• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Hifiman Edition XS - poor highs?

oleg87

Senior Member
Joined
Sep 26, 2022
Messages
374
Likes
685
Location
California
Caffeine? No drugs here. Hate to alter my hearing by not being sober.

Yes the Drop DCA Aeon X Open (planar) is one of the most realistic sounding headphones ever and is proven objectively by its low distortion performance. Only other headphones that I see that beats the X Open in terms of “realism” (subjective) is the Warwick Aperio and Hifiiman Susvara (I haven’t heard the Sennheiser HE-1 though).

Never heard a DD headphone driver sound truly realistic. Planars and some E-stats have achieved this realistic, life-like, spooky good tonality/timbre etc. that it gives me goosebumps on how crazy realistic they sound to me, as if the singers are singing in front of me, the old cliche “speakers diasappear” from being perceived, a perfect sonic illusion. HD 600/650/6XX, AKG headphones have fake sounding bass due to high distortion in the bass, Focal Clear/Utopia and HD800S can sometimes sound fake in the treble (resonance) and bass still sounds anemic compared to Planars. Measurements clearly show that lowest distortion = most realistic sounding
Low distortion performance does not "objectively" prove any such thing, with the levels/distortion profiles we're usually talking about, it's a pretty marginal component of overall sound quality. Put a vanishingly low distortion driver into a headphone design that gives you awful frequency response and it will sound... awful.
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
8,658
Likes
13,990
Always sobering to see the audibility threshold of distortion in headphones:
unknown-2.png
 

majingotan

Major Contributor
Forum Donor
Joined
Feb 13, 2018
Messages
1,561
Likes
1,861
Location
Laguna, Philippines
Low distortion performance does not "objectively" prove any such thing, with the levels/distortion profiles we're usually talking about, it's a pretty marginal component of overall sound quality. Put a vanishingly low distortion driver into a headphone design that gives you awful frequency response and it will sound... awful.

Bass distortion figures are excellent. Aeon X Open is tuned close to Harman Curve with the signature DCA mid-bass hump hence you don't need a ton of adjustments if you want the closest to Harman Target. I never EQ as I love the stock tuning. It has all that high-end sound of electrostatics on a planar driver that subjectivists call "ethereal" sound (confirmed by A/Bing my Aeon X Open with Stax SR-009, Warwick Aperio)

1679709746101.png



1679709239106.png
 
OP
P

PH14

Member
Joined
Feb 27, 2022
Messages
36
Likes
11
I requested the refund from Amazon already. I can always cancel the return should I change my mind in the next days. But while this is just one particular thing, it's one of those that you can't unhear once you're aware of it and it pops up unexpectedly everywhere. Shame really, if not for this, it would be a really nice headphone.

I'll do some more testing with left and right channel separately, thanks for the tip! If that turns out to be the same on both sides, back it goes.

Should I then completely forget about Hifiman, i.e. this is just part of their house sound? Is this particular to planars as a whole perhaps - this is my first planar?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
16,353
Likes
37,402
Location
South Holland
Should I then completely forget about Hifiman, i.e. this is just part of their house sound? Is this particular to planars as a whole perhaps - this is my first planar?

Well... you cannot trust measurements when it comes to frequencies above 6-8kHz and certainly not between 8kHz and 12kHz.
The sound aspects you (rightfully) complain about are in the 6kHz to 12kHz range, sibilance 6-8kHz and sharpness between 8kHz and12kHz.
As these are not reported 'properly' in the familiar industry standard plots you may well be lead to believe there is no peak in that part of the frequency range while in reality there is.
So what you hear is in the 6kHz -12kHz range and before dismissing the whole hifiman range (when you like the comfort and other sound qualities) I would recommend to experiment with some filtering in that part of the frequency range.
Why do I say such things.

Below a bunch of hifiman all measured on the same fixture which is not industry standard.
sundara-comparo.gif


Please note the treble range of all these different hifiman headphones and see what they all have in common in the 6-12kHz range.
Also note that quite a lot of people actually like a bit elevated treble as a positive side effect it can sound 'highly detailed' because of this.

Now... below the OLLO S5X compared to a bunch of other headphones (not hifiman) all measured on the same fixture and notice the 6-12kHz range.
You may see some well known headphones passing by that are reported to sound 'sharp'.
compare-s5x.gif

Note the OLLO S5X is a studio monitor and studio monitor headphones almost always have elevated treble anyway so one can easily pick out 'sharpness'.
This purpose is different from hifi music enjoyment.

So yes, all hifiman have 'highly detailed' sound (treble) but what is highly detailed to person A (which may have a dip in their hearing exactly at the peaking frequency) may sound 'sharp' to someone else who happens to have the 'dip' (like the one you see in some industry standard fixtures at 10kHz) at another frequency and they may be bothered by it.
At least I am.

Note I have not measured the XS (yet, maybe someday someone in Europe sends one in) but did listen to it and heard the familiar 'treble peaking' for certain.
 

GM3

Active Member
Joined
Nov 26, 2022
Messages
155
Likes
179
Well... you cannot trust measurements when it comes to frequencies above 6-8kHz and certainly not between 8kHz and 12kHz.
The sound aspects you (rightfully) complain about are in the 6kHz to 12kHz range, sibilance 6-8kHz and sharpness between 8kHz and12kHz.
As these are not reported 'properly' in the familiar industry standard plots you may well be lead to believe there is no peak in that part of the frequency range while in reality there is.
So what you hear is in the 6kHz -12kHz range and before dismissing the whole hifiman range (when you like the comfort and other sound qualities) I would recommend to experiment with some filtering in that part of the frequency range.
That would explain some things. How valuable are EQ curves from the web then? Think there was a project where people tested EQ curves, not sure if it reached a conclusion. Even if you had perfect mic, wouldn't inconsistencies/variations in 2 headphones of the same model, the shape of your head, headphone placement on the measurement rig, etc., would just mean that there's too much variation to rely on such measurement?

rtings has a consistency graph, and it looks to me like it's +/- 1.5dB <10K (ex; Hifiman 400i) for the same measuring rig & same headphone... So any FR graph is more like an average rather than a clear 'objective'/final measurement. So let's say you take the Peaks/Dips Graph, and you manage to create a perfect inverted EQ curve, how perfectly neutral would it be for your own headphones and head? It might hurt more than it helps... Best bet might be to look at different measurement sets and try to get an idea, but again, but yeah, in my experience, half measures are best, and nearly impossible to EQ for specifics...

Yeah I've overlapped the 2016 Hifiman 400i rtings + ASR FR 400i graph (both SPL &) , and there are significant differences:
Green + Red = ASR, the blue/grey lines from rtings L channel
1679754185380.png



Here's same graphs, with ASR on bottom and rtings on top:
1679754923954.png


Here's an overlay from diyaudioheaven measurements of same 400i vs ASR, ASR on bottom:
1679768507005.png

Here, the shape is completely different... Likely due to types of graphs/measurements (ASR = standardized Gras 45C) but again, to fix with EQ, if you look at 7-9 k the response is so different, hell, so completely different... Not sure what you could do with that. Even 300-600, so drastically different..!
So given your own headphones & head would be different, if you EQ'd for either graphs you'd like make things worse...

So yes, all hifiman have 'highly detailed' sound (treble) but what is highly detailed to person A (which may have a dip in their hearing exactly at the peaking frequency) may sound 'sharp' to someone else who happens to have the 'dip' (like the one you see in some industry standard fixtures at 10kHz) at another frequency and they may be bothered by it.
At least I am.
Personally, owning 400i and having somewhat the same experience as OP with harshness, kinda glad I didn't go with XS... What's weird is that you kinda rarely hear it in reviews, maybe sponsored reviews are to blame, maybe it's personal sensibility... But, what also skews things is that people might also be used to listening to gear that isn't neutral and, and have a relaxed sound...... So to them, neutral would sound harsh. And even slight bump in FR would just jump out as being 'awful', while in fact, what they're used to listen to might be the most flawed. The reverse is true too, hearing Senn 650 after being used to Hifiman, AKG, etc., they sounded poor to me; incredibly dark & unresolving.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
16,353
Likes
37,402
Location
South Holland
That would explain some things. How valuable are EQ curves from the web then?
They are valuable between 100Hz and 6kHz which is the major part of tonality.

Think there was a project where people tested EQ curves, not sure if it reached a conclusion. Even if you had perfect mic, wouldn't inconsistencies/variations in 2 headphones of the same model, the shape of your head, headphone placement on the measurement rig, etc., would just mean that there's too much variation to rely on such measurement?
The measurements should not be scrutinized under a microscope and people should not try to EQ the small wiggles they see on a single produced plot.
Instead, look for common deviations (preferably from more than 1 measurement) and just compensate the 'broad' picture.
That should be good enough. It is silly to take a single plot and create the exact opposite (and add a target) and think this is accurate. Well... it would be accurate if that was applied to that specific measurement. certainly not when using 7 or more parametric bands. Sure.. it'll make a difference and be be closer to 'ideal' than when it was not done.
People just should not believe it is now as accurate as they believe it is.

rtings has a consistency graph, and it looks to me like it's +/- 1.5dB <10K (ex; Hifiman 400i) for the same measuring rig & same headphone... So any FR graph is more like an average rather than a clear 'objective'/final measurement.
There are audible and measurable differences between copies of the same headphone and also between the same headphone in different positions on the head.

So let's say you take the Peaks/Dips Graph, and you manage to create a perfect inverted EQ curve, how perfectly neutral would it be for your own headphones and head? It might hurt more than it helps... Best bet might be to look at different measurement sets and try to get an idea, but again, but yeah, in my experience, half measures are best, and nearly impossible to EQ for specifics...

Such EQ does make a difference and is most likely closer to 'ideal' than when EQ was not applied. But yes, some things will not be EQ'ed or might even be EQ'ed incorrectly due to not detected measurement errors, seal issues or too much difference between an industry standard fixture and real ears.

Personally, owning 400i and having somewhat the same experience as OP with harshness, kinda glad I didn't go with XS... What's weird is that you kinda rarely hear it in reviews, maybe sponsored reviews are to blame, maybe it's personal sensibility... But, what also skews things is that people might also be used to listening to gear that isn't neutral and, and have a relaxed sound...... So to them, neutral would sound harsh. And even slight bump in FR would just jump out as being 'awful', while in fact, what they're used to listen to might be the most flawed. The reverse is true too, hearing Senn 650 after being used to Hifiman, AKG, etc., they sounded poor to me; incredibly dark & unresolving.
There are plenty of people that aren't bothered by the Beyer treble peak and like it just the way it is.
I know when I first bought my DT990 (about 35 years ago) I liked how it sounded. Elevated bass and treble... yeah.. bassy, excellent detail.
The exact same headphone I still own today now sounds muddy in the bass and sharp and sibilant in the treble.
So there's that too. Not all people are bothered by the typical Hifiman treble and like it because of the 'hyper detail'. Those people will never report 'sharpness' but will say it sounds crisp and detailed.

And yes, depending on where one comes from (listened to for quite a while) the brain needs some time to adjust.
So listening to an AKG or AT-ADxxx and then swithing to a HD6*0 may result in not liking it (too dark, no detail) while the other way around one will likely find AKG and AT headphones to sound bright and sharp but in the first moments might like the 'hyper details' until the sharpness and/or sibilance becomes annoying.
 

NexusOne

Member
Joined
Dec 20, 2021
Messages
63
Likes
7
I can't decide between XS, Ananda Stealth and Arya Stealth. But I think I will start with the cheapest one. It seems to be that the difference is not very big and the Arya Stealth is only a smaller upgrade. The only problem is that it feels loose on smaller or average heads. I know there is a construction for the XS. With this it is as comfortable and also for normal sized heads?
 
OP
P

PH14

Member
Joined
Feb 27, 2022
Messages
36
Likes
11
I would like to partially retract my statement that these were very comfortable for my head. Now that I have been extensively testing them for long periods at a time, I have to say I do get a hotspot on top of my head after an hour or two. The worst of it is the pads, that are rough (cloth on part that touches the skin) and very hard. It causes pain on the lower part of my jaw when worn for extensive amount of time. And that is with quite a low clamp from this headphone, so it really is just the pads that cause this.

If I were to keep it, I would try and find an aftermarket pair of pads with memory foam and leather covering on the skin part. For the headband I don't know if an option exist, but if presented the choice I most definitely prefer the older style that the Ananda and others have (the same as AKG uses and I had no issue with those years ago). Still, all this is only after a few hours, so it's quite different from e.g. the HD580 or HD6xx from Sennheiser that have hard pads that touch my ears and clamp very hard. They supposedly soften over time but mine didn't even after several hundred hours so I gave up.
 

NDRQ

Active Member
Joined
Nov 24, 2018
Messages
181
Likes
249
Unfortunately, I'm in the EU, and Drop has lousy service and return policies for EU. No representation here. Is there a non-Drop model that you would recommend?
You can pick something from here, though i dont exactly knwo what is the normal equivalent of the Drop aeon open.
Trusted eu dealer.
 

NDRQ

Active Member
Joined
Nov 24, 2018
Messages
181
Likes
249
DCA Aeon 2 Noires



I enjoy both types immensely! Planar for the lowest distortion and highest fidelity (most realistic sounding headphone driver tech ever) while dynamic for on the go listening (dynamic drivers sound more unnatural in general compared to planars due to worse distortion performance)

I tried lots of planar, but none of them were really realistic, their sound is generally kinda plasticky, with more or less unnatural timbre, depending the model.
The warmer ones are better, because the thicker sound masks very well the weirdness in the upper mids/highs.
But they are usually more airy, open, especially hifimans, i guess that why many people think that they are more "natural". But even though that more "airy" and "detailed" sound coming from the combination of the usually boosted highs and lean lows.
 

IAtaman

Major Contributor
Forum Donor
Joined
Mar 29, 2021
Messages
2,465
Likes
4,341
I most definitely prefer the older style that the Ananda and others have

Hifiman was selling Anada style headbands at some point which you can buy and replace the original headband with.
 
OP
P

PH14

Member
Joined
Feb 27, 2022
Messages
36
Likes
11
Do they still do this? None of the dealers of Hifiman I checked out when looking at their stuff before buying the XS from Amazon did. How long ago was this?

Edit: scratch that, I forgot to check the home page. Hidden among the accessories:
comfort headband 75 USD

However, it's only for HE400i/HE560 - no mention of e.g. Ananda, so I guess this won't fit the egg shaped line. It's the only headband listed there.
 
OP
P

PH14

Member
Joined
Feb 27, 2022
Messages
36
Likes
11
You posted while I was editing. Your link to their store indeed has many more options. I checked all of them, but none are for the Ananda, which is the one I suspect would work on the XS. None for the higher end ones like Arya either. No idea if they would be willing to sell you a spare part for the Ananda if you don't actually have one.
 

NDRQ

Active Member
Joined
Nov 24, 2018
Messages
181
Likes
249
I can't decide between XS, Ananda Stealth and Arya Stealth. But I think I will start with the cheapest one. It seems to be that the difference is not very big and the Arya Stealth is only a smaller upgrade. The only problem is that it feels loose on smaller or average heads. I know there is a construction for the XS. With this it is as comfortable and also for normal sized heads?
Any Arya < OG Ananda < XS < Ananda Stealth.
 

IAtaman

Major Contributor
Forum Donor
Joined
Mar 29, 2021
Messages
2,465
Likes
4,341
You posted while I was editing. Your link to their store indeed has many more options. I checked all of them, but none are for the Ananda, which is the one I suspect would work on the XS. None for the higher end ones like Arya either. No idea if they would be willing to sell you a spare part for the Ananda if you don't actually have one.
I have seen people who changed the headbands with the Ananda style ones but not sure which one they used. I suspect @Blorg might be one of those peope but not sure. Maybe you can ask Hifiman customer service. I asked them a few questions in the past and they were responsive and helpful.
 

NexusOne

Member
Joined
Dec 20, 2021
Messages
63
Likes
7
Any Arya < OG Ananda < XS < Ananda Stealth.
You mean Arya > than the others and not < ;)? The problem is that a balanced 3m cable is expensive so it destroys the good price from the XS a bit.
 
OP
P

PH14

Member
Joined
Feb 27, 2022
Messages
36
Likes
11
I have seen people who changed the headbands with the Ananda style ones but not sure which one they used. I suspect @Blorg might be one of those peope but not sure. Maybe you can ask Hifiman customer service. I asked them a few questions in the past and they were responsive and helpful.
If I decide to keep the XS after all, I might consider that. As it stands now, small chance though. Tuning sofar has not helped. Especially the sibilance is an issue, looking into some folks saying this can be diminished with after market pads. But those would arrive too late for the return date so...
 

bodhi

Major Contributor
Forum Donor
Joined
Nov 11, 2022
Messages
1,054
Likes
1,541
If I decide to keep the XS after all, I might consider that. As it stands now, small chance though. Tuning sofar has not helped. Especially the sibilance is an issue, looking into some folks saying this can be diminished with after market pads. But those would arrive too late for the return date so...

Just return it and get something else. It's not like the XS is some endgame headphone anyway so why try to fix it's flaws?

The XS is one of those hyped headphones that has quite a fanbase. This kind of following is often partly result of self reinforcement bias by buyers who first went trough countless reviews before spending big bucks (for them) and want to keep on believing they made the right choice. When you have the headphones you just have to forget all the reviews and opinions and focus on if they really are worth it to YOU.

If you can fix the problems with EQ then that's good, most headphones need some anyway. But if you can't and there are other things like Hifiman QC and lacking comfort, and you can get your money back then just get rid of it and try again.
 
Top Bottom