• 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!

Dan Clark Expanse Headphone Review

Rate this headphone:

  • 1. Poor (headless panther)

    Votes: 10 2.7%
  • 2. Not terrible (postman panther)

    Votes: 12 3.3%
  • 3. Fine (happy panther)

    Votes: 65 17.7%
  • 4. Great (golfing panther)

    Votes: 281 76.4%

  • Total voters
    368

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
Is the Expanse an exception because it sounds worse than expected despite conformity above 400mHz, or better than expected despite increased bass below 200Hz? Sorry if I've missed something, I read through the last few pages and could only see some discussions about whether it is or isn't 'shouty'.
The former, presuming 400 mHz (= 0.4 Hz) is a typo ;) As I originally quoted him saying (from here):

Interestingly, it measures reasonably flat throughout the mids, but I definitely hear this one differently (it sounds much more shouty to my ear).

Faulty logic then leads him to conclude from this uncontrolled, sighted listening and the Expanse's measured closeness on the B&K 5128 in the mids / lower treble to his new, scientifically untested target that:

It makes me think there’s bound to be a high degree of variation to how it’s perceived.

Without considering that it's his untested target that may well be 'shouty', and that his impressions of the Expanse sounding shouty can already be perfectly well explained by the fact that its response peaks above the Harman target around 3 kHz by up to 3 dB over a fairly broadband range in the lower treble around the frequencies our ears are most sensitive, as shown by his own GRAS measurements no less. So looking at measurements on an industry standard GRAS system showing the deviation from the target (Harman's) that has over a decade of controlled, scientific research backing its high correlation with perceived sound quality, as opposed to a target that has none of this and measurements on the B&K 5128 that despite what some proclaim has no evidence backing its supposed increased accuracy as a whole system, the Expanse is not in fact an 'exception' at all, and his whole argument evaporates into thin air.
 
Last edited:

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,066
Likes
14,697
What I'd love for folks to try is do some EQ to the target with some of the new measurements we're putting out. Obviously we can't account for unit variation, and I don't recommend necessarily EQing to match any target past a certain point (because it's not your head/ears), but it would be good to get a read on how others perceive it.
Tell you what , you do an eq for hd600 that you think best represents Harman/gras and 5128 targets and I'll do some listening to both . Even better , do 650/6xx as well and I reckon plenty will give you feedback .Probably best in a new thread though . Will happily set up one for you.
 

Robbo99999

Master Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
6,874
Likes
6,672
Location
UK
What I'd love for folks to try is do some EQ to the target with some of the new measurements we're putting out. Obviously we can't account for unit variation, and I don't recommend necessarily EQing to match any target past a certain point (because it's not your head/ears), but it would be good to get a read on how others perceive it.
Sure, that's a good idea, start a new thread on the topic here at ASR - perhaps keep a running list that you keep updated in the original post, bullet pointing each of the headphone models you've measured so far, including links to subsequent posts where you show the averaged measurement vs your target along with your proposed EQ - that way people can quickly find the measurements & EQ for their headphone, and the extra detail of the measurement shown on your target will be interesting & useful, especially for people that might want to do their own EQ's to your target from your measurements - you could even include a config text file that describes your measurement & target so that people can import it into REW to play with it for generating their own EQ's, etc. I'd be up for trying it out on my various headphones: HD560s / K702 / HE4XX / NAD HP50 / HD600.

EDIT: drop a link to your new thread here in this one, so that we can find it. (if you create a thread for it).

EDIT #2: you could include a poll with the thread. Maybe something along the lines of, what do you prefer best: GRAS Harman EQ or B&K Resolve's target EQ. Actually you could measure the same units on your GRAS and likewise show those measurements vs the Harman Curve and provide an EQ for that. This way you're providing an EQ based on GRAS Harman and an EQ based on B&K Your Target from the same headphone unit, which is far more valid for comparison of which rig/target combination is better, because you're removing the element of unit to unit variation from your measurement - because if you were to rely on people using Oratory's EQ to compare against your B&K EQ then that's not apples to apples. That seems like a lot of work for you, but it might be useful. People could vote on which EQ they preferred between your GRAS Harman EQ & your B&K target EQ - for people with multiple headphones you could phrase the question in the poll as such:
  • Did you prefer B&K Target EQ 100% of the time vs GRAS Harman EQ (ie in all of your owned headphone models you preferred the B&K Target)?
  • Did you prefer B&K Target EQ 75% of the time vs GRAS Harman EQ (ie you preferred B&K Target in about 3 out 4 of your owned headphones)?
  • Did you prefer B&K Target EQ 50% of the time vs GRAS Harman EQ(ie you preferred B&K Target in about 2 out of 4 of your owned headphones)?
  • Did you prefer B&K Target EQ 25% of the time vs GRAS Harman EQ (ie you preferred B&K Target in about 1 out of 4 of your owned headphones)?
  • Did you prefer B&K Target EQ 0% of the time vs GRAS Harman EQ (ie you didn't prefer B&K Target in any of your headphones)
You'd try to devise the GRAS Harman EQ and B&K Resolve Target EQ to have the same output level - so level matched - this way the testers wouldn't need to do any level matching, instead they'd just keep all their volume settings the same between each EQ. I think this has the potential to be quite a powerful survey.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
What I'd love for folks to try is do some EQ to the target with some of the new measurements we're putting out. Obviously we can't account for unit variation, and I don't recommend necessarily EQing to match any target past a certain point (because it's not your head/ears), but it would be good to get a read on how others perceive it.

Do you plan to post both plots and 2 different EQ's based on 2 different fixtures with 'similar' targets and look for preference of users with the same model ?
 

markanini

Major Contributor
Joined
Feb 15, 2019
Messages
1,718
Likes
1,768
Location
Scania
Looks like an effort to use sighted listening data is being made before blind listening. Some would say that's less than optimal.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I think Resolve is just trying to get some more info besides the ears he has access to and simply trying to find out if he is in the right direction.
I don't think his goal is to create a new standard.

Truly blind headphone comparisons are impossible. You can always feel a difference anyway so would be a give away. Besides on which frequency to level match.
I don't think EQ'ing a HD201 to a similar response as the HD800 will make the HD201 sound the same as a HD800. They might have a somewhat similar-ish tonality though.
 

markanini

Major Contributor
Joined
Feb 15, 2019
Messages
1,718
Likes
1,768
Location
Scania
I don't think his goal is to create a new standard.
Certainly for reviewing purposes you don't need to. Still a five digit $ investment was made, so it's natural for questions to be raised as Resolve is initiating community discussion. With some problem solving, it's not inconceivable something like a new standard could be started as an indie effort in 2023. OTOH staying the course may be giving naysayers proof that the new rig is not all it's cracked up to be.
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,066
Likes
14,697
Truly blind headphone comparisons are impossible.
But truly blind tests of eq are very easy as long as you have someone switching for you. I think that is what he's after. Which eq does one prefer without looking at the squiggles (something I suspect a lot of is here are guilty of)
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
Looks like an effort to use sighted listening data is being made before blind listening. Some would say that's less than optimal.
Yeah it needs to be blinded. Biased sighted listening data leading to false conclusions is worse than no data. Similar protocols as Harman used in this paper should be followed as closely as possible. The EQs can be pre-applied to individual music samples (the same tracks as in that paper with known smooth wide spectra, original releases not remasters with destroyed dynamic range), loudness matched to the ITU-R BS 1770-4 standard, and instruction to set absolute playback level to an average level of 85 dB (slow, C-weighted) equivalent diffuse field level if possible (or nearest to that they feel comfortable with), estimated from the sensitivity of the headphone and the output voltage of the user's amp, which should also be specified as strictly having an output impedance below 1/10th of the nominal impedance of the headphone (so e.g. < 2.3 ohms for the Expanse) so the results are not confounded by non-flat headphone impedance and voltage division effects changing its frequency response. And obviously the source, DAC and amp having a known flat frequency response with all EQ and DSP off, with playback from Foobar2000 in WASAPI exclusive mode to avoid any Windows audio stack nonsense. The headphone should be specified to be 100% stock with no mods, including the original pads, the latter with < 1 year of use so results are not confounded by pad wear, which can significantly alter frequency response. (Obviously this is quite restrictive so in addition separate polls can be created for users with pads older than 1 year to increase participation at the expense of some potential confounding, and the results analysed accordingly). Testers should be instructed not to use analysis software to identify any differences in the files (of course some may still cheat, but there's not much you can do about that with this online testing format), and not to post any comments at all about what they hear in the poll thread. Apart from these specifications and of course the headphone model (including release year for those with revisions) each poll (for each headphone and test track) should contain no other information whatsoever about what's under test or the difference between the files in order to blind the test as much as possible and reduce cognitive biases to a minimum - no mention of Harman, GRAS, B&K 5128, headphones.com etc. Just present the two EQed files, to Harman (Oratory's EQ) and headphones.com's 5128 target, as A and B (order randomised for each poll) and have four clear options: prefer A, prefer B, equally preferred, don't hear a difference (the last option to exclude those who can't distinguish the files from confouding the results by choosing 'equally preferred'). And the poll should be posted by a 'new' member, with a neutral, nondescript name such as 'Blind Tester' so no-one can infer who they are or what's being tested from their post history and be biased by this, who posts zero comments apart from the initial test specifications so as not to in any way influence the poll or give away their identity. Obviously anyone reading this thread may know what's being tested but this would at least reduce the number of people knowing. These are all the minimum level of controls and scientific rigor that need to be implemented if this new target is to be taken seriously, which is claiming to challenge the Harman target with its years of comprehensive research behind it with provocative click-baity YouTube videos entitled Say goodbye to the Harman target - It's time to move on.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
But truly blind tests of eq are very easy as long as you have someone switching for you. I think that is what he's after. Which eq does one prefer without looking at the squiggles (something I suspect a lot of is here are guilty of)

Yes, a preference between 2 EQ's is certainly possible but may not answer the accuracy question.
Matching that to 'accuracy' (the goal of the target) is another matter as it requires comparison to a considered neutral reference and it is not clear which of both EQ's (if any) is the 'correct' one.

Chances are, when preference is the goal, the EQ with a certain bass boost for instance could be preferred over an EQ with a different bass boost.
In that case the target curve would be preference based. Remains the question of taste and used program material which can also skew preference.

it would be good to get a read on how others perceive it.

That's all Resolve is asking at this point, it can give him a clue as to what he is hearing/measuring is similar to other opinions.
You can't expect listeners to adhere to reference levels and leave a preference out of it. This is not a scientific endeavor but simply asking opinions from users. I don't think Resolve will base any target solely on this but may take results (when there are enough) along with his findings.

So I would say those that think they know how to contribute best should perhaps join the effort IF one wants this to be a success instead of trying to make the endeavor appear silly, flawed and futile.
 
Last edited:

Resolve

Active Member
Reviewer
Joined
Jan 20, 2021
Messages
212
Likes
531
I'll be starting up this process once I'm back, since I'm currently away on a trip and yeah some of these are great ideas. But I'm also not going to be posting it here. I think it's best we leave this thread to get back on track.
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,066
Likes
14,697
I'll be starting up this process once I'm back, since I'm currently away on a trip and yeah some of these are great ideas. But I'm also not going to be posting it here. I think it's best we leave this thread to get back on track.
Enjoy!
 

Robbo99999

Master Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
6,874
Likes
6,672
Location
UK
I'll be starting up this process once I'm back, since I'm currently away on a trip and yeah some of these are great ideas. But I'm also not going to be posting it here. I think it's best we leave this thread to get back on track.
lol, yeah, it is the wrong thread, but it's so darn interesting! I would endeavour to control your testing with some of the ideas presented here.

EDIT: and the DCA Stealth is kind of the king of the Harman Curve, so it does make some sense that targets were being discussed here.
 
Last edited:

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
I did not do a blind test, but it was level matched.
I asked if it was accurately level matched. You've given no details of your matching protocol.
The thing is, this much of a difference was immediately recognizable, to the point that it wasn't even close
Just like your 'night and day' impressions that headphones 'scale' differently with 'higher end' amps?
And... I wasn't the only one to have this same experience between the two.
You're not the only one to believe in 'scaling' either, looks like most of your headphones.com reviewer friends are afflicted with the same nonsense. This is what happens when sighted and expectation bias combine with groupthink, and is why you all need to do your listening tests blind, because you're evidently susceptible to these, as everyone is of course, although you and other subjectivist-leaning reviewers seem especially prone (this situation is well summed up here).
Moreover, you should know by now that it's not uniquely about level in one region of the FR
No, but the region over which our ears are most sensitive, at the nadir of the equal loudness contours, is paramount to get right, which it doesn't seem like your target does.
You can't just point at 3khz on our target without considering where the rest of the treble is.
I have considered the rest of the treble. If anything the lower-level treble of your target above the excessive 3 kHz peak may actually perceptually accentuate the latter even more due to reduced downward masking by higher frequencies. Which I see you yourself say a similar thing about the Harman 2013 (GRAS) target sounding shoutier than their 2018 target (which is a similar but actually even less in magnitude difference than between yours and Harman's 5128 targets at 3 kHz and above, see below):
I could’ve just gone with 2013 but I know a lot of us find that to be too shouty at 3khz, myself included, and I also suspect this is one of the reasons for the treble lift in 2018, since, as you increase the treble above 3khz, the glare goes away with lower harmonics not dominating as much over upper ones.

And this is the last time I'm going to be replying to you...
That won't be a huge change, considering your answers to my questions have been pretty vague and evasive.
Headphones behave differently on different heads/rigs.
We all know this is possible, to varying degrees in the general case. But here's your problem: you repeatedly use this as a crutch to justify any subjective claim you make from sighted listening (be that the Truthear Zero sounding shouty to you, which can in fact be explained by their deviation from Harman without you invoking your ear canal geometry; DCA headphones lacking 'macrocontrast'; or here the Expanse sounding shouty despite matching your 5128 target), without you giving specific evidence of this supposed high on-head variation being the primary cause in these particular cases, significant enough to dominate over other potential causes that you ignore, in this case sighted bias, and your new untested target being at fault. And there is evidence it is at fault, comparing it to Harman's 5128 target (reverse engineered from the error curves in this video, credit Sharur), which as Sean Olive showed produces a low mean predicted preference rating error of 5.8 points in relation to the predicted ratings from the original GRAS measurements and Harman target:

graph (9).png


Here we see your target peaks ~2 dB above Harman's around 3 kHz where our ears are most sensitive. Bringing it back to the Expanse, this likely explains why you find it shouty as it matches your target there. Further corroborated by your comments on what you find objectionable with the Expanse:
I find myself preferring the sound signature of the Aeon 2 Noire, which has a bit less focus towards the upper mids and a bit more focus towards the treble.
And the major difference between the Expanse and Aeon 2 Noire? That peak above Harman around 3 kHz again (which you classify as upper mids) present on the former but not the latter, as shown by good old industry standard GRAS measurements and deviation from the scientifically robustly, perceptually tested Harman target, without need for the claimed more accurate (despite lack of evidence confirming this) B&K 5128, an untested theoretical target, and invoking your assumed non-average hearing/preferences with no valid evidence for this in the form of controlled blind listening tests.

3505540f818a213bce3ff317adaad0b555f5a2f5.jpeg


Aeon 2 Noire:

A2N.png


It seems just like with most subjectivist reviewers/audiophiles, you really want your hearing and sound preferences to be 'unique' or somehow inexplicable by measurements, and you've latched on to what you fallaciously think is a catch-all objective 'cause' of this you can apply at will to explain away any subjective claim you make to objectivists, when that really won't wash with those who understand the basic scientific principles of controlled testing and providing specific valid evidence for your claims, and the more statistically probable truth you're not willing to accept is that, just like most other people, your hearing and sound preferences are likely pretty average, and it's your new untested target that may well not be representing that average.
 
Last edited:
  • Like
Reactions: jae

Resolve

Active Member
Reviewer
Joined
Jan 20, 2021
Messages
212
Likes
531
So I will not be discussing any of this further on this forum, for obvious reasons, but for those who are interested in further discussion, I will be putting together some more comprehensive information on this topic in a different place once I'm back from my trip. I welcome input from folks here as long as it's in good faith with an effort to move forward with the new standard. At the moment, there are three well-established evaluative outlets working on this, and we're all discussing with one another the best ways to develop a target curve, with a variety of interpretations of the same conceptual foundation. What we could of course use is additional input and feedback from the various audio communities to make sure the end result is a good fit, so the more of that the better.
 

Zenairis

Member
Joined
Jun 27, 2021
Messages
29
Likes
34
I think Resolve is just trying to get some more info besides the ears he has access to and simply trying to find out if he is in the right direction.
I don't think his goal is to create a new standard.

Truly blind headphone comparisons are impossible. You can always feel a difference anyway so would be a give away. Besides on which frequency to level match.
I don't think EQ'ing a HD201 to a similar response as the HD800 will make the HD201 sound the same as a HD800. They might have a somewhat similar-ish tonality though.
This, different driver designs and headphone designs change how the sound in ways that cannot be measured by the harmon curve.

It's why I tell people you should test different things and find what you like best. If you buy something solely based off of measurements that doesn't mean that it suits your preferences and if you're spending that much money it had better be something that you know you like especially on TOTL builds.
 

gallionetech

Active Member
Joined
Jan 19, 2023
Messages
138
Likes
117
Location
South Florida
Was curious and emailed Dan Clark about their recommendations for pairings to use with this headphone. They sent me a document stating some recommendations for it. Note that even though it doesn't mention the Expanse, they said it applies:
"The headphones are relatively low impedance,

-Aeon is 12.5ohms
-Stealth is 22.

-A good rule of thumb is the best amps will double their power as impedance drops from 32 to 16 ohms.
-For Stealth we recommend at least 500mW at 32ohms
- for Aeon 500mW at 16 ohms. Also its quite important for the amplifier to have low output impedance, preferably
1ohm or lower.

Below are some links to amps we like but there are many others
For Mobile Use:

DAC/Amps

Chord Electronics MOJO
https://chordelectronics.co.uk/product/mojo

Chord Electronics Hugo-2
https://chordelectronics.co.uk/product/hugo-2

DAPS

ASstell&Kern SP2000T
https://www.astellnkern.com/product/product_detail.jsp?productNo=22

Basso DX300Max

https://ibasso.com/product/dx300max/

For Desktop


SPL PHONITOR

https://www.moon-audio.com/spl-phonitor-xe-headphone-amplifier-and-dac.html

RME

https://www.rme-usa.com/adi-2-dac.html

Chord Electronics ANNI

https://chordelectronics.co.uk/product/anni

Chord Electronics Hugo TT and Mscaler combination

https://chordelectronics.co.uk/product/hugott2
https://chordelectronics.co.uk/product/hugo-mscaler
"

Very odd recommendations.. I tried the Stealths with the SP2000T at the Florida Audio Expo at Moon Audio & it had a really hard time powering them.
 

jae

Major Contributor
Joined
Dec 2, 2019
Messages
1,208
Likes
1,508
Very odd recommendations.. I tried the Stealths with the SP2000T at the Florida Audio Expo at Moon Audio & it had a really hard time powering them.
SP2000T Unbalanced:
UNBAL-THDN-Ratio-vs-Measured-Level-2.jpg

Balanced:
BAL-THDN-Ratio-vs-Measured-Level-3.jpg


We have 200 mw here with this (less unbal). That should be quite enough to play them loud enough at the end of the volume range, but maybe limited headroom at very loud volumes. Bear in mind when listening in noisy environments like an expo we will usually play much louder than normal to mask the outside noise and to also compensate for listening fatigue and the changes in responses between trying a lot of headphones. You'll maybe want more juice if you're using EQ in conjunction with features like replaygain etc. which will cut output a bit more. Harman-like curves usually sound best at higher average volumes too so there is that to consider (compared to other headphones). Remember to protect your hearing, it is easy to turn up the volume to damaging levels especially with the clean output of planars in these situations.

Most daps are way overpriced for what you get, you are mostly paying for the physical quality/prestige of an audiophile device, as an example the Fiio btr5 or similar basically has the same output at a fraction of the price, there are good usb dongles out there as well that can be used with a mobile phone. I've found an older generation android phone you can find on ebay for $100-200 (or maybe even free sitting in your drawer) + some external dongle and external battery case is infinitely more useful than most daps, if you can deal without having a nice jog wheel or robust physical buttons.

Personally, I would stay away from all chord products.
 
Top Bottom