• 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 planars for music production

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,151
Location
Singapore
Did you even read my post? I even put the takeaway in bold for you. The black curve is not Harman's headphone preference target curve (that would be the blue one). The black curve is purely the response of a neutral loudspeaker (anechoically flat) in a good room measured by an ear-simulator within a HATS (Head and Torso Simulator). Etymotics have bass 5 dB below this response. If this is what you prefer, then that means you do not like accurate, neutral sound, but instead prefer bright, bass-light sound. If you also prefer this bright sound in speakers (read: flat in-room response i.e. upwardly tilting anechoic response), it's likely you have noise-induced hearing loss or presbycusis, both of which primarily affect higher frequencies so reduced bass can make these relatively more audible. If however you actually prefer accurate, neutral loudspeakers, yet you still maintain you also prefer the bass-light sound of Etymotics, the only reason I can think for that is subconscious bias from their visually flat bass / their false reputation as 'neutral' influencing your judgements.

Did you read my post where I stated that I find the ER4SR a better representation of the music I listen to? Why would I be benchmarking against speakers when I attend concerts on a regular basis (COVID permitting). Why would I be benchmark headphones against speakers rather than against real music?
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,350
Likes
1,850
Did you read my post where I stated that I find the ER4SR a better representation of the music I listen to? Why would I be benchmarking against speakers when I attend concerts on a regular basis (COVID permitting). Why would I be benchmark headphones against speakers rather than against real music?

Because unless you are comparing the sound of a concert you attended with a recording of that same concert played back on your ER4SR, your comparison is not valid. And even then you have no idea what the mixing/mastering engineers intended anyway (maybe they did not intend a 100% lifelike, concert-like sound). Here's a question that will get to the bottom of this: what are your preferred speakers / speaker frequency response when listening to your music that come closest to the ER4SR's accurate (according to you) representation of the music you've heard live?
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,151
Location
Singapore
I tend to listen with either my ER4SR or Arctis Pro most of the time. When I use speakers it is mainly Technics C700
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,350
Likes
1,850
I tend to listen with either my ER4SR or Arctis Pro most of the time. When I use speakers it is mainly Technics C700

And how would you compare the sound of all three?
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,151
Location
Singapore
The Arctis Pro has a DAC with EQ as part of the package. The default is bassy and high, I generally dial it down with the flat option. The C700is reasonably neutral and quite to my ears for a speaker, however I suspect that may be because I am comparing it with much bassier speakers as the alternatives I listen to are either home cinema set ups or my brothers system which uses old JBL L100s with bass dialed right up via EQ for his hip hop, rap and dance music.
Other headphones I own and use include the DT1990 (lots of bass plus Beyerdynamic highs), Audioquest Nighthawk (very different), Austrian Audio Hi-X55 (bit more bass than my ideal but a very nicely made product with a pretty balanced sound) and antiquated HD580.
 
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
Not really, no. Because you keep saying things like this:

Which does not make sense, because:
https://www.reddit.com/r/oratory1990/comments/dpp813/_/f5xlkoi There is no possible audible difference in transient response speed between any two headphones that can reproduce frequencies up to the audible limit of 20 kHz (i.e. any headphone that isn't broken). Whatever audiophiles think they're hearing that they describe as 'transient response' or 'speed', is actually just a feature of frequency response and/or distortion, likely a described 'fast transient response' just being good high-frequency extension and low distortion, but trying to accurately correlate the amorphous vocabulary of audiophiles with physical reality is a fool's errand.
Not really, no. Because you keep saying things like this:

Which does not make sense, because:
https://www.reddit.com/r/oratory1990/comments/dpp813/_/f5xlkoi There is no possible audible difference in transient response speed between any two headphones that can reproduce frequencies up to the audible limit of 20 kHz (i.e. any headphone that isn't broken). Whatever audiophiles think they're hearing that they describe as 'transient response' or 'speed', is actually just a feature of frequency response and/or distortion, likely a described 'fast transient response' just being good high-frequency extension and low distortion, but trying to accurately correlate the amorphous vocabulary of audiophiles with physical reality is a fool's errand.

I always thought that an impulse transient will sound different based off of the frequency response, which makes sounds like layered kick drums perceptibly different.

EDIT: I just re-read your post and you seem to be thinking way too much in terms of frequency. The time-domain and the frequency domain are related to each other. A transient will literally have a different shape because the frequency response is different. We hear this different shape. Now, is it perceptibly different? Unless every headphone has the same frequency response within the limit of human hearing, we do detect changes in the time-domain.
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,151
Location
Singapore
On the Technics SB C700 I bought it as I was offered a great package price from a dealer wanting to offload them along with the matching Technics CD player/amplifier. They had wife acceptance factor, were beautifully made and sounded good. They seem quite rare in Britain but are really very nice, a pretty well done coaxial design. They are mainly used by the wife and kids but I sometimes use them. Although they are in a 40 foot container somewhere on the high seas just now.
 

cany89

Active Member
Joined
Dec 22, 2020
Messages
251
Likes
127
@Goschie

In here the distributor of Audeze is a well-known recording studio. I asked them for setup advice and they told me they are using LCD-2 or X with Can Openers + EQ. (One extra step they do and I don't is they probably fix the Can Opener's effect on FR. I'm just a casual listener so I don't bother) Just FYI.
 

Dealux

Active Member
Joined
Aug 31, 2019
Messages
175
Likes
195
Location
Arad, Romania
Why would I be benchmark headphones against speakers rather than against real music?
That never made any sense to me. Do you only listen to live music? Well, that's not a true representation of what recorded music sounds like. Recorded music isn't mixed in a free field environment. Neutral speakers are still 100% the only true reference for tuning headphones.

I always thought that an impulse transient will sound different based off of the frequency response, which makes sounds like layered kick drums perceptibly different.

I just re-read your post and you seem to be thinking way too much in terms of frequency. The time-domain and the frequency domain are related to each other. A transient will literally have a different shape because the frequency response is different. We hear this different shape. Now, is it perceptibly different? Unless every headphone has the same frequency response within the limit of human hearing, we do detect changes in the time-domain.
It is funny that you mention sampled instruments. When you look at a waveform you're literally just seeing frequency response. The tones and overtones of an instrument are also just frequency response. Two layered kicks? Different frequency response.

The only correct definition of "transient" is the one oratory posted. I don't understand why audiophiles keep misusing that term. Also, time domain of what? Your brain perceives FR. It can't perceive the mechanical motions of a driver because it is by no means a sophisticated measuring device.
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,350
Likes
1,850
The Arctis Pro has a DAC with EQ as part of the package. The default is bassy and high, I generally dial it down with the flat option. The C700is reasonably neutral and quite to my ears for a speaker, however I suspect that may be because I am comparing it with much bassier speakers as the alternatives I listen to are either home cinema set ups or my brothers system which uses old JBL L100s with bass dialed right up via EQ for his hip hop, rap and dance music.
Other headphones I own and use include the DT1990 (lots of bass plus Beyerdynamic highs), Audioquest Nighthawk (very different), Austrian Audio Hi-X55 (bit more bass than my ideal but a very nicely made product with a pretty balanced sound) and antiquated HD580.

I don't know how else to say this, and I'm honestly not trying cause offense, but there must be something seriously wrong with either your ears (noise or age-induced hearing damage), or more likely your judgement of sound quality (probably due to subconscious cognitive biases), if you think the Hi-X55 has a 'pretty balanced sound':

Harman 2018-Austrian Audio Hi-X55.png


This increases the likelihood that your judgment of your Etymotics is as I thought heavily influenced by subconscious bias, and if presented with them blind in order to remove these biases, you would likely find them bass (and treble) deficient.
 
Last edited:
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
That never made any sense to me. Do you only listen to live music? Well, that's not a true representation of what recorded music sounds like. Recorded music isn't mixed in a free field environment. Neutral speakers are still 100% the only true reference for tuning headphones.


It is funny that you mention sampled instruments. When you look at a waveform you're literally just seeing frequency response. The tones and overtones of an instrument are also just frequency response. Two layered kicks? Different frequency response.

The only correct definition of "transient" is the one oratory posted. I don't understand why audiophiles keep misusing that term. Also, time domain of what? Your brain perceives FR. It can't perceive the mechanical motions of a driver because it is by no means a sophisticated measuring device.

Your brain perceives whatever it wants to perceive (psychoacoustics). Your ear perceives the time domain.

EDIT: Also, layered kicks in music production have higher frequency transients. Transient means sounds that last a short time. The narrower a transient, the more frequency spectrum it takes up. You can demonstrate this with a gaussian and a fourier transform.
 
  • Like
Reactions: Bop
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
What is time domain in music? I promise you this will get back to frequency response.

Do you even know what a frequency response is? Its a linear transform (the fourier transform) from the time domain to a frequency domain. You take a time domain signal (the physical thing that exists) and take the inner product of it as a function with a periodic function with a variable frequency. This lets you study the signal in its frequency components. You can transform the signal using other kinds of transforms, but what is nice about the fourier transform is it that it has an inverse.

Its pointless semantics to claim we perceive anything other than what physically exists.
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,350
Likes
1,850
I always thought that an impulse transient will sound different based off of the frequency response, which makes sounds like layered kick drums perceptibly different.

EDIT: I just re-read your post and you seem to be thinking way too much in terms of frequency. The time-domain and the frequency domain are related to each other. A transient will literally have a different shape because the frequency response is different. We hear this different shape. Now, is it perceptibly different? Unless every headphone has the same frequency response within the limit of human hearing, we do detect changes in the time-domain.

As a physicist, I'm sure you're aware that with Fourier analysis and the superposition principle any transient can be decomposed into a series of sinusoids, each with a specific frequency and amplitude. If the transducer can reproduce all of those frequencies (and no others) at the correct amplitude (which is dependent on its frequency response), then it can perfectly reproduce the transient. If there are frequencies in this decomposition above 20 kHz it cannot reproduce this does not matter, because they will be beyond the limit of audibility. If it produces spurious frequencies that are not part of the decomposed series, this is distortion. So, once again, any differences in transducers' reproduction of transients is fully determined by their frequency response and distortion.

Do you even know what a frequency response is? Its a linear transform (the fourier transform) from the time domain to a frequency domain. You take a time domain signal (the physical thing that exists) and take the inner product of it as a function with a periodic function with a variable frequency. This lets you study the signal in its frequency components. You can transform the signal using other kinds of transforms, but what is nice about the fourier transform is it that it has an inverse.

Its pointless semantics to claim we perceive anything other than what physically exists.

It's not just semantics, because the ear itself is a Fourier analysis device, so we do actually perceive sound in the frequency domain.
 
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
As a physicist, I'm sure you're aware that with Fourier analysis and the superposition principle any transient can be decomposed into a series of sinusoids, each with a specific frequency and amplitude. If the transducer can reproduce all of those frequencies (and no others) at the correct amplitude (which is dependent on its frequency response), then it can perfectly reproduce the transient. If there are frequencies in this decomposition above 20 kHz it cannot reproduce this does not matter, because they will be beyond the limit of audibility. If it produces spurious frequencies that are not part of the decomposed series, this is distortion. So, once again, any differences in transducers' reproduction of transients is fully determined by their frequency response and distortion.



It's not just semantics, because the ear itself is a Fourier analysis device, so we do actually perceive sound in the frequency domain.

Well I stand corrected on the ear part. I originally thought only our eyes could do what is a called a spatial fourier transform.

But back on the topic of transients and transducers, once again, you are just reiterating the same thing I am saying, but because you keep living in fft land, you keep thinking there is a disagreement, when there is not. I'm just stating things in terms of transients and the time-domain because I think that is what those audiophiles think they are hearing and trying to attribute as "fast", but I can safely say its the coloring of the transients. The faster the transient response is (in fft land, how flat the frequency response is) means that the transient sounds will be more neutral. So a layered kick bass which contains much more frequencies is going to sound more neutral. And I think that is all it is. They are just very neutral sounding kick drums, but probably a little less sub bass because the hifiman planar magnetic headphones need some sub-bass correction according to EQ settings people have published.
 

Dealux

Active Member
Joined
Aug 31, 2019
Messages
175
Likes
195
Location
Arad, Romania
But audiophiles use that word completely wrong. They will claim that a specific driver has more speed or better transient response because subjectively it sounds "fast" in music even though the two phenomena could be completely unrelated and you can't actually determine if your ears and brain hear the actual mechanical motions of drivers.

Then they will provide as an example the decay of a cymbal or the ability to retrieve fine reverb/room cues in music, things that are indeed better on some headphones (e.g. IEMs), but those things don't really suggest that the driver is faster in any way. The ability to hear fine details in music is also a function of frequency response. If a cymbal sounds particularly fast or clear on a specific headphone, you could isolate that sound and look at its individual components. A headphone's ability to produce the tones and overtones in music, i.e. the things that actually give you detail/information, is limited by its frequency response.

Most audiophiles will deny that there is any relationship between frequency response and detail. I believe this is mostly because we suck at reading/interpreting graphs and perhaps they're not as accurate as they should be.
 

FrantzM

Major Contributor
Forum Donor
Joined
Mar 12, 2016
Messages
4,372
Likes
7,863
Late to the party and owner of HiFiMan headphones.
I have the HE-560 v2 ( I think)
and the original HE-6 (I think it was purchased in 2014~5).

I wear eyeglasses and in both cases not an issue, seal can be imperfect. With the HiFiman headphones, General response and bass in particular, doesn't change noticeably.
The HE-6 is an endgame headphone, IMO better than anything I've heard, including the famed Stax 007. Plus it has the specs and credible (ASR among others but @solderdude too) measurements to back that up. I find it difficult to drive even with the marvelous and cheap JDS Lab Atom but that plays loud enough... Reviews from @amirm confirms my subjective impression, with his measurements and final impressions: One of the lowest ( perhaps the), THD you would find in non-DSP, actually any, headphones, extended frequency range, superlative "Spatial Qualities". When used with some EQ, close to perfect. The HE-560 is easier to drive and when EQ'd is not fas good but close enough.

The HE-560, IMO a fabulous headphones, is often on sale at Adorama for around $280, even less at times. I've seen $260. At those price, it represents the proverbial steal. I prefer it to the fabled Sennheiser HE-600 which I find tizzy and light when non-Eq and even when EQ'd , seal must be perfect for good bass. Something that can be difficult for eyeglasses wearers.
As for the HE-6 it is a rather unique headphone. Competition is few and can be counted on one hand. Objectively.
I did experience the famed " spatial qualities" of the Sennheiser HD-800 and those of the HE-6 are in my opinion similar/comparable. As for the bass of the HE-6 ,it is a galaxy better than the HD-800 or , perhaps, any other headphones.
The HE-6SE has been reviewed by Amir here, and is at top also in term of objective and subjective performance. In term of objective performance it stands with the HE-6 as one of the top 5 headphones reviewed by ASR.

Many of the qualities of the HE-6, HE-6 and HE-560 are found in the dirt cheap HE-400, which can be found often around $150. When equalized the 400 are no-brainers entry to great sounding headphones.
I have no experience with the new top fo the line offerings, Susvara, Sundara, HE-1000, etc or the stupendously expensive Shangri-La... IMO the HE-6 and HE-6 SE are endgame headphones. I am there and have no need to change to anything.

Peace.
 
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
But audiophiles use that word completely wrong. They will claim that a specific driver has more speed or better transient response because subjectively it sounds "fast" in music even though the two phenomena could be completely unrelated and you can't actually determine if your ears and brain hear the actual mechanical motions of drivers.

Then they will provide as an example the decay of a cymbal or the ability to retrieve fine reverb/room cues in music, things that are indeed better on some headphones (e.g. IEMs), but those things don't really suggest that the driver is faster in any way. The ability to hear fine details in music is also a function of frequency response. If a cymbal sounds particularly fast or clear on a specific headphone, you could isolate that sound and look at its individual components. A headphone's ability to produce the tones and overtones in music, i.e. the things that actually give you detail/information, is limited by its frequency response.

Most audiophiles will deny that there is any relationship between frequency response and detail. I believe this is mostly because we suck at reading/interpreting graphs and perhaps they're not as accurate as they should be.

From what I have read, they might just be borrowing the lingo from someone who first used the term "faster transient response" to describe a the time-domain equivalent of a very flat frequency response, and they have probably lost that direct translation in their own language. The original definition of "fast" was probably correct but used as a marketing scheme, because flat does not really have an emotional appeal.
 

Dealux

Active Member
Joined
Aug 31, 2019
Messages
175
Likes
195
Location
Arad, Romania
From what I have read, they might just be borrowing the lingo from someone who first used the term "faster transient response" to describe a the time-domain equivalent of a very flat frequency response, and they have probably lost that direct translation in their own language. The original definition of "fast" was probably correct but used as a marketing scheme, because flat does not really have an emotional appeal.
Crinacle actually wrote an article trying to explain it but in trying to do so he made it even clearer that not only is it all pure conjecture but the terms are just plain and simple wrong. Also, "time domain" is just a buzzword that doesn't mean anything.

But then he had this to say, which is honest but confusing.
Yes I know that scientifically speaking, transducers such as headphones and IEMs are (generally) minimum phase devices. Whatever exists in time domain will be reflected in the frequency domain for these transducers, so all this talk about transients and time-domain are technically completely inaccurate in a truly objective sense. However, I can’t really come up with an alternative for the phenomena that I’ve experienced over the years that I’ve always attributed to time domain stuff, so all the things I’ve talked about here are essentially placeholders terms for the time being.
So he admits it's technically wrong but he still uses technical terms badly because...word salads = good?

This is a bit like caveman thinking. We're wired to make assumptions about things we don't fully understand. Like when people used to think that thunder came from angry gods that wanted to punish humans. Or that respiratory infections were caused by cold weather because you feel cold when you have a fever. Except that the two are basically completely unrelated. You feel cold because your body expels heat at a rapid rate when it raises its temperature to kill of the virus. So clearly humans are really bad at inferring objective facts from subjective experiences.

If our ears and brains essentially perceived things completely flat like microphones we wouldn't be having this discussion. Make the interaction a bit more complex with some fairly complex resonances due to ear anatomy and people suddenly think that headphones have to have magic drivers to sound detailed. Make these people do a blind test with speakers and they will most likely pick the flattest speaker (there are plenty of studies on that). With headphones the concept of flat is more complicated because the interaction with the ear is more complicated. Complicated. Not magical.
 
OP
G

Goschie

Member
Joined
Aug 8, 2021
Messages
29
Likes
6
Crinacle actually wrote an article trying to explain it but in trying to do so he made it even clearer that not only is it all pure conjecture but the terms are just plain and simple wrong. Also, "time domain" is just a buzzword that doesn't mean anything.

A fast transient response being related to how flat the frequency response is, is not just conjecture. Its a fact that the more perfect or flat the frequency response of the transmission space is, the narrower the temporal transient response of the space is. A perfectly flat infinite frequency response would have the transient response of a dirac delta function, which can not possibly exist.

Time domain is the time varying portion of a wavefunction. It helps to say time domain when you are switching between frequency and time representations of a signal and trying to communicate it with other people.

So he admits it's technically wrong but he still uses technical terms badly because...word salads = good?

This is a bit like caveman thinking. We're wired to make assumptions about things we don't fully understand. Like when people used to think that thunder came from angry gods that wanted to punish humans.

I've read the article, and he makes no mention of fourier analysis.

But I think its important to stress that great scientific innovation always starts with observations and theories. Audiophiles are making observations and even though the "elitist" in me wants to correct people, we must respect those observations until they are scientifically tested as being false; certainly not assumed to be true, but respected. Sometimes people experience things that are real, and those things are waiting to be properly discovered through the lens of scientific rigor. A caveman made the observation of thunder, and we knew that thunder was real. Poor conclusions like thunder being the product of gods is a result of dismissing the observation rather than confronting it; because people didn't care, the poorest quality conclusion was made.
 
Top Bottom