Where is the science in EQ of headphones?

[preload]

I don't even know what this thread about anymore.

 

[GaryH]

Who said anything about listening that loud all day?

BTW: I doubt that many people listen to live levels for 6-8 hours. That volume leads to fatigue very quickly. Even musicians wear plugs during their work. Ever heard a Piccolo's higher registers next to you? Not something you want to experience for hours on end.

Please read the sources I post. The sub-bass stimulus investigated in the paper I linked was 90 seconds (bit shorter than 6 hours...).

How can a sensory cell be "stimulated", to the point of taking damage and yet not be stimulated enough to transmit the signal of the stimulus to the brain?

Any takers on that one?

As I understand it there's a threshold of stimulation that will result in a signal being transmitted to the brain, but repeated stimulation below this threshold can still be damaging to the hair cell as it is cumulative. As the resonant frequency of the cochlea decreases from the base to the apex, lower frequencies will resonate the hair cells at further positions from the base, but will still vibrate hair cells on the way (short of the higher intensity resonance elicits). This is one of the reasons age-related hearing loss is primarily at high frequencies. As the audiologist I linked to said:

If you look into the biology of it and how we hear sounds, some might argue that low freq sounds are more damaging than high. If you look at the cochlea, you have hair cells which convert the physical sound waves into electrical impulses which get sent to the brain. Sounds enter at the base and depending on what freq the sound is, will be picked up at the right point. You can see that lower freq sounds have to pass allllll the way through to reach the right point. Although the other hair cells are not 'hearing', they're still being stimulated. This is normally why age related hearing loss is normally a high frequency loss.

Here are some other sources you may be interested in:
https://academic.oup.com/icb/article/58/2/282/5039866
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6556399/

I suggest directing any more questions on this to an actual professional in the field like that audiologist, rather than relying on audiophiles/engineers in no way qualified to give advice on the risks of sound-induced hearing loss.

 

[GaryH]

You were responding to:

“Besides that, passive headphones still have plenty of confounding factors. Bass distortion or low power handling can easily prevent you from using the full Harman bass shelf.”

In other words, distorting AFTER EQ.

No, please do not truncate the quote I was responding to, it's misleading. And please do not use all-caps, as online they represent shouting at someone, which is bound to provoke. I was responding to this (full quote):

Besides that, passive headphones still have plenty of confounding factors. Bass distortion or low power handling can easily prevent you from using the full Harman bass shelf. Cancellation dips can cause unfixable nulls. It's also not uncommon for a headphone's treble to be full of narrow-ish Q peaks that can't be fixed without without killing all the treble because they shift too much with the position of the headphone on your head.

And I made clear that bass distortion may not be a problem without EQ if it is low enough:

apart from possibly bass distortion depending on its level

But the other issues mentioned above will be, even without EQ. I hope this has cleared things up.

 

[JustAnandaDourEyedDude]

How can a sensory cell be "stimulated", to the point of taking damage and yet not be stimulated enough to transmit the signal of the stimulus to the brain?

Any takers on that one?

Ultraviolet light (into the eye)?

 

[solderdude]

How can a sensory cell be "stimulated", to the point of taking damage and yet not be stimulated enough to transmit the signal of the stimulus to the brain?

Any takers on that one?

Looking into a fiber optic cable that has considerable power between 1310nm and 1625nm can burn out your retina without you realizing it before it is too late. Hence the warning to never look into a fiber optic cable.

With the hearing ultrasonic sounds at high level also may not be good for the hearing in general.

 

[JustAnandaDourEyedDude]

As I understand it there's a threshold of stimulation that will result in a signal being transmitted to the brain, but repeated stimulation below this threshold can still be damaging to the hair cell as it is cumulative. As the resonant frequency of the cochlea decreases from the base to the apex, lower frequencies will resonate the hair cells at further positions from the base, but will still vibrate hair cells on the way (short of the higher intensity resonance elicits). This is one of the reasons age-related hearing loss is primarily at high frequencies. As the audiologist I linked to said:

I wonder what it is that prevents the high-frequency waves from traveling all the way to the apex and damaging the low-frequency hairs and receptors there (and others on the way)? Attenuation by dissipation?

 

[maverickronin]

Please read the sources I post. The sub-bass stimulus investigated in the paper I linked was 90 seconds (bit shorter than 6 hours...).

The idea is prima facie implausible. It very unlikely that such an effect could exist without it being common knowledge and experience by now. How many sources incredibly loud and completely unblockable sources of 50/60Hz hum are there in the world now?

Also, the study is only collecting data on a possibly proxy for hearing damage rather than any actual change in the subjects' hearing.

Rather than just providing links without comment about what you find relevant, do you have any sources that indicate high SPLs at low frequencies can cause permanent threshold shifts? Because I've never seen any such data.

 

[GaryH]

I wonder what it is that prevents the high-frequency waves from traveling all the way to the apex and damaging the low-frequency hairs and receptors there (and others on the way)? Attenuation by dissipation?

Yes, the basilar membrane of the cochlea is stiffer at the base than the apex. Stiffer means higher resonant frequency, so high frequencies will resonate near the base, which will dissapate the sound energy there before it can reach further down towards the apex:

 

[GaryH]

The idea is prima facie implausible. It very unlikely that such an effect could exist without it being common knowledge and experience by now. How many sources incredibly loud and completely unblockable sources of 50/60Hz hum are there in the world now?

Also, the study is only collecting data on a possibly proxy for hearing damage rather than any actual change in the subjects' hearing.

Rather than just providing links without comment about what you find relevant, do you have any sources that indicate high SPLs at low frequencies can cause permanent threshold shifts? Because I've never seen any such data.

I'll repeat (one last time...) the salient comments from an audiologist I posted earlier:

To put it bluntly, any frequency at a loud enough intensity can damage your ears.

If you look into the biology of it and how we hear sounds, some might argue that low freq sounds are more damaging than high.

You can see that lower freq sounds have to pass allllll the way through to reach the right point. Although the other hair cells are not 'hearing', they're still being stimulated. This is normally why age related hearing loss is normally a high frequency loss.

So, they're saying the main reason age related hearing loss is primarily in the high frequencies (which is a widely observed 'permanent threshold shift') is the fact that lower frequencies pass all the way through the entire cochlea before being dissapated, so have a higher chance of damaging more hair cells, and the high-frequency region of hair cells get stimulated and progressively damaged over time more than others. If you know better than a qualified audiologist, please explain what exactly is wrong with their explanation with its firm grounding in auditory physiology (as depicted in the above video).

 

[maverickronin]

I'll repeat (one last time...) the salient comments from an audiologist I posted earlier:

So, they're saying the main reasons age related hearing loss is primarily in the high frequencies (which is a widely observed 'permanent threshold shift') is the fact that lower frequencies pass all the way through the entire cochlea before being dissapated, and so the high-frequency region of hair cells get stimulated and progressively damaged over time more than others. If you know better than a qualified audiologist, please explain what exactly is wrong with their explanation with its firm grounding in auditory physiology (as depicted in the above video).

An anonymous audiologist on reddit?

I'll try asking one more time and as specifically as possible.

What SPLs at what frequencies and durations are you claiming as a risk for/cause of noise induced hearing loss and what data or authority are you basing that claim on?

 

[jae]

Thanks for the input.

I’m 56.

I wonder what the age distribution curve is here.

I do find it strange that people not getting enough treble would compensate by turning down the bass, rather than turning up the treble.

Hearing loss is an extremely complex problem, full of non-linearities and counter-intuitive quirks that vary between person to person based on genetics, severity, and nature of the hearing loss. Especially true of sensioneural hearing loss, losses at certain frequencies can cause the perception of higher or lower sensitivity at others and also variable levels/types of noise and distortion at the location of the loss and even other related/unrelated frequencies on the spectrum.

It can also have a large, non-linear impacts on the harmonic, temporal, and spatial resolution of sound. Some of these characteristics are especially significant in what makes headphones sound good or natural, such as what the harman curve tries to accomplish as far as FR, but this research is only applicable to those in normal hearing ranges and still has many limitations.

The brain can compensate for hearing loss in some ways, and while it is good in some aspects it is not great at doing so in others. In a real world scenario, many with hearing loss simply find lowering the bass (i.e. flattening the bass to be closer to their upper range loss) and perhaps a volume increase across the board will result in a clearer, more natural sound than simply fixing their points of "deficiency" as it is shown in an audiogram of a typical presbyacusis case. Even though it may look similar, an audiogram used to determine hearing loss is not like the frequency response of an audio component and it is more like of a measure of dynamic range at limited frequencies that does not factor in any other measure of sound like distortion or how sound is processed and perceived by the brain.

 

[Aerith Gainsborough]

Looking into a fiber optic cable that has considerable power between 1310nm and 1625nm can burn out your retina without you realizing it before it is too late. Hence the warning to never look into a fiber optic cable.

That's only because of the thermal component of the laser energy. Ever held your hand into a strong laser beam?
I would not recommend it. I worked with Excimers and we used it to cut ceramics.

This does not apply to acoustic energy, most certainly not the weak levels we are talking about with speakers/headphones. So I am still interested in the actual mechanism that lets bass energy of around 115dB (which is what you have in most movies) damage our ears.

 

[solderdude]

Of course but you stated 'sensory cell' which the eye has plenty off and was an open door that begged to be kicked in .
Much less powerful lasers are used in fiber optics which are invisible but can be a few Watt concentrated on a small area.
The folks that sometimes stare into the low power 820/850nm MM transmitters (in the dark) to see if they are 'on' might ruin their vision (in one eye) staring into a seemingly not functioning transmitter that is in fact a few Watt SM laser output that is supposed to couple in a fiber.
We have meters or fluorescent cards to check for optical power.

115dB SPL in music at lowest octave bass frequencies won't damage your hearing when listened to now and then.
Hearing gets damaged anyway over time.
I never met someone old that can hear as good as a young person (at least not without a cranked up hearing aid).
I bet those older folks didn't listen much to very loud music when they were young.

 

[Aerith Gainsborough]

The idea is prima facie implausible. It very unlikely that such an effect could exist without it being common knowledge and experience by now. How many sources incredibly loud and completely unblockable sources of 50/60Hz hum are there in the world now?

Also, the study is only collecting data on a possibly proxy for hearing damage rather than any actual change in the subjects' hearing.

Rather than just providing links without comment about what you find relevant, do you have any sources that indicate high SPLs at low frequencies can cause permanent threshold shifts? Because I've never seen any such data.

He did provide the link.
The study assesses sounds made by the human ear (don't ask me how, first time I've heard of that one) and how they change when the ear is presented with a low frequency stimulus.

Figure 1. Left column: representative pre- and post LF sound exposure examples of pre-existing SOAE level and frequency changes as a function of time from four different subjects. The grey bar indicates the presentation of the LF stimulus (30 Hz, 80 dB(A), 90 s). The timescale is centred around the LF sound offset at 0 s. Right column: same as in the left column, but for new SOAEs which only appeared for a short period after LF exposure.

The sound level of the intense 30 Hz stimulus was 120 dB SPL, corresponding to an A-weighted level of 80 dB. The accumulated daily LF sound exposure was monitored and controlled to be well within the daily exposure limit for normal working conditions in Germany.

So first of all: most music does not incorporate 90s long 120dB peaks of 30Hz. This study is NOT directed at music at all but it is directed at the safety regulations for people that work in noisy environments. They say that due to the fact that the sounds change and that recovery takes longer than the stimulus, the LF sounds are affecting the ear, presumably in a negative way, even if the volume is not perceived as uncomfortable. In that respect, they may be on to something... or not. More research is needed.

Here is the original link: https://royalsocietypublishing.org/doi/10.1098/rsos.140166

 

[solderdude]

Why don’t headphone makers provide various EQ scenarios for their products? Just curious…

The only one that does is Audeze (afaik) and it is sorely needed (reveal plug-in) as well for many of their models.

Once someone creates EQ adjustment list isn’t there an objective way to measure the suggested SQ improvements?

Yes, but there is a snag. Not all measurement fixtures are the same so when you base EQ on test fixture A it will not be the same on fixture B, C to Z.
One can use an industry standard one (so others using the same test fixture get comparative similar results) but that doesn't mean the 'measured response acc. to a specific and described test fixture' is similar to your ears. Especially above 8kHz things are far from accurate.
That's also where your ear differs the most and often differently than some average standard.

This is why when one EQ's various headphones to a standard it measures great on that rig but you can still hear differences tonally. The differences are much smaller than without EQ so there is improvement.

Then there is taste (sound one is used to) and recording quality which often require tone control (not EQ in general) which complicates matters.

So there certainly is science involved in EQ but there is one less sci-ency aspect as well. The listener, the seal, preference, way of wearing it etc.

 

[restorer-john]

There’s nothing scientific about correcting to a preference curve which you don’t like.

The most scientific thing to do would be to correct to what the Harman research had before they asked people what they thought and gave them treble and bass controls to play with.

The closest you’ll get to that is pretty much Harman, but flat from 200 down to 20, instead of bass rising.

And that might not be your preference either, but it’s a reasonable starting point.

The Harman target is just boom/tizz to my ears. Not unlike the smiley face graphic EQ settings from the 70s. Hardly an accurate or sophisticated preference in my experience. Might sell to the "beats" crowd or the EDM aficionados.

But each to their own. I take no notice of the so-called Harman preference curve.

Harman could have given their test subjects a single midrange control and achieved essentially the same curve.

 

[GaryH]

An anonymous audiologist

So what? The brilliant NWAVGuy was anonymous. Paul McGowan isn't. Should we trust the latter more just because he's put his name to his BS? You are free to believe who you want, but I will take the explanations and advice of an 'anonymous audiologist' on the matter of hearing loss, grounded in known auditory physiology as shown, over a named audiophile/engineer not at all qualified as an authority in this specific area. As I said before, if you have any further questions on the matter, I suggest you contact that (or another) audiologist yourself, as I doubt anyone on here is qualified to answer.

 

[maverickronin]

The study assesses sounds made by the human ear (don't ask me how, first time I've heard of that one) and how they change when the ear is presented with a low frequency stimulus.

I looked though it, but as far as I can tell from other searching SOAEs are only a possible proxy for hearing damage. The discussion section includes a lot of hypothesizing how about this could cause hearing damage and suggests further things that could be studied, but the study itself doesn't demonstrate actual harm.

As I said earlier I find it unlikely that this could be such a hazard but remain undiscovered until now.

 

[maverickronin]

So what? The brilliant NWAVGuy was anonymous. Paul McGowan isn't. Should we trust the latter more just because he's put his name to his BS? You are free to believe who you want, but I will take the explanations and advice of an 'anonymous audiologist' on the matter of hearing loss, grounded in known auditory physiology as shown, over a named audiophile/engineer not at all qualified as an authority in this specific area. As I said before, if you have any further questions on the matter, I suggest you contact that (or another) audiologist yourself, as I doubt anyone on here is qualified to answer.

It's fine to be anonymous when you present data or a reasoned argument.

That post is a bare assertion without any references other than implied authority from their username. I'm perfectly willing to give them the benefit of the doubt that they are a real, qualified audiologist but it still doesn't come down to anything more than, "my doctor said so".

More importantly for the question at hand, is the complete lack of quantification.

To put it bluntly, any frequency at a loud enough intensity can damage your ears.

should really be re-frames as

To put it bluntly, any frequency at a loud enough intensity can damage your ears.

Since that's the real issue. What SPL, at what frequency, for how long?

 

[GaryH]

What SPL, at what frequency, for how long?

Ask them. I'm sure they would be happy to explain to you. I find it funny how you are willing to believe without evidence the bold assertion on here that high SPL (sub)bass frequencies won't cause hearing damage because our hearing threshold is high there, an inference that is not supported by known auditory physiology, yet you require much higher standards of evidence from an actual audiologist whose position is grounded in this known biology.