between 500 Hz and 3 kHz? I'm no expert on this stuff--just speculating.

Seth195208's picture

.. between .5khz and 3khz, Is due, I believe, to complex interference from the dummy head to the individual R & L signals before entering the respective ears. The deepest null occurs at about 1.8khz, about 7.25" in length, which is probably the width of the head itself.

Rillion's picture

Reply to Seth195208,

That is what I was thinking: that much of the interference would be from the delay between the left and right ear, but apparently there is more going on than just that.

Rillion's picture

The "delay between the left and right ear" is not exactly what I meant. To state it more accurately: the delay between the near speaker sound and far speaker sound to a single ear (which is related to the distance between the ears via geometry).

Tyll Hertsens's picture
Yeah, I think you guys are right.
Seth195208's picture

How wide is the dummy head, Tyll?

Rillion's picture

Did you try recording any music samples through the dummy-head in the listening room? This could be useful in figuring out the optimal amounts to weight the panned and stereo runs--perhaps by doing a mid-side decomposition. If you only used tones or noise, music could probably be simulated with mixing and equalization after-the-fact but that would be more complicated. I have a WAV file containing 4 minutes of random 1 second samples from my entire music collection that I could share. I have similar files split into classical, rock, and piano "genres". The average spectra are surprisingly similar between these genres.

Rillion's picture

I should have addressed that to Tyll, for clarity.

Tyll Hertsens's picture
No, sorry, not enough time to do too much playing around that day.
Hands's picture

First, everything below 400Hz looks way off on these headphones based on what I've heard. From what I've experienced with measurably mostly-flat speaker setups, using the most typical measurement methods that everyone and their mom knows about (for better or worse), all of these cans should have at least decent bass extension. Even the HD600 should be relatively extended down to 30-40Hz relative to mids and treble, and both it and the HD650 share that fun bump around 100Hz relative to a flat speaker setup. Sure, you won't get the same feeling as a subwoofer, but the extension should be there regardless. (Though the fun bump on the HD6X0 does sorta help.)

Then again, I'm not at all sold on this bass boost that Harman or whoever keeps pushing. All of the headphones I've heard thus far that are supposed to be close to the Olive Welti target curve sound bassy as hell and overly thick to me, like the sort of dual, ported subwoofers high schoolers run in their cars. I'm probably missing the point, but if we're looking for compensations to best match neutral/how we hear it, I'm not feeling this bass boost even if it's used on headphones to try to recapture the feeling of a subwoofer, which they'll never replicate in the first place.

All that aside, the left channel results on the HD600 most closely match what I heard on my pair: relatively even/flat-ish overall in the mids and treble, with a bit of an upper-mid boost (between ~1-4KHz or so on my pair), a lower treble notch, and a bit of "wiggle" in the otherwise balanced spot around 6-10KHz. It's not an exact match but, then again, I am not using the same measurement rig.

Tyll Hertsens's picture
Thanks Hands. I'm thinking hard on the bass problem. I don't think manufacturers have caught up yet with techniques to provide a reliable boost---even a small one---without things going out-of-phase or getting mushy. It may be the thing to do is not have the boost in the compensation and let folks mentally add it in if they like it. I think the overall tilt is probably good to keep.
Hands's picture

To be honest, I'm familiar enough with your results that you could throw in an entirely wonky compensation curve, and I'd still be able to go off raw results pretty well.

Ultimately, it's your call to do what you think is best, being the expert, and you might be exactly right with your thoughts on techniques to boost bass without mucking things up.

Another idea worth considering is posted graphs of 2-3 different compensation curves. I've seen a couple other sites that do this, and it's at least interesting. Might mean more work than it's worth...

skris88's picture

I hated the fun bump on my HD-600s and switched to the HD-650s. A tinny weenie bit flatter but still gets me that unless and find and twist a 125Hz knob or slider on an EQ by 3db, I am not satisfied. Have listened the the HiFiMAN HE400Ss and while they measured a flatter bass, they seemd to have a one note bass signature, perhaps it was my own hearing bias after 10+ years of the HD-6X0s. The clincher was that the sheen of high treble (and I'm old, I know my hearing is going) was missing on the HE400Ss. So I'm stuck with my HD650s and look for an EQ whenever I listen to them - which is less and less :-(

Apart from the Fun Bump they are marvelous! A silky smooth top end nothing can beat. And, if at first they sound dull to you, give them a chance, it'll take some 15 mins or so - if you can bear it. Then you'll wonder how you put up with the poor treble of all those other cans.

Argyris's picture

To no one's surprise, the HD 600 lines up quite nicely with these compensation curves, apart from the bass rolloff. I look forward to the promised plots for the other headphones.

Also, without wanting to come across as demanding or imposing, is there any possibility that a different HD 600 can be used? Maybe somebody reading could send theirs over? The channel imbalance in the particular specimen measured has long been a confounding factor when I've tried to read any plot using its raw data.

Tyll Hertsens's picture
Yeah, I'll try to get my hands on a nice specimen to measure. They're a good reference for this sort of thing as so many people have heard it and like it.
Argyris's picture

I also have a suggestion for one of the headphones to test: either a Sony V6 or 7506. Not because it's a great headphone (I don't think it is), but because it's a prototypical example of DF (or strongly DF-like) tuning. It would be useful to set the differences between the newer curves and the older DF standard in relief. Again, I'm not trying to be difficult, and if you don't have easy access to one or have other plans, that's perfectly fine. It's just something I would be interested in seeing because it draws such a pretty flat line using the usual compensation, yet it comes across as upper-middy, bright and grainy to me.

Tyll Hertsens's picture
I have V6 measurements...I'll try to remember to include them with the next go 'round. Whack me on the head if I forget---it's a good idea.
Aufdemaury deus ex machina's picture

Sorry But Why Is only the only company starting to raise flags and concerns about Tyll's system the very instant he, for the first time out of a myriad Audeze reviews, finally gives them a mediocre review.
Sry but I've seen some very very credible comments in Audeze's initial post from the community &
I just don't buy it, there's a huge conflict of interest here.

If Audeze came out to defend other negatively reviewed headphones that would seem far more ernest on their part. When

Tyll was gushing about the Audeze LCD 2's & 3's as well as x's and all other new updates, Audeze didn't say A Single Word questioning the merits of his measurements. Then suddenly
Tyll comes out with an Audeze Lcd 4 review then two weeks later Audeze comes out with a post,
raising concerns of the merits of his measurements process. To me, it's seems defending and promoting their products is their only interest and priority.

If you accepted glorious praise from Tyll in the past and kept your mouth shut then,
then you should accept his criticism when he's trying to give you critical feedback,
that's constructive and helpful. Tyll has also kept close in touch with many of his Audeze
contacts, so it's not as if he isn't attentive or ignorant. As a company who strives for identity
in the objective accuracy, Audeze shouldn't engage in a huge conflict of interest as this.

It would also be polite and supportive of Audeze to keep an open conversation with Tyll here at Innerfidelity to analyze the data he's getting an pinpoint certain problems or things to improve etc.

I emplor anyone to read Audeze Post very carefully, look at the comments, read and learn from them,
as their is a lot of useful measurement and reference Audeze own plots graphs they provide to their customers (this can be found with a simple google search) to Tylls measurements. Also you can do the same with Sennheisers Hd800 Plots they provide their customers that they distribute. I really dislike it
when companies engage in such politics.

Also Examine The Headphones Graphs in question:
Audeze Lcd 4
Audeze El8

Then reference these headphones to the likes of other similar headphones graphs:

Most resources can be found at innerfidelity.

In Addendum, yes I believe measurements WILL vary from lab to lab - facility to facility.
Factors such as the Concha, Inner ear modelling, and pinna will also have an affect,
but as I've brought up before in Audeze's initial post, From person to person IRL these
WILL ALL VERY REGARDLESS, so it's far more important to have a head/model that is accurate
to the most common response across peoples hearing - Essentially a measurement rig that is standardized to measure very similarly to the larger majority of people.

I've disagreed with some of Tyll findings in the past, when referencing other measurements, as we'll as his own data base, his interpretations of them etc. Being an avid user of an equalizer as well as someone who has a tone generator with sweeps, square waves, noise, i have to say my own observations and experiences with the gear that Tyll has measured in his database (HD 800, HE 500, HD 650, Dt880, Stax 009, Audeze LCD 3, 2 & X, AKG Q701) I've more or less found my own impressions of these headphones quite consistent with Tyll's observations. I've also demo'd the
Audeze El8 with music of course, next music with an equalizer,then white noise,pink noise, then last with a linear scale sweep from 20 to 17500hz. & if I found anything is that there's was a huge dip in the 6-8khz reason, this being privy to Tyll's measurements. I Invite anyone to try this simplistic test with the EL 8 and share your opinion.

I also have Audeze plots graphs that I received from Audeze themselves as well as sennheisers own plot graph of my Hd 800, and found them both to be consistent with the measurements here at inner fidelity) I can send anyone a picture of my headphones serial number along with the corresponding graph Via email.

in the end it's healthy and helpful to earnestly reach out to Tyll to give him the rundown and tips to improve his measurements rig, that's all good business, but the way Audeze approached this is shows semblance to bias, politics of proprietary interest & their Agenda, not to mention a huge conflict of interest. Tyll's truly an earnest guy to keep soldiering through these rough trials he's working towards and confronting as well as trying to resolve this issue with answers and data.

That's my take on it

Aufdemaury deus ex machina's picture

Can Be Found Here

http://www.innerfidelity.com/content/how-do-measurements-sound-audeze#xz...

Read the comments, Read the review

Tyll Hertsens's picture
Yeah sure, their writing comes from their perspective, but they were simply trying to clue me in on something...and, it turns out, they were cluing me into something important that I didn't have a grip on.

This exercise has been eye opening and I thank Audeze for bringing it up. Don't read there stuff with a jaundiced eye.

kais's picture

So far you always used left ear/left speaker as "main" or dominant source.
One has to expect room modes and other sources of asymmetries.
Use right ear/right speaker the same way, then average the two.
This will increase the accuracy by a factor of about two and even out a lot of whiggles, so smoothing will be a lot easier.

arnaud's picture

I am not sure Tyll has this data for the other speaker. But, in regards to making the interpolation job less of a random quest, I would be tempted to average out the various headings first and apply a cubic interpolation at last (basically going back to 1st article tyll published but replacing the moving average by cubic interp.).

kais's picture

Looking at the example you show it's obvious that neither "Tikhonov method" smoothing does represent an average of the raw curve.
Choosing some special kind of processing of which you don't even know how it works, just because it looks good, is not a correct scientific approach.
The smoothing may not even go beyond what's done later in the headphone measurements.
Otherwise you would provoke "noise" (as you usually call it) in the measured curves that is not representing the headphone, but uncompensated irregularities in the heads response.
In the low range (ca. below 500 Hz) one can assume, proved by looking at the numerous headphones measurements that are already existing, that the head is linear.
Even looking at your raw room curves points into that direction.
If you want to go for a "taste" compensation here (Harman Target or your own), you can just dial in what you like.
BTW I'm not 100% convinced that this is the right approach, but be it.

Mauro's picture

A better way to remove random noise would be doing several acquisitions in the same condition and then average them. This method allows to avoid introducing errors with criptic averaging post processing methods, as it is commonly seen in the audio community.

Do you have many acquistions of the same configuration?

kais's picture

Sorry, little misunderstanding, because I used the term "noise" here like Tyll often uses it.
He calls the little wiggles in the higher range of the frequency response "noise".
This has nothing to do with real random noise.
Random noise does not seem to be a problem with this measurements.

Rillion's picture

Here's a fairly simple technique for smoothing that has worked well for me when the data (dB vs frequency) are at many equally-spaced frequencies. It is even simpler if the frequencies are already logarithmically-spaced.
1. Select a fairly coarse set of logarithmically-spaced frequency bands. I use 61 bands, but you can experiment with that.
2. If the data are already logarithmically-spaced (assumed to be greater than 61 bands), then do simple averaging of the data within each of the 61 bands. If the data are instead at equally-spaced frequencies then weight each point when averaging by 1/frequency. In mathematical terms, this weighting factor comes from the Jacobian for converting an integral (approximated here by a sum) from a linear to a logarithmic scale.
3. Now that your data have been reduced to 61 (or whatever number) logarithmic bands, you can use a common interpolation scheme (such as cubic) for intermediate points. Interpolation is often available in spreadsheets, at least for graphing purposes.

arnaud's picture

Indeed, the cubic interp of some nth octave band averaging may be cleaner. [tyll, maybe this will give you what you what you are looking for at low frequencies].
Tbh, I did this at night and early morning on a week end when kids were asleep so I haven't really explored all options ;-).
Arnaud

Tyll Hertsens's picture
Dude, you did me a solid. Your smoothing did very nicely in revealing the trends in the midrange and treble. We'll chat by email about where to go from here.
tony's picture

Science and Engineering should be feelingless!

2+2=4, period!

However our music releases Dopamine in our Brains and that IS feelings, the better the gear the better the buzz.

But there seems to be quite a few folks out there buying gear for ( I think ) Social reasons, to show off, to write on Head-fi about all their stuff and to tell what they think about the new released products ( like these Audeze 4 transducers ) or the $4,000 Focal Utopias, etc., etc........

I spent 4 years in Research, the most reliable findings we could get were from using Calibration Standards to calibrate our instrumentation. Actual measured test performance is surprising and educational. ( which is why we measure )

Those HD600s deliver a wonderful high or buzz.
The HD800 delivers astonishing emotional highs, I doubt that better headphones exist.

Better Recordings do exist and that is where a better "listening" experience is to be found.

If Engineering becomes exciting, it's when the Engineers develop HD800 headphone experience levels using the iPhone. ( which I think we are seeing now)

Someone has to figure out how to EQ these headphones properly - Harmon to the rescue.

Someone has to understand it and explain it - Tyll to the rescue.

Thank you for all this hard work.

Tony in Michigan

arnaud's picture

I agree with Kais about smoothing being a dangerous correction to an imperfect recording. From email exchange with Tyll, issue is that he used a single sine sweep so there's no time averaging. In my (little) experience from frf measurements in somewhat reverberant rooms, you never get smooth looking curves however.

I also agree with hands that the equalized sample phone results look very off at lower frequencies. Actually, I think the issue is that speaker response at the head location is not flat to start with (harman bass boost), hence the measured responses can't be used as is. We need to normalize the measured response on the dummy by single mic at head center location when head is not present (e.g. get some HRTFs from Tyll's measurement).

I think Tyll wanted to find a way to get a better normalization curve for his head, taking into account the work harman has done, hence went there to try get measurements done. My understanding is that he should add this harman eq (such as the bass boost) after normalization of his headphones (white noise) measurements by above HRTF (or rather the suitable average of various headings HRTFs).

At the end of the day, we may be heading back toward the ID eq. tyll has been using for years but, at least now he can clearly show which room / heading averages he used for normalizing his headphone measurements.

Cheers,
Arnaud

Pages

X