HiFiMAN Sundara Around-Ear Open Planar Magnetic Headphones Measurements

Click on graphs image to download .pdf for closer inspection.

Raw frequency response plots show very little change in response with position change on the head. Reponse from 20Hz to 600Hz is quite flat. The Harman target response would indicate a gentle rise from 600Hz to about 1.5kHz is desired, but the Sundara remains prett flat in this area producing a somewhat laid back presence response. Vocals sound a bit distant to me.Though the rise to the peak at 3.3kHz is a bit abrupt, the peak is about 13dB above baseline...pretty much exactly where I like it. The falling response above 3.3kHz is nearly perfect as well, but for some glitching between 9kHz and 12kHz, which is almost certainly a result of normal (and usually inaudible) ear canal resonances.

But for a slightly too high initial transient, the 30Hz square wave has good shape for an open headphone with little sway-back and remaining above zero until the very end of the waveform. Though it's worth mentioning that Audeze headphones can sometimes do a bit better.

300Hz square wave initial transient and following rings are a bit ragged, and subsequent noise a bit high, but frankly, it sounded a somewhat better than this looks. I'm not saying the headphones sounded liquid smooth, they didn't, but I would have expected them to be a bit more grainy sounding than they were given the measurements. Impulse response is likewise somewhat noisy. It's guessing we'll see some some "Ortho Wall" action 2kHz, 6kHz, and 9kHz when the SBAF boys do their CSD plots, but I reckon it'll be fairly low level stuff.

Distortion is fairly low across the board, but a slight rise at 2kHz and 6kHz may evidence some mild break-up at those frequencies.

Impedance is dead flat at 43 Ohms, with only a gentle change in electrical phase in the top octave. The lack of visible features in the impedance plot usually points towards a well behaved headphone.

Isolation is minimal and typical for an open headphone. Needing 245mVrms to drive these headphones to 90dBspl at the ear indicates it will not be easily driven to loud levels from portable devices. I'd recomend and amp or dedicated DAP for portable use.


Suuup's picture

Utopia btfo.


HE6 btfo.


What a marvel. Definitely agree with the WoF this time. Might have to get a pair.

zodiac1234's picture

Since English is not my mother-tongue, some of the things I mentioned on the headfi forum (Sundara review page) might have been interpreted as offensive, Specially with regards to Mr Tyll Hertsens.

I assure all of the readers and Mr Hertsens that it was most certainly not my intention.

I have followed the wonderful work Mr Hertsens has done since Innerfidelity started, and have been a great admirer of his relentless efforts and hard work to create a center for higher fidelity head-audo lifestyle and sciences.

I have specially enjoyed reading all the technical articles he has kindly provided for the community, and enjoyed all the informative videos he has created and shared with the rest of the world.

If my comments on the headfi forum - which was meant as just being funny - in any way upset or offended Mr Hertsens - please accept my apologies.

Best wishes:
headfi member zodiac1234

Impulse's picture

I don't miss this kinda Head Fi drama, not even a little bit...

Tyll Hertsens's picture
No worries. My motive was to point out the type of errors that people unwittingly make when they're making assumptions. Sorry that put you in the cross hairs.
arthur li's picture

It is likely that Hifiman tunes the Sundara to match the Harman Target Response Curve, since researches showed that average listeners prefer this tuning. If that is case, Tyll may actually exert some influences on HFM's decision. After all, Tyll is one of the major advocates of the Harman Target Response Curve. The speculation that HFM tunes their headphones in a specific way just to please Tyll is absurd though.

南开米饭's picture

you cnnut..

Impulse's picture

Speaking for HFM?

arthur li's picture

I didn't say Hifiman tuning the headphone to please Tyll. You misunderstand my words.

By the way, why are you using the account of Dr. Fang on Weibo as your account on this website? You are clearly not Dr. Fang. You seems to be a troll.

lambdastorm's picture

Save your words and frick off somewhere else. The community doesn't need another Fang impersonator when the real Fang's busy gluing drivers together.

jherbert's picture

Distortion seems to be a bit high for higher frequencies. I would have expected these to be much lower given the new and thinner diaphragms.

南开米饭's picture

LCD on WOF is big big mistake!

Impulse's picture

In the 2nd sentence of the paragraph under "What does all this have to do w/Sundara", it seems like there's a word missing in the phrase "I don't mean the Sundara brighter than I like" (maybe 'sounds' after Sundara?).

In the second to last paragraph before the summary, 3rd or 4th sentence, I think you meant to write alley and not ally (right up your alley). Great review btw, nice to see more heated competition in what used to be a dead $500-1K segment.

Wick's picture

Not to mention complement instead of compliment. But why are we discussing this?

Impulse's picture

He'd wanna fix them for posterity? /shrug

mike138's picture

I wonder if this is supposed to be the spiritual successor to the HE-400i? When the 400i was first released, I think it was the same price as the Sundara. Of course, within a year or two they were going for about half that....

crazywipe2's picture

It's really a bad message for the community to think that manufacturer alters the frequency response to match Tyll taste. The clarification was very good.
Honestly, I liked all the headphone I bought that were on the "Wall of Fame". I can't say the same for the headphone recommended on other websites (in most cases I didn't like them at all) This is a testament (for me) of the great objective work of Tyll.
I can understand that sometimes some "Guru" or "Fanboy" is upset for the bad review, for a childish reaction. (just look at the Sennheiser hd660s thread :D)

Keep the good work Tyll!


krimsonking's picture

I've always wondered: why do the frequency response graphs go "crazy" in the upper registers? Even the best headphones don't seem to have flat responses up there.

Tyll Hertsens's picture
A couple things: First there are the ear canal resonances that occur...around 5k, 9-10k, and 15k usually. In the top octave the wavelengths are short enough to stimulate modal resonances and cancellations somewhat randomly in the volume of the earcup.
Iliketrains's picture

Do you hear any of the 7k/10k peaks in this case? thx

krimsonking's picture

That makes sense. I had a feeling it had something to do with resonance due to the short wavelengths. Has any company managed to "solve" this issue or is it not worth the time/effort?

AMW1011's picture

I'm not the person you responded to from Head-fi, but I'd like to give a possible explanation for their take on neutrality.

Frequency response neutrality is at best an ill defined concept, and it generally comes down to what target response curve you identify with. You, Tyll, seem to prefer the Harman Target preference curve and Sean Olive's work. By these metrics something like the HE-1000 is indeed bright in treble response. However, another very popular target response is the diffuse field target, of which the HE-1000 much more closely aligns to.

Here is your measurements of the HE-1000 compensated around the diffuse field target: https://cdn.discordapp.com/attachments/243448552313847808/42020295354299...

Here it is aligned to the Harman preference curve: https://cdn.discordapp.com/attachments/243448552313847808/42020303650239...

I generally align much closer with the diffuse field target myself.

arthur li's picture

The Harman Target Response Curve isn't developed as a target curve for neutrality in the very first place. It is nothing more than a particular tuning of headphone that pleases some small samples of listeners. Every time I see Tyll writing about approaching neutrality, I am confused - how can we reach Rome if the only thing we know about Rome is that it's called Rome?

AMW1011's picture

That's a fair point. I don't think Tyll considers the Harman preference curve to be neutral, but what he does consider neutral seems fairly close. Certainly not the diffuse field target. Maybe he can elaborate.

thefitz's picture

In the time I had my HE-1000, I "heard" it much like the Harman curve. For all this talk about smooshiness, man was it harsh.

chik0240's picture

Hi Tyll I know it’s pretty much comparing with ancient dinosaur, but I look at the fr curve it looks like the he-500 which I am still enjoying, any idea how close or different they sound? To me the HE-500 was pretty close to neutral like the uerr but with less details and tighter bass

maelob's picture

Tyll, apologize if you have already discussed this, but I would be interested in finding out what is your preferred target response. It is interesting to me, that while the Sundara measures closer to the Harman target (which i think you are a big advocate) than the Aeon Open, you still prefer the Aeons which does not measure as well. I know you mentioned that you prefer a warmer tilt, but how much different from the Harman? To me this is a good example that not everything is measurements , and that sometimes headphones that don't measure "well" still are enjoyable.

buckchester's picture

I have a question about innerfidelity's measurements. They consistently show the bass to be flat down very low on planar. For example, the HE400i. However, I bought Sonarworks EQ, which indicates a sharp roll off in these headphones just below 100hz. Their measurements of many other popular headphones show bass roll off in much higher frequencies than Innerfidelity's measurements.

When I play bass test tones on my he400i headphones, it seems to support Sonarworks measurements since the volume of the bass seems to decrease quite dramatically and progressively below 100hz.

So why do innerfidelity's measurements show a pretty flat response down to 20hz?

crenca's picture

I don't own a HE400i, but I do own a HE4XX (massdrop version - apparantly slightly modified) and I find (via subjective experience, though I am not a "noob" when it comes to HP's) the bass to be right where Innerfidelity's measurements say it should be. Of course, the difference between measurement rigs/methods is also widely know. I wonder how Sonarworks measures HP's, and do their measurements correlate with others (such as Innerfideality's) more often than not, or is there a consistent difference across many model/makes of HP's?

Skycyclepilot's picture

Good question. My only complaint about my 400i headphones is that they lack punch in the very low bass range. My Oppo PM3 set is much better in that regard.

Brad331's picture

Sonarworks' other measurements don't have this characteristic. I think they just failed to get a good seal when measuring HE400i. https://sonarworks.zendesk.com/hc/en-us/community/posts/360006146374-Hif...

gefski's picture

For me, these reviews (and the follow-up posts) have become far less fun and informative than they used to be. Similar, in fact, to the dreary "impressions" on Head-Fi, most of which are only about frequency response. It's as if F/R measurements and impressions tell us all we need to know about headphones.
I'm at
Truth of timbre, texture, touch, graceful and natural dynamic flow, real instruments in space, the perception of hall sound even in "silence", are just a few things that matter. Such music listening easily reveals, for example, the HE-1000's price justification compared to the HE-560 and HE-400 from the same manufacturer.

How about more subjective impressions of products' real musical merit??

arthur li's picture

Your comment is spot on. Frequency response is important but it isn't everything. Think about tube amps. Many fantastic sounding tube amps have rolled off bass and high frequency. On the other hand, most modern solid state amps have flat response from 20 to 20000 hz. Despite this, many people prefer tube over solid state.

crenca's picture

to the "objectivist" response that FR combined with distortion pretty much is everything - it is the reproduction of the waveform (i.e. sound - the signal being sent to the HP electrically) and there is nothing "in between" the waveform, no mysterious non-waveform "timbre, texture, touch, grace" etc.?

gefski's picture

Still have one more album to listen INTO tonight.


buckchester's picture

I agree the difference must be explained through different measuring techniques. But like I said, after listening to test tones from 100hz on down, each time I lower the tone by 10hz below about 80hz, the sound becomes significantly and progressively fainter, seeming to corroborate Sonarworks measurements.

What is it exactly that makes you believe Innerfidelity's bass measurements are representative of your HE4XX headphones?

Phoniac's picture

Your comment is spot on. The sine wave test is easy to do (you can even use YouTube for that) and reveals exactly what you describe. IMHO this is a problem of how the ear/brain works. Lowest frequencies are not received with the same volume as others, while the air pressure captured by the measurment systerm stays the same - therefore you get a straight line. To really represent what humans are hearing the graph needs compensation below 100 Hz. IMHO Sonarworks not only measures, but puts a self defined compensation curve (like Harman etc) on their measurement rig's output - to better show what you hear. Tyll does the same, but forgot about low bass (bewlo 50 Hz) - he's not much interested in that, it seems.

buckchester's picture

Thanks for your comment. What you say makes a lot of sense. I will also add that after applying Sonarworks EQ, the headphones sound much better. The biggest difference is the increased bass. It sounded too bassy at first, but after some time, I cannot go back to the stock sound of these headphones.

Also, when I play the bass test tones with the EQ on, each frequency below 100hz, down to about 30hz now sounds about even.

crenca's picture

...I got crosswired, thinking the 400s measurements were the 4XX's

buckchester's picture

The 4XX has been measured here, actually. Even though it doesn't appear in the databass, Tyll did do a write up on them last year and he links to their measurements in that article. Their measurements are very similar looking to the 400i.

e_resolu's picture

Head-fier sucks !
They gave me so bad advices I wasted money because of them
It seems I have exactly the same taste than Tyll forvheaphonevsignature and i’ve Never been disappointed keep up with the good job

Huntersknoll's picture

I have a pair of HiFiman HE400s's. I actually got them after reading the review here. My first pair of hifi headphones and I love them. However I was thinking about upgrade. Would the Sundara be a good upgrade from the HE400's? I am driving them with a Schiit Stack

Chris Mini's picture

While bouncing around Amazon there were a lot of reviews that were less than favorable, mostly due to sloppy QC. Everything from one channel dead right of the box to one channel getting weaker and then nothing. When I first saw them it was love at first sight especially since I have been using the original he-400's for 6 years. Anyone had these issues? Now I am looking hard at the new Senn series 6xx