Well, as he mentions there’s also a question of what typical loudness you listen at: if you’re a quiet listener like me, you need more boosted bass and a little extra treble for things to sound flat/natural, just because of how equal-loudness contours work at different loudness levels (he calls them Fletcher-Munson curves). People who blast themselves with decibels will prefer a flatter bass (in the raw response) and somewhat less treble, so the warmer LDC-2 style Audezes will sound closer to perfect for them than for me.
1 Like
I have gone through this whole thread, and these are the replies that stood out the most to me. They revolve around the same findings from my own thread. However I want to have some short questions, so I will proceed to repeat-quote some of the replies I quoted from above:
-
This is a question for @Polygonhell, and @abm0’s 6th post. So, you can conclude that a “neutral” headphone is not achievable without calibration to the individual’s ear (personalization). You can also conclude that the original target curve (frequency response) of the headphone will sound (be perceived by human ears) differently than the original target curve/frequency response - this is what I believe HRTF is, the frequency response that we actually perceive.
The question is, if you were to calibrate a headphone individually to ones ear, would you want a “perfectly” flat frequency response if you wanted the headphone to be neutral/reference-grade/uncolored?
The reason why I am asking this is because you brought up a good point, that speakers are tuned to be flat in an anechoic chamber, but would you also want a “neutral” headphone to be perceived with a flat frequency response? -
If not, what is the frequency response that can be classified as “flat” for headphones (in the case that it is not the same as speakers)? Now, I understand that all of the frequency curves of headphones cannot be taken into consideration because they are for mass-production headphones that are not individually calibrated to the user’s ears, but if you were to calibrate them, what frequency response would you want the user to hear?
From my understanding an anechoic chamber isn’t needed for heaphones (unless they are open-back, correct?) since you do not face reflections or stuff like that when the headphones are on the measurement rig (I am specifically talking about a measurement rig that does not have external pinnae). But it is also very clear to me that no matter how you tune the headphones on these measurement rigs, the very frequency response of the tuning will get distorted and that is not how the user will hear it. This being said, I am very interested to get the answer to the 1st question, because if a flat frequency response (HRTF) is not what is neutral, I would like to understand why - since this is what you want in a “good” speaker, or studio monitors, why wouldn’t it be something that you want to hear (not talking about manufacturer’s frequency response that they got from the measurement rig)?
No I"m suggesting any definition of Neutral is suspect, and in itself it shouldn’t be a goal.
My point on speakers is a bit more nuanced.
At some point a neutral response on speakers was defined to be flat in an anechoic chamber, but no one listens to speakers in an anechoic chamber, so because of room interactions no speaker measures flat in a room. So what was the goal of defining it that way. And most good speakers don’t measure flat in an anechoic chamber either.
All the headphone targets try to match speaker responses either in a chamber, or near field, or …, usually with some “consumer preference” mixed in.
When you actually look at speaker frequency responses the parts that tell you anything useful are, what do the crossover points look like and how jaggy is the treble, because those are what get in the way of listening.
The question you have to ask is what are you trying to achieve, then you can answer how do you measure that.
In your case you want headphones for say Mastering, (although most engineers still master with near field speakers), so what are you looking for, it’s not Neutrality, it’s a presentation that lets you project how a consumer will consume it through FM radio (less important now) off a CD through an Ipod, from a phone with crappy earphones etc.
Very few recordings are mixed for the Hi-Fi enthusiast, but if you can make it sound good there as well great.
The harder to answer parts then becomes what characteristics of a headphone let you do that, and I would argue it’s predictability (no big spikes or lulls to have to account for, no excessive frequency tilt, smooth reproduction across the frequency range). Neutrality might give you that, you could even define it to be that, but it’s just a word if you don’t define it.
If you were using them for mixing rather than Mastering, resolution might be the most critical feature.
2 Likes
As you quoted, MY definition (not everyone’s) for neutrality can be summed up by a flat PERCEIVED frequency response, or no stand out frequencies. So I would say yes, personalization is required. Whether that be through EQ or just finding the right headphone is up to the individual. Everyone’s ears are different to the point at which an individual can have two different ears (like me) so no headphone can be universal. Hell, that’s pretty much what thus hobby is all about. The hunt for the right sound.
On a side note. I kind of like what you said in your thread about measurements taken without a fake ear. To me, it makes sense to measure the headphone’s frequency response by its self, then let the people reading the graph use their knowledge of their own ears and preferences to decide if it’s what they want.
1 Like
Perfect point, speakers are not meant to be listened to in an anechoic chamber.
But then again, if you look at it as to how the consumer is listening to your track, it’s highly difficult. Some listen in the car, some on 5$ earbuds, some on 5000$ headphones, some on 1,000,000$ loudspeakers.
Wouldn’t the most unbiased way to create a track be if the source of sound (headphones) had a flat frequency response? I know this wouldn’t be ideal because no consumer listens to the track like that, but it would be the “rawest” form, wouldn’t it be?
I guess this is the same problem as tuning headphone for mass-production, you cannot know the HRTF of the user, you don’t know what frequency response/tuning he will be hearing - but would this flat calibrated/tuned headphone (according to your HRTF) achieve anything?
Not sure if you got what I was saying. It would be a lengthy process, but a possible one (and expensive too, that’s why you do not see it being done by company’s) - somebody would need to make a mold of your whole ear structure (outer, middle, inner), measure your own HRTF, and probably other factors like pressure. With the mold and this data, you could “most accurately” be able to tune the headphones to your very own and personal ears, this would be the solution of “altered original frequency response” that is caused by these factors. If you know these factors and are tuning the headphones only for yourself, you would be able to achieve any desired and wanted frequency response, and it would be how you would hear it (because you took all the subjective factors in consideration) and you took advantage of them to make the product for your ear.
To clarify, I meant I agree with you concept of not using a fake ear for measurments. The user can EQ or just look for a headphone that is close to what they think is their “neutral”. The way you do this at home is with a tone generator and you raise or lower the frequencies that stand out until it sounds about equal the whole way from 20 to 20k. It’s not perfect but it’s good enough for most. Your right though that nothing short of full custom made and tuned headphones could be a perfect signature out of the factory.
2 Likes
Yes, I got that. But I am referring to a “custom tuned” headphone that is tuned to your ears exclusively.
In this case, would you want them to have a flat frequency response?
Technically no, it would not measure flat on an earless rig once it’s been customized. If you started with a hypothetical perfect flat FR, every custom headphone would have certain frequencies raised or lowered to mach that individual’s ear related gain. It would even have frequencies that are louder in one ear compared to the other since an individual (like me) can have two different ears.
I think the closest thing there can be to a “universal neutral” is a flat FR as measured by an earless rig, but, just like the Harman curve, it will never be perfect, or even close for everyone.
The reason I think this is pretty much the fact that this is as close to “output=input” as is possible. Maybe I’m missing something important but it makes sense to me that if the transducer reproduces all frequencies at the same volume when they where all imputed at the same volume, then it would sound as close to the source as possible.
1 Like
IMO harman is an all-rounder sound signature…coming from a person who listens to EDM,KPOP,COUNTRY,ROCK,HIPHOP LOL
That’s kind of funny to me since I made this thread precisely because the harman curve was a tuning I felt failed miserably at certain genres (specifically metal and heavy synth type stuff). I find a more “Sennheiser” like tuning is a better all rounder, but kind of boring at times.
3 Likes
You misunderstood me, but it’s okay. I am not talking about the earless rig. An earless rig would be used to tune a headphone for mass-produciton. If you wanted a custom made headphone it would be tuned with a mold that exactly replicates your ears and also several other factors (data of your ears) would be used.
The headphone would be tuned according to your HRTF, and then you would be hearing the “perfectly flat” response that was present on the earless rig, but you would HAVE to calibrate it and tune so it sounds like it sounds to the microphones on the earless rig. We have ears, just watch the video from my thread, you will understand why ears play a big role.
It will sound different to everybody. It’s a complete match to some, complete miss to others. This is due to both HRTF and preference.
But if you had a headphone be tuned to a flat frequency response, people will get different variations of this flat freq response, but it would be the closest to “uncolored” sound signature.
Yes. That’s what I mean. The custom headphone will be perceived as perfectly flat but no longer measure perfectly flat to the microphone.
1 Like
Yup. But the microphone is not important at that point - because it is a custom made headphone just for your ears, the masses should never hear it. But I do get your point, it would indeed be a different freq resp on the mic rig.
You need to do this individually, this is the main reason why there are so many opinions and experiences regarding headphones (and other audio equipment too), we hear it completely differently. You ideally want the product to be custom tuned to your ears. Otherwise you have vastly different experiences from people - some love it, some hate it, for some it’s just okay.
A very complicated subject, at least business-wise.
And that’s the hobby. If it was possible to have one tuning that was objectively perfect then there would actually be a “best” headphone just like there is a “best” graphics card. But since ears are fucky, we have an entire community set up around trying and collecting things that make noise.
2 Likes
Just like this discussion, there never is a definitive answer to the GPU question.
1 Like
And headphones don’t obey Mores Law, if you went back to the days of 3DFX Voodoo 1 there was a very clear best graphics card. but it’s no longer about color depth, or filtering implementations.
2 Likes
Already answered everything you’re asking but you keep repeating the same questions. I think you’re simply refusing the notion that “flat” for headphones is not objective or universal and it’s not a graph that looks anything like a straight line. When people add EQ to mass-produced headphones to make them sound more like open-space sound from a flat-tuned speaker, they get various curves that you can see in a Griesinger lecture that MaynardGK quoted in your other thread: How to approach to review and judge a studio/reference headphone? How to tell if something really is "reference-grade"? - #15 by MaynardGK
I don’t know how many more ways I can say this. The point of “neutral” is to not sound different from reality. To achieve that, headphones have to have a response with hills and valleys, not a straight line. The hills and valleys will be different for each headphone because their drivers and cups are also doing different things to the sound, and also different for each user because their ears will be doing different things to the sound until it reaches the eardrum. What will natural FR look like on a headphone? A curve with hills and valleys, not a straight line. I don’t know how else to answer this.
That’s an equal-loudness tuning, it’s not natural sound, not even close. If you’re going to put in the work to do equal-loudness tuning, do it with the chosen headphones, do it again for good speakers in a good room, then derive the final EQ curve by subtracting the speaker equal-loudness EQ values from the headphone equal-loudness values, and you will get the tuning you have to apply to those headphones to make them sound more natural to you, more like listening to speakers. This is Griesinger’s method.
1 Like
Alright, so this changed my opinion. Neutrality is a different concept than I thought so, and it is probably impossible to incorporate a universal “neutral” for headphones - it sounds too complex and subjective to be able to apply a certain formula that “truly” is objective.
However, coloration is something that should be referencing the frequency response graph. This being said, would a headphone that was tuned to your ear (with the help of ear data, molds, and your HRTF - instead of tuning it on a measurement rig (e.g. G.R.A.S stuff)) and had a flat frequency response, would this mean that it is the most uncolored headphone?
While neutral is referring to life-like, which usually refers to a particular setting with loudspeakers, coloration should mean boosting certain frequencies. So if a headphone was calibrated and tuned just to your ears and it had a flat frequency response, could it be the fine definition of an uncolored headphone?
Without personal calibration/tuning of headphones, this would be impossible to achieve because of the HRTF - you cannot know how your headphone’s frequency response will sound to the human listener because you tuned the headphones on a measurement setup… which is not the human ear. But with the tuning/calibration, you should be able to achieve a flat freq resp and uncolored freq resp, correct?
1 Like
Different ears gang 
lol.
2 Likes