Dan Clark Expanse Headphone Review

[GaryH]

more preference score shenanigans:

https://media.discordapp.net/attachments/1012030761395826718/1012031563380306001/HD60_Preference_Score_Resolve_vs_oratory1990-4x_foolhardy_Remacri.png

Where's the 14 points difference between these two HD600 units?
If you say it's the HF resonance, then that's a bad algorithm.

The algorithm ignores frequencies above 10 kHz. What it does do is give around equal weighting to the total spectral tilt (slope of the error curve) as the standard deviation of the error curve. Pychoacoustically this makes sense as our hearing is most sensitive to lower Q deviations, and you can't get much lower Q than a spectral tilt across the entire frequency range. As Dr Floyd Tools says in his book:

The simplest deviation from flat is probably a spectral tilt. There is some evidence that we can detect slopes of about 0.1 dB/octave, which translates into a 1 dB tilt from 20 Hz to 20 kHz — not much.

It's quite obvious from looking at those two HD600 units that Resolve's pair is spectrally tilted 'anticlockwise' relative to Oratory's i.e. on average it has higher treble and lower bass. It also seems to have on average greater deviation from the target (the peak between 4.5 and 6 kHz likely the biggest contributor). Considering all this, a difference in predicted preference rating of 1.4 on a 10 point scale sounds about right to me.

 

[MayaTlab]

Yeah, the Expanse is not the first, that goes probably to the Stealth for passive headphones and maybe the N700NC (active) overall,

Most assuredly not once it's on my head. With blocked ear canal entrance mics, in active mode, vs other Harman HPs (normalised across an octave centred at 500Hz) :

Vs several arbitrarily constructed averages that collectively provide a vague estimation of what the Harman target is supposed to measure like in situ with that methodology :

Some notes on how these averages were derived : https://audiosciencereview.com/foru...ctivist-bang-for-your-buck.34596/post-1207611

If (if !) that experience is shared by others, the N700NC is a terrible pair of headphones to evaluate what the Harman target is supposed to sound / measure in situ like. Stick to large open over-ears EQed to the target, preferably several, averaged together, to get a decent estimation of it.

but the preference rating is bad and should not be relied upon at all.

If you take it with a pinch of salt it can be statistically decently valid if the alternatives you're presented with are reproduced via Harman's virtual headphones methodology. The main problems are sample variation (albeit not a problem for all headphones) and coupling of the real headphones with your head (and in my opinion active headphones with a feedback mechanism such as the N700NC are trickier than what the benefits of a feedback mechanism could theoretically suggest in that regard).

Using it to assess a single pair of headphones against a target is an unrigorous stretch in my opinion.
As an illustration of that, a pair of headphones that scored 52 (in blue) vs the averages above, and the N700NC :

 

[solderdude]

yep, a target on a specific test fixture does not guarantee the same tonal balance on ones's head.
It just kind of guarantees that this particular headphone will measure similarly on the same type of test fixture and one would hope closely to that on other test fixtures (post compensation).

So... exact EQ is rather pointless. Just look at commonalities in FR deviation between a few different test fixtures.
Work out what 'general' EQ is needed and maybe tune to taste from there.

The question will be asked why people say.. (well most say this anyway) how come there is always an improvement and I like it better with EQ.
That is because that 'exact' EQ will also include a 'close to' needed compensation anyway. It is better than nothing.

The next question that will pop up is: why do I prefer or hear an improvement when I use exact EQ over a 'general EQ'.
The reason for this is, believe it or not, that there will be an audible difference and that knowing the 'correct EQ/target' has been applied makes you feel secure that IS the correct tonal balance.
It is also the same reason why 2 different headphones, EQ'ed to the same target and look similar after measurement/calculation, still can sound different.

The expanse will not need much EQ, maybe only when a not so perfect seal can be achieved. Afterall this is a semi-open headphone.

 

[Dro]

Interesting measurements, especially considering that ANC employs active feedback to reach the target frequency response.

 

[jae]

The biggest thing I have against closed-back headphones for home use is that I often have talks or meetings during the day, or Discord calls in the evening for gaming, and I'd like to use these for all of that. But closed-back makes it so I sound weird to myself, which I am not a fan of. Speaking of which, I even sound different when I wear my SR-007 instead of my HD800S, both open-backed headphones. Kind of crazy to think about.

Is that possibly the most overkill use for TOTL headphones? Yes. But my desk is cluttered enough as-is. I have 3 pairs of 'phones within arm's reach, I'd like to reduce that to 1.

Besides, these are allegedly properly comfortable. Though I doubt more comfortable than the HD800(S)...

One solution is just to monitor your microphone's signal with some gating

There are many competitors at this price point who are making very fine headphones.

Many? As far as my opinion goes the Stealth and LCD-5 are the only ones that are remotely justifiable at ~$4k, after their prospective discounts.

 

[Deleted member 4708]

These Expanse headphones make me consider selling the Stax Sr-009BK (+ SRM 727II). The Expanse are planar, ultra low distortion and follow Harman curve so I don't have to EQ them. It would be a much simpler setup. The cheap gaming headphone looks is not doing them justice; however, the materials seem of good quality.

 

[srkbear]

One solution is just to monitor your microphone's signal with some gating


Many? As far as my opinion goes the Stealth and LCD-5 are the only ones that are remotely justifiable at ~$4k, after their prospective discounts.

Well, you’re unquestionably entitled to your preferences! I have the 2020 Utopias and I’ve previously own Susvaras—I greatly prefer my Utopias to the LCD-5, which I also auditioned. I feel strongly that my $4,000 was more than well spent.

The Meze Elites are fairly conclusively a very fine headphone—they haven’t been put to the test here, but their published tunings are not far from the Clark’s in their compliance with the Harman Target. There are some folks who love their Susvaras; although their tuning is flatter and less prominent in the low end, their detail retrieval, speed and soundstage are astonishing—I just didn’t like how much power it took to drive them so I sold them for the Utopias. Plenty of folks rave about the Abyss Phis. A select few think the RAAL SR1as are the bee’s knees. Then electrostatics are a whole different discussion.

I have no grounds to quibble with your position that the Dan Clark’s and the LCD-5’s are the only two headphones worth 4 grand. Since that price point is arbitrary, unless you can produce some evidence supporting your conclusions, I think your assessment is equally arbitrary. What criteria are you employing to justify your dogmatic assessment here?

 

[srkbear]

$4000 you get the first True Harman target hp without any EQing, plain and simple, no mumbo, no jumbo

maybe change the name to dca harman is better, simply dcah

I’m still unclear as to when achieving the Harman Target was the determining factor in headphone quality assessments. We’ve been tuning headphones for decades. Not needing EQ is nifty, but it certainly doesn’t tell anything near the whole story for me when I’m investing in a TOTL headphone.

Just for provocation’s sake, here’s Jaakkopasanen’s ranking of headphones calculated by their degree of deviance from the Harman Curve, The Stealths come in around 30th on this list. The majority of those that make the top twenty are cheap cans, like the Shure SRH440’s which cost $79—which may shed some sunshine on the fact that there is much more to the listening experience of headphones than aligning with the Harman Target.

Actually Clark’s Aeon 2 Closed rank 20 steps higher on this list compared to the Stealths—I don’t think the link has been updated yet with the Expanse. I don’t think you need golden ears to appreciate the difference in quality between the Stealths or Meze Elites and the Sundaras and Shures, irrespective of their respective tunings.

AutoEq/results/RANKING.md at master · jaakkopasanen/AutoEq

Automatic headphone equalization from frequency responses - jaakkopasanen/AutoEq

github.com

 

[spookyblack]

Think you could run these with a Schiit Magni?

 

[majingotan]

Think you could run these with a Schiit Magni?

You should have no issues whatsoever. I tried them with the 1V Apple Dongle and they sound perfectly fine and loud enough during my demo with it.No detectable clipping as well

 

[jae]

Just for provocation’s sake, here’s Jaakkopasanen’s ranking of headphones calculated by their degree of deviance from the Harman Curve, The Stealths come in around 30th on this list. The majority of those that make the top twenty are cheap cans, like the Shure SRH440’s which cost $79—which may shed some sunshine on the fact that there is much more to the listening experience of headphones than aligning with the Harman Target.

I have no grounds to quibble with your position that the Dan Clark’s and the LCD-5’s are the only two headphones worth 4 grand. Since that price point is arbitrary, unless you can produce some evidence supporting your conclusions, I think your assessment is equally arbitrary. What criteria are you employing to justify your dogmatic assessment here?

I think it is fair to say that in general headphones are designed to be sold and used as-is, i.e. without the assumption that they will be EQ'd by a customer. I would expect a top of the line headphone worthy of paying an exorbitant price like this should at least sound great (tonally accurate) out of the box, amongst other things. That is to say the least amount of problems out of the box for the average person. Since is it unreasonable for me to assume the company is tailor-making headphones to my physiology or preferences, I would assume this would be some sort of average response that is well-regarded in a large swathe of the population. So, Harman. That being said, overall adherence to the curve is less important than how and where it adheres, especially if EQ is a consideration. One has to look at it holistically. A headphone might follow the target quite close and then have uncorrectable offensive resonances and other non-linearities in the midrange or ear gain region which may cause it to sound clearly wrong to many people, even if it is flawed in some other ways like lacking bass or issues in the midrange.

Most people are not going to want to touch a headphone's response beyond 8k unless they really know what they are doing and tuning by ear, and the validity of Harman in headphones in these high 10k+ frequencies is already low, so I think it important that the high end on a headphone beyond ear gain region be as correct as possible out of the gate (if even subjectively), or at least as smooth as possible as to at least give the option for a possible broader-band treble tilt if desired. A "correct" deviation from Harman here can be viewed as a positive thing, even if it would reflect poorly on a score.

Closed headphones are already biased for higher scores because it is easier to achieve low end SPL in a sealed cup (despite a lot of them having worse treble), which is a significant factor in scoring, but one issue with this is that leakage/seal variability and consistency have the potential be all over the place in a closed headphone (head/jaw shape, spectacle wearing etc). Even the Stealth is subject to this to some degree, although there are design considerations in this product to mitigate this along with other design flaws that are generally present in other closed headphones. When EQing, If you don't like as much bass as Harman it is obviously more desirable to reduce it than to boost it. If you aren't EQing, I would say it would be better to have too little bass than too much. From a psycho-acoustic standpoint, too little bass only really affects bass appreciation while too much takes away from bass appreciation and also masks higher frequencies and destroys the overall tonality of the headphone. Stealth (and Expanse) also have deliberate deviations from Harman in certain bass regions citing psychoacoustic considerations, so they are already nerfing their own potential score by implementing those changes.

Shure SRH440 is not a bad headphone from a tonality standpoint, price and everything considered. It meets bass, mid, and even upper targets decently well, but it is still lacking energy below 50hz. It even has the similar 'psychoacoustic' bump at 100-200hz like DCA adds. But it has significant resonances, especially 6k and beyond which are gross and not correctable, and more treble fall off. Obviously, as a $75 headphone you are not getting any premium materials or any of the considerations one would be getting with a premium product, but I would say its a good product. Some things to think about: channel imbalance, sibilance, higher SPL distortion and impaired ability to accurately fix treble.

Meze Elite- FR is an uneven mess for the price, ignoring the subjective lack off bass, muddy/forward and resonances in midrange and what is going on at 1.5-2k?? While bass might be subjective and ear gain region is mostly based on physiology, Harman research (as corroborated by many studies on HRTF measurements, more importantly) shows very low variability in people up to around 2.5k in this band, so I can't imagine this sounding normal tonally to most people, I find those that deviate in this region tend to prefer marginally more energy, especially if upper bass and mids are already boosted. While the lower ranges might be mostly fixable with EQ, most people may want to correct the ~5k presence region to be closer to Harman or beyond will also have to suffer from that 8k resonance which is not correctable but will be further exaggerated, and will particularly affect consonant sounds in vocals and speech. Measurements here beyond 10k are less reliable but while it seemingly has a bit better extension it is obviously more resonant compared to the stealth, for example.

Susvara- Again lacking 1.5k-2k energy, this just doesn't sound accurate, especially for people that have earlier canal gain. Why are there so many resonances throughout the entire FR? While the audibility of these may be up for debate due to amplitude, it clearly does not demonstrate any sort of marvel in engineering when $100-200 headphones manage to not have this 'issue'. Sticking with the same brand, Is there any value proposition compared to Sundara as far as overall tonality? Adjusted for inflation at release date, these headphones cost $1000-3000 more than Stealth/Expanse do today. Why do I see so many complaints about hifiman build quality and qa on every audio forum, even for their more expensive headphones?

Utopia- FR is so similar to the well-regarded HD650/6XX throughout most of the range that one begins to question if the better materials/build quality are worth the price tag? Furthermore the HD650 has a smoother response so you could tonally match it quite close with some minimal EQ. Utopia has better lower low end distortion at higher SPL and better build quality for similar tone/SQ, some might argue its worse. It's higher sensitivity which precludes the need for more powerful amplification which is a plus. If you really love this sound and you have money then it could be a decent purchase. But as a holistic product, is it doing anything special to justify the price other than being the flagship product of a well-regarded company?

I can argue expanse is worth $4k ($3.4k with discount) because it's a planar open back that actually has bass, low distortion, EQable, strongly adheres to harman, extremely comfortable (subjective, but many people agree DCA is comfort king), great build quality using premium materials, foldable, as lightweight or lighter than is competitors, looks unique. No other product is doing as many or all of these things, let alone well. The worst thing about it objectively speaking is that it in insensitive and thus has a higher requirement for amplification than many other headphones to play loudly. There's obviously way more careful consideration into many aspects of the design compared to the other expensive headphones. They are also manufactured in the US so comparatively higher costs. By this logic every other headphone should clearly be under $3.5k.

 

[solderdude]

The SRH440 suffers from compression (higher 3rd order distortion) above 85dB SPL (in the bass) the expanse does not.

Also it is susceptible to seal breakage:

So while the FR may well be following a certain target on a specific test fixture and gets a 'preference rating' stuck to it there is technically (and IMO audibly) a lot wrong with the SRH440. Not in the least longevity of the pads and headband and elevated not very refined treble. Fine or studio monitoring (not mixing) on a budget b.t.w.

Do I find the Expanse over-priced... you bet I do and are firm in my belief that it (as well as Stealth) could easily be sold for around $ 1k but then would have to be made/sold in larger quantities, requiring a different production line/method. That investment would only pay itself back after quite a few years alone.

I reckon the biggest part of the price one pays for Stealth/Expanse is to recoup the many, many, many hours of tinkering, prototyping, testing, listening, changing something and loop back.
That has to be paid in the end and Dan is not going to be the one that does it.

 

[edahl]

The SRH440 suffers from compression (higher 3rd order distortion) above 85dB SPL (in the bass) the expanse does not.

Also it is susceptible to seal breakage:

So while the FR may well be following a certain target on a specific test fixture and gets a 'preference rating' stuck to it there is technically (and IMO audibly) a lot wrong with the SRH440. Not in the least longevity of the pads and headband and elevated not very refined treble. Fine or studio monitoring (not mixing) on a budget b.t.w.

Do I find the Expanse over-priced... you bet I do and are firm in my belief that it (as well as Stealth) could easily be sold for around $ 1k but then would have to be made/sold in larger quantities, requiring a different production line/method. That investment would only pay itself back after quite a few years alone.

I reckon the biggest part of the price one pays for Stealth/Expanse is to recoup the many, many, many hours of tinkering, prototyping, testing, listening, changing something and loop back.
That has to be paid in the end and Dan is not going to be the one that does it.

If they did that they would completely disrupt the high end market, that would be such a beautiful thing.

 

[solderdude]

It would be like offering a Focal Utopia for the price of a Clear or offering a Susvara for the price of the Arya, or the LCD5 for the price of an LCD2.
It looks like for (at least) perceived increased performance the market expects you to pay a lot more, obviously it works so why deviate ?

 

[srkbear]

I think it is fair to say that in general headphones are designed to be sold and used as-is, i.e. without the assumption that they will be EQ'd by a customer. I would expect a top of the line headphone worthy of paying an exorbitant price like this should at least sound great (tonally accurate) out of the box, amongst other things. That is to say the least amount of problems out of the box for the average person. Since is it unreasonable for me to assume the company is tailor-making headphones to my physiology or preferences, I would assume this would be some sort of average response that is well-regarded in a large swathe of the population. So, Harman. That being said, overall adherence to the curve is less important than how and where it adheres, especially if EQ is a consideration. One has to look at it holistically. A headphone might follow the target quite close and then have uncorrectable offensive resonances and other non-linearities in the midrange or ear gain region which may cause it to sound clearly wrong to many people, even if it is flawed in some other ways like lacking bass or issues in the midrange.

Most people are not going to want to touch a headphone's response beyond 8k unless they really know what they are doing and tuning by ear, and the validity of Harman in headphones in these high 10k+ frequencies is already low, so I think it important that the high end on a headphone beyond ear gain region be as correct as possible out of the gate (if even subjectively), or at least as smooth as possible as to at least give the option for a possible broader-band treble tilt if desired. A "correct" deviation from Harman here can be viewed as a positive thing, even if it would reflect poorly on a score.

Closed headphones are already biased for higher scores because it is easier to achieve low end SPL in a sealed cup (despite a lot of them having worse treble), which is a significant factor in scoring, but one issue with this is that leakage/seal variability and consistency have the potential be all over the place in a closed headphone (head/jaw shape, spectacle wearing etc). Even the Stealth is subject to this to some degree, although there are design considerations in this product to mitigate this along with other design flaws that are generally present in other closed headphones. When EQing, If you don't like as much bass as Harman it is obviously more desirable to reduce it than to boost it. If you aren't EQing, I would say it would be better to have too little bass than too much. From a psycho-acoustic standpoint, too little bass only really affects bass appreciation while too much takes away from bass appreciation and also masks higher frequencies and destroys the overall tonality of the headphone. Stealth (and Expanse) also have deliberate deviations from Harman in certain bass regions citing psychoacoustic considerations, so they are already nerfing their own potential score by implementing those changes.

Shure SRH440 is not a bad headphone from a tonality standpoint, price and everything considered. It meets bass, mid, and even upper targets decently well, but it is still lacking energy below 50hz. It even has the similar 'psychoacoustic' bump at 100-200hz like DCA adds. But it has significant resonances, especially 6k and beyond which are gross and not correctable, and more treble fall off. Obviously, as a $75 headphone you are not getting any premium materials or any of the considerations one would be getting with a premium product, but I would say its a good product. Some things to think about: channel imbalance, sibilance, higher SPL distortion and impaired ability to accurately fix treble.
View attachment 232445

Meze Elite- FR is an uneven mess for the price, ignoring the subjective lack off bass, muddy/forward and resonances in midrange and what is going on at 1.5-2k?? While bass might be subjective and ear gain region is mostly based on physiology, Harman research (as corroborated by many studies on HRTF measurements, more importantly) shows very low variability in people up to around 2.5k in this band, so I can't imagine this sounding normal tonally to most people, I find those that deviate in this region tend to prefer marginally more energy, especially if upper bass and mids are already boosted. While the lower ranges might be mostly fixable with EQ, most people may want to correct the ~5k presence region to be closer to Harman or beyond will also have to suffer from that 8k resonance which is not correctable but will be further exaggerated, and will particularly affect consonant sounds in vocals and speech. Measurements here beyond 10k are less reliable but while it seemingly has a bit better extension it is obviously more resonant compared to the stealth, for example.
View attachment 232444

Susvara- Again lacking 1.5k-2k energy, this just doesn't sound accurate, especially for people that have earlier canal gain. Why are there so many resonances throughout the entire FR? While the audibility of these may be up for debate due to amplitude, it clearly does not demonstrate any sort of marvel in engineering when $100-200 headphones manage to not have this 'issue'. Sticking with the same brand, Is there any value proposition compared to Sundara as far as overall tonality? Adjusted for inflation at release date, these headphones cost $1000-3000 more than Stealth/Expanse do today. Why do I see so many complaints about hifiman build quality and qa on every audio forum, even for their more expensive headphones?
View attachment 232451

Utopia- FR is so similar to the well-regarded HD650/6XX throughout most of the range that one begins to question if the better materials/build quality are worth the price tag? Furthermore the HD650 has a smoother response so you could tonally match it quite close with some minimal EQ. Utopia has better lower low end distortion at higher SPL and better build quality for similar tone/SQ, some might argue its worse. It's higher sensitivity which precludes the need for more powerful amplification which is a plus. If you really love this sound and you have money then it could be a decent purchase. But as a holistic product, is it doing anything special to justify the price other than being the flagship product of a well-regarded company?
View attachment 232452

I can argue expanse is worth $4k ($3.4k with discount) because it's a planar open back that actually has bass, low distortion, EQable, strongly adheres to harman, extremely comfortable (subjective, but many people agree DCA is comfort king), great build quality using premium materials, foldable, as lightweight or lighter than is competitors, looks unique. No other product is doing as many or all of these things, let alone well. The worst thing about it objectively speaking is that it in insensitive and thus has a higher requirement for amplification than many other headphones to play loudly. There's obviously way more careful consideration into many aspects of the design compared to the other expensive headphones. They are also manufactured in the US so comparatively higher costs. By this logic every other headphone should clearly be under $3.5k.

That’s a mouthful to take in, although I managed to get through it, and I can’t fault you without being a total hypocrite about the average length of my own comments on here. You still seem to be doubling down on the Harman tuning though, to nth level granularity, while underplaying all the other factors to which listeners aspire when shelling out four figures for a flagship headphone, and I’m not sure why.

There’s a whole universe online led by a group of headphone gurus who have compiled an exhaustive list of convolution files and PEQ settings to “correct” every imaginable headphone one can buy, going back about 30 years, to align with the Harman Target. Every time a new Harman Curve revision is unleashed they release a whole new set of algorithms.

Audiophiles and hobbyists alike have been utilizing these methods for years, and it certainly appears that quite a few of them are satisfied enough with the results—otherwise everyone would be unloading their existing phones in favor of a rarified few options with “compliant” tunings right out of the box. You may prioritize this criterion as a deal breaker for assigning 4 grand worthiness, but your methods are clearly not universal to all headphone enthusiasts, and we’re not all schmoes or pinheads for buying a Utopia or a Susvara and utilizing Roon’s DSP engine to fine tune them to taste. And despite your quibbling with the Meze Elite’s tuning quirks, you still haven’t offered me a convincing argument for why it’s flying off the shelves. It’s a beautifully designed, ingeniously engineered and thoughtfully executed piece of gear, clearly making a lot of people happy—and it’s unlikely that all of them are sheep.

I’m not disregarding the ingenuity of the Harman Curve. I would say that it’s the best method yet invented to provide some form of industry standard for a segment that otherwise has few standards. And we certainly can measure some objective headphone performance attributes such as distortion, dynamic range, frequency response, SNR, etc. But when it comes to tuning, I still maintain that the human ear itself is not “standardized”—an endless number of factors influence how a particular headphone will sound from person to person. Age, head shape, ear shape, depth and diameter of the ear canal and other anatomical variations, musical genre preferences and simple matters of taste are all confounds complicating the subjective experience of any unique listener—when it comes to tuning preferences.

Beyond the tuning, I don’t see how anyone can argue with assigning greatness to a headphone that outperforms others in speed, accuracy, detail, imaging, instrument placement, soundstage—not to mention fit and comfort. I can’t view anyone as on the level if they’d argue that a near-perfectly tuned headphone like the Sundara, that aside from tuning falls short in just about every other aspect, is superior to the Susvara, which dazzles across the board but needs a bit of PEQ. Is this the sort of iconoclastic polemics you’re suggesting here?

 

[DJBonoBobo]

@Dan Clark
I read a lot on other forums about how important "burn in" is in Stealth and Expanse. I always thought that was an audiophile myth. However, I now saw that this is apparently also recommended by your company in the Expanse (and/or Stealth?) manual.
Since my impression is that you usually have good technical reasons for your actions, could you please explain what is happening and why this is necessary? Is this different with planars than with dynamic headphones?
I ask because this may not be unimportant when listening to a new product for tryout.

 

[preload]

Anyone know how the bass on the Expanse (open back) compares to the Stealth (closed back)?

 

[DJBonoBobo]

Anyone know how the bass on the Expanse (open back) compares to the Stealth (closed back)?

Dan Clark Audio EXPANSE Review: Interview, Measurements, Impressions

Its amazing how you can create a $500 headphone and convince people to pay $4000 for it...

www.head-fi.org

 

[staticV3]

lol

that's some self awareness

 

[Grobbelboy]

Would a Topping DX3 Pro+ be sufficient to drive these?