Oratory1990 has REMEASURED a load of headphones and updated his pdf's

[Feelas]

With that said, you have just given me a download link for Python 3.9, and AutoEQ states clearly that even Python 3.8 doesn't work yet.

Aaaaaaaah... oops! Forgot to check whether it matters, oy, it's late here!

Guess you found it anyways, yet here goes 3.7.9 (direct link to installer).

 

[Jose Hidalgo]

To get back on topic, today was an interesting day.
I'm currently evaluating the new Oratory preset for HD600. Then it will be the turn of the Sundara.

It will take some time to form a realistic opinion. So far I think I prefer the new HD600 preset vs. the old one. Today's session with the EQed HD600 was quite satisfactory, especially on guitars and vocals. It's easy to see why the Senn is considered a master at those : the midrange texture just has something more that makes those instruments come to life while being gently pushed at you.

Tomorrow and this WE, more listening ! Yay !

 

[Honken]

You didn't find the bass boomy with the new HD600 preset? I've been going back and forth between his three presets, the old Harman preset, the new Harman preset and the Optimum Hi-Fi preset, and personally I've come to enjoy the last one the most.

I suppose I could just change the 1st band of the new Harman preset to match the first band from the Optimum preset, but I'm a bit too lazy for that now that I've entered the Optimum preset into my EQ.

 

[Jose Hidalgo]

You didn't find the bass boomy with the new HD600 preset? I've been going back and forth between his three presets, the old Harman preset, the new Harman preset and the Optimum Hi-Fi preset, and personally I've come to enjoy the last one the most.

I suppose I could just change the 1st band of the new Harman preset to match the first band from the Optimum preset, but I'm a bit too lazy for that now that I've entered the Optimum preset into my EQ.

So far I didn't find the bass boomy, but I need to keep listening and comparing.
Now here's what I think about this matter :

1. I'm no expert, just an enthusiast who listens to what the real experts here have to say.
   
   
2. With that said, I've read several threads on other forums, where people wonder why the Harman curve is considered as "neutral" when it's "obviously V-shaped". I wonder if those people know about Fletcher-Munson and the subsequent studies up to ISO 226-2003. Bass always needs to be boosted in order to be perceived as "equal loudness" as the rest. By how much, it depends on the listening loudness level, but it always needs to be boosted (I suppose headphone manufacturers can take care of that too, but do they all really and to what extent ?). AFAIK there is a often-used reference loudness level at about 83 dB, so I suppose that may be a good starting point for designing curves such as the Harman one. I hope that makes sense.
   
   
   
3. The Harman curve is "somewhat close" to the 80 dB equal-loudness curve, and so it can be considered as fairly neutral. It may not be completely neutral, because it's based on listeners subjective preferences. For example it seems to have more emphasis on the higher frequencies with that bump around 3-4 KHz. Maybe the reason of that emphasis can be found in point 7 below, who knows ?
   
   
4. The "Optimum Hifi curve" does look less neutral to me relatively to the equal-loudness curves, with no bass boost at all below 100 Hz, and emphasizing the 100-800 Hz range compared to the Harman curve (look at the comparison below). It has a 3-4 KHz bump similar to the Harman one, but since it has less bass, I guess you hear the bump more. I don't know where that "Optimum Hifi curve" comes from exactly (who designed it and why/how). Maybe someone has some info on that matter ? I'd love to have some info before actually listening to it (I haven't tried it yet).
   
   
   
   
   
   
5. Last but not least, age may be a factor here. I guess when we become old, we progressively lose our hearing ability on higher frequencies. Maybe in that case older people enjoy more a treble boost : because for them it's not really a boost but a compensation for their hearing loss to make the sound neutral... to them. I hope that makes sense.
   
   
6. As for me personally, I'm very sensitive to treble. Maybe that's one of the reasons why I do enjoy my Nighthawk Carbon : they may have slightly recessed mids, and yet those mids are enough for me to perceive all the information : it's there if you want it, it's just not "in your face" like with other headphones. The HD600 are very different, and I like them very much too for other reasons, mainly because of their great midrange I guess. But maybe their highs can cause me some fatigue whereas the Nighthawks tend to cause me none.
   
   
7. I don't claim to have better ears than anybody, it may even be the opposite, lol. But I know that I've always tried to take care of them. I didn't always succeed, but at least I tried, lol. For example I've rarely listened to music at very high volumes/for long periods of time, or been to night clubs (I hate that), or to concerts, etc.. Lots of young people nowadays have destroyed their ears because of iPod abuse. As a consequence they need more and more treble to be able to actually hear something. That's a real pity.

 

[thewas]

The Harman curve is "somewhat close" to the 80 dB equal-loudness curve, and so it can be considered as fairly neutral. It may not be completely neutral, because it's based on listeners subjective preferences.

You are mixing up different things there, the equal-loudness curve should never be a loudspeaker or headphone target because its already considered when mixing and mastering the music we listen to, we only have to consider the differences of the different curves when we listen at a different SPL, for example only 70dB vs the 80-85 dB most music is usually mastered at and then you get something like this:

 

[Jose Hidalgo]

I would agree @thewas Except maybe for one thing : the equal-loudness isn't necessarily what the mixing/mastering engineers have in mind. Theoretically yes, but in the real world, albums aren't necessarily mixed/mastered to sound equally loud at all frequencies. Other criteria come in play depending on the type of music, not to mention the engineer's own freedom to do pretty much as he wishes (and which may sometimes result in a poor final result, like heard on many albums, unfortunately). I don't think we should take that part for granted. Don't you think so ?

And a question : what is the reason for the 3-4 KHz bump present in both Harman and "optimum Hi-fi" curves ? (other than a personal preference of the listeners). According to your curves there shouldn't be any.

 

[thewas]

Theoretically yes, but in the real world, albums aren't necessarily mixed/mastered to sound equally loud at all frequencies. Other criteria come in play depending on the type of music, not to mention the engineer's own freedom to do pretty much as he wishes (and which may sometimes result in a poor final result, like heard on many albums, unfortunately). I don't think we should take that part for granted. Don't you think so ?

Yes, but all that has nothing to do with with equal loudness curve form you mention.

And a question : what is the reason for the 3-4 KHz bump present in both Harman and "optimum Hi-fi" curves ? (other than a personal preference of the listeners). According to your curves there shouldn't be any.

This is the inner ear canal resonance and has nothing to do with the loudness curves, see this article and especially the parts
What are "raw graphs"?
Compensations
Measuring Systems
https://crinacle.com/2020/04/08/graphs-101-how-to-read-headphone-measurements/

A frequency neutral source like a loudspeaker always makes this bump when we measure inside our ear but since we are born with it our brain compensates it and we hear it as neutral.

 

[Jose Hidalgo]

Oh yes, the ear canal compensation, of course. Silly me.
You see, that part has always puzzled me. Why would the ear canal resonance always be located at the same frequency ?
For example, I have unusually tiny ear canals (as said by doctors), which makes it more of a challenge to remove all the wax. Wouldn't that mean that my resonance frequency is higher than most people ? (and not by just some percent, I mean by a lot). It could help explain why I'm more sensitive to treble than most people. And it could also mean that the Harman bump isn't ideally located for my ears ?

 

[thewas]

Oh yes, the ear canal compensation, of course. Silly me.
You see, that part has always puzzled me. Why would the ear canal resonance always be located at the same frequency ?
For example, I have unusually tiny ear canals (as said by doctors), which makes it more of a challenge to remove all the wax. Wouldn't that mean that my resonance frequency is higher than most people ? (and not by just some percent, I mean by a lot). It could help explain why I'm more sensitive to treble than most people. And it could also mean that the Harman bump isn't ideally located for my ears ?

Yes, this is also the reason why headphone EQ based on such fixed curves doesn't work similarly well for everyone.

 

[wasnotwasnotwas]

Yes, this is also the reason why headphone EQ based on such fixed curves doesn't work similarly well for everyone.

There are 2 separate questions for me with any eq target, Harman included.

1. Does it make the hp subjectively sound better than stock? I suspect the answer for most people, even those with unique ears, is yes. If not, just don't use it

2.is the selected target the best eq for that one person? Far harder to answer if it indeed answers number 1. That's where user knowledge and experience plus a willingness to fiddle comes in. Harman might have an advantage over other targets as it factors in preference and covers a wide spectrum of the user base.

Some users may well stop at question one and (wrongly) assume that because it sounds better than stock, then that's as good as it can get.

 

[Jose Hidalgo]

Thanks guys

One last thing : why wouldn't the ear canal bump be already taken into account by the equal-loudness curves ? If that were the case there wouldn't be any need for it in the Harman / "optimum HiFi" curves.
Plus if somehow it wasn't taken into account by the equal-loudness curves, then why wouldn't it be taken into account by the mixing/mastering ? Is it something specific to headphone listening ? Why would it be ?

I'm actually surprised by the things we need to take into account that haven't been considered previously.

 

[wasnotwasnotwas]

Thanks guys

One last thing : why wouldn't the ear canal bump be already taken into account by the equal-loudness curves ? If that were the case there wouldn't be any need for it in the Harman / "optimum HiFi" curves.
Plus if somehow it wasn't taken into account by the equal-loudness curves, then why wouldn't it be taken into account by the mixing/mastering ? Is it something specific to headphone listening ? Why would it be ?

I'm actually surprised by the things we need to take into account that haven't been considered previously.

I believe the bump is needed for HP because the sound is coming at right angles , directly into the ears, bypassing some of the effect of ear shape for more "normal" front on listening with speakers. Dont forget Harman for HP is an attempt to replicate "good" speakers in a room with some user preference wiggle room.

 

[Jose Hidalgo]

Got it.
I had to read you twice, because here in France HP actually means "Haut-Parleur", that is "loudspeaker".
Once I remembered that in english HP means "Headphones", all was good.

 

[Dreyfus]

Just a side note:
Keep in mind that the Fletcher-Munson curves were captured using pure sine tones.

Equal Loudness Contours - Purely Misleading

It appears that band-passed random noise is better at simulating music due to its wider bandwidth.
I would be careful about using the Fletcher-Munson curves for headphone corrections.

 

[Robbo99999]

@Jose Hidalgo , this is my take on it. You can't use Fletcher Munson curves as a Target Curve for headphone corrections because they're created (I assume) from measurements from a mic in free air at a listeners position and correlated to the listeners reported perceptions, they're not measured with a mic at the eardrum....so you can't use them for correction purposes (in terms of correcting headphones that have been measured "on a rig") because it's not taking into account all the things that user "thewas" mentioned in this post: https://www.audiosciencereview.com/...phones-and-updated-his-pdfs.18152/post-594435. So they can't be related to headphone measurements done on any rig, it's just not the same thing...... There are some elements that might be noticed from the Fletcher Munson Curves that could be seen as their origins in the Harman Headphone Target Curve for instance, but it's just not the same thing.

In terms of Fletcher-Munson though, I think there are some algorithms around in some software that can change the tilt of an existing frequency response to try to account for playback at different playback volume levels. So if you were listening at low volume levels it might decide too boost the bass by tilting it downwards automatically. I don't know a whole lot of specifics on these software algorithms, but I've heard of them in passing. Either way Fletcher Munson Curves are just not the same thing as a Harman Headphone Target Curve or indeed any Headphone Target Curve.

 

[Jose Hidalgo]

Just to clarify, I don't intend to use Fletcher-Munson curves for any of my headphones. I intend to use them only for my loudspeakers.

I'm currently trying the Oratory presets (Harman) on both the HD600 and the Sundara. I need more time to be sure, but so far I could sum them up as follows : (1) they improve things slightly (both the HD600 and the Sundara are fairly neutral without EQ), especially for the HD600 (better bass extension), and (2) there may still be a little too much treble for my taste. No harsh sound, no sibilances, just maybe a little too much treble. So maybe it's the 3-4 KHz bump that bothers me, I don't know.

 

[Robbo99999]

Just to clarify, I don't intend to use Fletcher-Munson curves for any of my headphones. I intend to use them only for my loudspeakers.

I'm currently trying the Oratory presets (Harman) on both the HD600 and the Sundara. I need more time to be sure, but so far I could sum them up as follows : (1) they improve things slightly (both the HD600 and the Sundara are fairly neutral without EQ), especially for the HD600 (better bass extension), and (2) there may still be a little too much treble for my taste. No harsh sound, no sibilances, just maybe a little too much treble. So maybe it's the 3-4 KHz bump that bothers me, I don't know.

Re the HD600, an EQ to the Headphone Harman Curve actually reduces the treble, so that should have improved it for you. If it still doesn't help you could keep the bass Low Shelf boost and then just experiment with High Shelf cuts above 1kHz.....I'd probably start with High Shelf 1100Hz Q0.707 G-1......this would be on a stock headphone, maybe not too much point in using Oratory's EQ if it doesn't match your HRTF. You could experiment with decreasing the High Shelf in 1dB increments and then 0.5dB increments when you are closer to perfection. Or if you like it wouldn't hurt to experiment with just dumping this same High Shelf cut on top of Oratory's EQ if we assume the general shape of the 1-6kHz bump is a more optimum shape. Or you could experiment with shifting the 1-6kHz bump further to the right and designing EQ's to that new Target Curve, because I think you & thewas concluded that the position of the bump can shift depending on the width of your ear canal (although I can't vouch for that), and you concluded it could be shifted to the right. If you can find any resources that show the size & positioning of that hump in relation to ear canal width that would give you an idea on how to the shape the bump in terms of general shape and how much it can shift - so perhaps this would allow you to shape the Target Curve more accurately.

 

[Jose Hidalgo]

Yes, the Harman curve reduced the treble slightly (that's why I said "there may still be a little too much treble for my taste").

The position of the bump depending on the width of the ear canal is just a hypothesis, I don't have any data on that. Maybe some of the experts here can confirm. It makes sense to me since a resonance frequency should be dependant on some physical dimension.

Thanks a lot for the tips, I will try them. I'm trying to see if I can get used to the Harman curve first.

 

[bidn]

If anyone has a collection of proper measurements like Oratory, let me know if you want yours on the site as well. Hopefully we can gather as many comparable measurements as possible.

I do also have all the Innerfidelity measurements ready to plug into the site, but I'm not sure I'm allowed to. Does anyone know who owns those measurements now that Tyll stopped at IF?

Thank you for your work, Suuup.

Re. copyright, it does not apply to measurements.
Copyright became law based on the idea that creations by humans are " proper" (→ some kind of limited property), specific re. their human author, i.e. a book is written according to the proper, supposedly unique, personal writing style of the human author, and much later this was extended to works of art (painting, musical composition, and more recently to photographs).
Copyright does not cover works created by a machine (even creations by so-called "artificial intelligence" do not fall .under copyright).

Headphones measurements are not the specific personal artistic expression of a human artist, but are automatically generated by a device, hence they do not fall under copyright .

 

[bidn]

More stuff on copyright:

# 1 Copyright does not cover facts.
Contrary to personal expressions, scientific measurements are facts, i.e. they cannot be copyrighted.

# 2 Copyright does not cover scientific principles.

# 3 Back to the requirement of personal human authorship,
let's take the country where you can copyright the most things, the USA.
There is a US governmental copyright compendium :
official pdf here = https://www.copyright.gov/comp3/docs/compendium.pdf

It states:

306 The Human Authorship Requirement
The U.S. Copyright Office will register an original work of authorship, provided that the work was created by a human being.

The copyright law only protects “the fruits of intellectual labor” that “are founded in the creative powers of the mind.” Trade-Mark Cases, 100 U.S. 82, 94 (1879). Because copyright law is limited to “original intellectual conceptions of the author,” the Office will refuse to register a claim if it determines that a human being did not create the work.

taken here:
https://law.resource.org/pub/us/compendium/ch300.html#s306


From the same source, re. facts:

313.3(C) Facts
Facts are not copyrightable and cannot be registered with the U.S. Copyright Office. “No one may claim originality as to facts . . . because facts do not owe their origin to an act of authorship.” Feist, 499 U.S. at 347 (internal citation omitted). A person who finds and records a particular fact does not create that fact; he or she merely discovers its existence. As a result, facts “are never original” and Section 102(b) of the Copyright Act “is universally understood to prohibit any copyright in facts.” Id. at 356. “[This] is true of all facts – scientific, historical, biographical, and news of the day.” Id. at 348.

https://law.resource.org/pub/us/compendium/ch300.html#s313.3(C)