Dan Clark Stealth Review (State of the Art Headphone)

[Robbo99999]

Well if that happend maybe because the science itself is not complete? As you can see the harman target itself keeps evolving. Which one is correct? 2013, 2015, 2017? Not sure what is the latest revision or what it looks like anymore...

About half a year ago I did some EqualiserAPO config files that will turn a 2018 Harman EQ into a 2013 Harman EQ, and likewise into a 2015 Harman EQ. I'll attach the config files to the end of this post. I do prefer the 2018 Harman EQ to the others. If you use the files below you'll have to have already EQ'd your headphone to the Harman 2018 Curve before you can convert it using one of the attached config files. Also note that these config files contain Shelf Filters, and when you import config files that include Shelf Filters into EqualiserAPO then those Shelf Filters don't get activated properly - you'll probably need to manually input the Shelf Filters into EqualiserAPO - you'll be able to tell if the Shelf Filters imported wrongly because they'll be a different typeface & format in the EqualiserAPO window - in which case just input those Shelf Filters manually and delete the Shelf Filters with strange typeface.......bizarre I know!!
EDIT: I also attached one that converts a 2018 Harman to one that combines 2013 Bass with 2018 treble - apparently that's Oratory's personal target curve of choice for his own music listening - he doesn't publish that curve of course though, all his EQ's are done to the 2018 Harman Curve.

 

[PeteL]

I also get what you're you're saying about distortion, however as someone who never goes above 80 dB (typically lower), I'm guessing this won't be such a big problem?

According to Fletcher Munson, If you are listening to a wide band signal at 80 dB SPL A-weighted, or what sounds like a 80 phon volume, You are developing 110 dB SPL in the sub Bass region so that's the distortion in the Bass at 110 dB SPL that you should account for. Solderdude point was that if you run into distortion in the Bass, the rest of the spectrum will get dynamic compression.

 

[Oceanvibe]

Thank you.

I checked their internet site, it states:
- 'vegan suede and "leather" ear pads'
- 'Composite synthetic suede and protein-leather pads'
- 'Ear Pads: Synthetic Suede and Leather'.

The second statement is clearer but the first one (and also the third one) is unclear, misleading. I interpret it as meaning that the ear pads are made from fake, vegan, artificial, smooth protein material looking like leather ( = marketing talk to hide that they are made from fake leather). The second statement makes it luckily clear enough that the pads are made with fake leather.

My experience with vegan fake "leather" ear pads is that they never last more than 6 months before entirely falling apart, unlike ear pads made from genuine leather which I have never seen falling apart, even after many years.
( normally fake leather pads do not fall under warranty: there is no assembly fault, but these are parts submitted to wearing off through use).

Sorry, but I am not wasting $ 4000 on such an expensive device which will quickly fall apart.
I can understand that manufacturers use vegan leather on bluetooth headphones, having to save every dollar to keep the price under a few hundreds dollars.
But to save these few bucks to maximise profit on $ 4000 headphones is something not acceptable for me.

My partner is vegan and doesn’t like to buy leather shoes. A couple of years ago I bought her some vegan doc martens - they actually wear incredibly well. They’re also more expensive than the leather versions! I personally hate to see leather used where a more environmentally friendly and more ethical material could be used and don’t understand why headphone manufacturers still use it - it seems outdated to me. Of course, the pleather on these is likely loads with plastics so also not environmentally friendly.

 

[solderdude]

Could this just be down to unit variance?

COP

COP-Pro

They differ the most in the 1 setting but yes, there are variables. Pad condition/seal may be one of them

After modding 2 of them they measured the same though.

 

[PeteL]

Thank you.

I checked their internet site, it states:
- 'vegan suede and "leather" ear pads'
- 'Composite synthetic suede and protein-leather pads'
- 'Ear Pads: Synthetic Suede and Leather'.

The second statement is clearer but the first one (and also the third one) is unclear, misleading. I interpret it as meaning that the ear pads are made from fake, vegan, artificial, smooth protein material looking like leather ( = marketing talk to hide that they are made from fake leather). The second statement makes it luckily clear enough that the pads are made with fake leather.

My experience with vegan fake "leather" ear pads is that they never last more than 6 months before entirely falling apart, unlike ear pads made from genuine leather which I have never seen falling apart, even after many years.
( normally fake leather pads do not fall under warranty: there is no assembly fault, but these are parts submitted to wearing off through use).

Sorry, but I am not wasting $ 4000 on such an expensive device which will quickly fall apart.
I can understand that manufacturers use vegan leather on bluetooth headphones, having to save every dollar to keep the price under a few hundreds dollars.
But to save these few bucks to maximise profit on $ 4000 headphones is something not acceptable for me.

The protein-Vegan Leather pads on My Fostex TH-X00 are 4 or 5 years old and looks like new. No sign of wear. What is your experience with Vegan Leather that don't last 6 month? Which Model?

 

[Dro]

According to Fletcher Munson, If you are listening to a wide band signal at 80 dB SPL A-weighted, or what sounds like a 80 phon volume, You are developing 110 dB SPL in the sub Bass region so that's the distortion in the Bass at 110 dB SPL that you should account for. Solderdude point was that if you run into distortion in the Bass, the rest of the spectrum will get dynamic compression.

You won't get 110 dB SPL in bass with an 80 dBA average. A punchy bass drum has an RMS 6-10 dB higher than the average RMS of the loud part of a reasonably dynamic rock track. Generally I would say the 94 dB SPL distortion measurements are the only relevant ones for you with the Stealth if you listen at 80-85 dBA.

 

[Jave]

@amirm - Do you plan review the Audeze CRBN and compare with these?
From Jude's measurements they seem to perform about the same and are around the same price.

Dan Clark Audio Stealth Review, Interview, Measurements - YouTube
Audeze CRBN Review, Interview, Measurements - YouTube

 

[PeteL]

You won't get 110 dB SPL in bass with an 80 dBA average. A punchy bass drum has an RMS 6-10 dB higher than the average RMS of the loud part of a reasonably dynamic rock track. Generally I would say the 94 dB SPL distortion measurements are the only relevant ones for you with the Stealth if you listen at 80-85 dBA.

I didn't talk about rms level, nor average, I talked about A Weighting. There is no direct relations. Nothing says that the RMS level of a recording correspond to the 1kHz average amplitude, perceived loudness does. in fact it's lower than that, recordings contain more energy in the 100-400Hz, but it don't matter. In fact, correct me if I am wrong but I have never seen RMS levels expressed in dBA. It's a time domain calculation, it doesn't know anything about frequency or perceived loudness. The LUFS unit does, but it is still not a direct relation between the Loft normalization and equal contour curves.

 

[woof!]

About half a year ago I did some EqualiserAPO config files that will turn a 2018 Harman EQ into a 2013 Harman EQ, and likewise into a 2015 Harman EQ. I'll attach the config files to the end of this post. I do prefer the 2018 Harman EQ to the others. If you use the files below you'll have to have already EQ'd your headphone to the Harman 2018 Curve before you can convert it using one of the attached config files. Also note that these config files contain Shelf Filters, and when you import config files that include Shelf Filters into EqualiserAPO then those Shelf Filters don't get activated properly - you'll probably need to manually input the Shelf Filters into EqualiserAPO - you'll be able to tell if the Shelf Filters imported wrongly because they'll be a different typeface & format in the EqualiserAPO window - in which case just input those Shelf Filters manually and delete the Shelf Filters with strange typeface.......bizarre I know!!
EDIT: I also attached one that converts a 2018 Harman to one that combines 2013 Bass with 2018 treble - apparently that's Oratory's personal target curve of choice for his own music listening - he doesn't publish that curve of course though, all his EQ's are done to the 2018 Harman Curve.
View attachment 149065

Thanks for the info! I was actually looking something like this. I actually bought HE6SE v2 because I wanted to learn more about eq.

 

[Hifihedgehog]

Jude

Also Jude

From Jude's measurements they seem to perform about the same and are around the same price.

If you are choosing to use Jude's measurements, even there, they are not even close rivals. The Audeze CRBN lacks a significant amount of bass. The CRBN also has the Audeze characteristic dip at around 2 KHz, meaning a suppressed or muffled upper midrange. It is far from neutral, especially in the deepest bass region. The Dan Clark Audio Stealth is far and away better, no contest.

 

[Robbo99999]

Thanks for the info! I was actually looking something like this. I actually bought HE6SE v2 because I wanted to learn more about eq.

Nice! I guess that's different the HE6se then which was reviewed here:
https://www.audiosciencereview.com/forum/index.php?threads/hifiman-he6se-review-headphone.23576/
https://www.dropbox.com/s/638c1dgypzw7eit/Hifiman HE6se.pdf?dl=0
Certainly the HE6se is a fantastic headphone, but I don't know about the v2, haven't seen any measurements for it.

 

[Dro]

I didn't talk about rms level, nor average, I talked about A Weighting. There is no direct relations. Nothing says that the RMS level of a recording correspond to the 1kHz average amplitude, perceived loudness does. in fact it's lower than that, recordings contain more energy in the 100-400Hz, but it don't matter.

The original question was about bass distortion when listening at 80 dBA. You will not encounter 110 dB SPL in bass when listening at 80 dBA. Maybe ~95 dB SPL when you listen with a Harman curve target for your headphones.

 

[PeteL]

The original question was about bass distortion when listening at 80 dBA. You will not encounter 110 dB SPL in bass when listening at 80 dBA. Maybe ~95 dB SPL when you listen with a Harman curve target for your headphones.

It's an interesting theory, but I'd like to know how you arrive to this conclusion, your RMS, average, vs bass notes don't explain that. I can word it differently, if you play a 20 HZ tone, and your SPL meter show 80 dBA, which is not very loud, the measured unweighted SPL level, which is also how we measure distortion. is at 110 dB SPL. Now how much content in the subs depend of the song, of course, but it says nothing about this that kick drums are 6-7 dB louder than the RMS level. For what you claim to happen, you would need the bass region would be less loud than the rest of the music to be only a few dB of more amplitude. But yes some acoustic music don't have subs so it wouldn't matter for that.

 

[SeriousSam70]

Also Jude

View attachment 149091




If you are choosing to use Jude's measurements, even there, they are not even close rivals. The Audeze CRBN lacks a significant amount of bass. The CRBN also has the Audeze characteristic dip at around 2 KHz, meaning a suppressed or muffled upper midrange. It is far from neutral, especially in the deepest bass region. The Dan Clark Audio Stealth is far and away better, no contest.

View attachment 149093

View attachment 149094

Let's not compare apples to oranges.
Audeze CRBN has an exemplary frequency response in the bass area (very linear and very well extended to the lowest sub-bass area, almost a straight line with a very slight downward slope), especially considering that it is an OPEN-BACK type headset and has an electrostatic driver.
You can't compare that to the augmented sub-bass and bass typical of closed-back headphones, as the DCA Stealth frequency response chart shows.
I also could say that it has a fairly well-extended midrange FR, at least for an Audeze headset.
Below I combined the two FR graphs into one (intersect at 1 kHz), so you can better see the differences:

 

[gvl]

Given its basically a custom tuned product, how much of unit to unit variation should be expected?

 

[Dro]

It's an interesting theory, but I'd like to know how you arrive to this conclusion, your RMS, average, vs bass notes don't explain that. I can word it differently, if you play a 20 HZ tone, and your SPL meter show 80 dBA, which is not very loud, the measured unweighted SPL level, which is also how we measure distortion. is at 110 dB SPL. Now how much content in the subs depend of the song, of course, but it says nothing about this that kick drums are 6-7 dB louder than the RMS level. For what you claim to happen, you would need the bass region would be less loud than the rest of the music to be only a few dB of more amplitude. But yes some acoustic music don't have subs so it wouldn't matter for that.

How I get there:

• Take the loudest section of a rock track (modern mastering, but not exactly a loudness war victim) with a punchy bass drum
• Measure its RMS (vs a 0 dBFS sine wave, but that is standard). Result is -8 dB.
• Apply EQ that is roughly equivalent to A weighting and measure the RMS again. Result is -11 dB, 3 dB lower.
• Under the assumption that this section of the track is played back at 80 dBA (weighted), I can deduct that 83 dB SPL (unweighted) corresponds to -8 dB RMS in the audio file.
• Now I look at the spectrum and see that bass frequencies max out at about -2 dB during a bass drum transient, 6 dB higher than the -8 dB unweighted for the longer section.
• Based on this, I assume 89 dB SPL would be the max resulting SPL in bass frequencies when played back with a perfectly linear (in sound pressure, not perception) headphone. With the Stealth, add 6 dB for 95 dB SPL due to its frequency response.

I think where your method goes wrong is that you assume we want to listen to a pure 50 Hz note at 80 dBA, which indeed would require 110 dB SPL. but we listen to wideband music at 80 dB, and the SPL for the whole is larger than its individual components. My -8 dB RMS section has no single frequency above -20 dB RMS, but the sum comes out to -8 dB.

 

[bidn]

The protein-Vegan Leather pads on My Fostex TH-X00 are 4 or 5 years old and looks like new. No sign of wear. What is your experience with Vegan Leather that don't last 6 month? Which Model?

Oppo PM-2, PM-3,
All Bluetooth headphones with fake leather.

 

[Dro]

@Dan Clark do you mind sharing the frequency response of the Stealth compared to the Ether 2? It would be nice to compare the amount bass to a known quantity.

 

[amirm]

@amirm - Do you plan review the Audeze CRBN and compare with these?

I have asked for sample but availability is tough. As to response company told me it doesn't have Harman bass response so can't be like Stealth.

 

[Dan Clark]

Due to people trying to game the Club DCA system we have changed the policy so that your first discount is available 30 days after the first purchase.