[makne]

This is exactly the point I was trying to make, except ^^ makes it so much better than me. Why is it that the closer to neutral, the “better” the headphone is measured (and often reviewed)?

Here’s a test - take a slightly warm of neutral/bright of neutral (whichever your preference) headphone and pass it through Sonarworks to “flatten” the FR as close to studio neutral as possible. If you think it improves the sound, great, that’s your thing. But every time I’ve tried it, it broke what made that particular headphone special or different.

Case in point, someone here said to try the Elear with Sonarworks and you’ll get an almost indistinsuishable sound to the Clear. I tried it, back and forth, with different tracks, then switched it off and uninstalled Sonarworks.

Maybe this is the point, that neutrality is a target for studio professionals and is therefore considered to be the baseline for how music is mixed. Except who’s to say that’s how ‘most’ people like to hear their music?

So without getting too OT, whenever I see a headphone lauded for its neutrality, it usually means I’ll need to tweak it to my taste, or look elsewhere.

The harman curve is a target curve that’s supposed to show what sounds optimal for most people. IE. The «optimal FR for most people».
«Slightly warm», «slightly bright» etc. Are all descriptions that would make no sense without an understanding of what is neutral. Neutral is the baseline that stuff is evaluated from. So yeah, I think it’s fair to say that a neutral headphone (again, according to Harman or another target curve) is OBJECTIVELY better, since after OBJECTIVE experiments we’ve found that that’s what most people think sounds the best.

What you’re talking about, that you don’t like sonarworks for instance, and a neutral sound in general, is a SUBJECTIVE opinion.

 

[gLer]

The harman curve is a target curve that’s supposed to show what sounds optimal for most people. IE. The «optimal FR for most people».
«Slightly warm», «slightly bright» etc. Are all descriptions that would make no sense without an understanding of what is neutral. Neutral is the baseline that stuff is evaluated from. So yeah, I think it’s fair to say that a neutral headphone (again, according to Harman or another target curve) is OBJECTIVELY better, since after OBJECTIVE experiments we’ve found that that’s what most people think sounds the best.

What you’re talking about, that you don’t like sonarworks for instance, and a neutral sound in general, is a SUBJECTIVE opinion.

Fair enough. Then it’s my subjective opinion that the Clear doesn’t measure objectively better than the LCD-3, although it does objectively measure differently.

 

[Phronesis]

The harman curve is a target curve that’s supposed to show what sounds optimal for most people. IE. The «optimal FR for most people».
«Slightly warm», «slightly bright» etc. Are all descriptions that would make no sense without an understanding of what is neutral. Neutral is the baseline that stuff is evaluated from. So yeah, I think it’s fair to say that a neutral headphone (again, according to Harman or another target curve) is OBJECTIVELY better, since after OBJECTIVE experiments we’ve found that that’s what most people think sounds the best.

What you’re talking about, that you don’t like sonarworks for instance, and a neutral sound in general, is a SUBJECTIVE opinion.

Beyond preferences deviating from neutrality, a few things to consider:

- Frequency response needs to be measured using a defined method, but this will generally not capture how the FR and acoustics vary with outer ear geometries; so perfectly neutral for one person may not be so neutral for another

- There's substantial variability in tonal balance of recordings, so the 'best' sound may be achieved with a headphone tonality which compensates for the tonality of a recording

 

[gLer]

There's substantial variability in tonal balance of recordings, so the 'best' sound may be achieved with a headphone tonality which compensates for the tonality of a recording

Excellent point, which explains why some genres (sometime down to individual tracks) sound 'better' with one headphone over another, even if the other headphone OBJECTIVELY measures 'better'.

 

[up late]

it would be helpful to know which measurements you guys are arguing over

 

[gLer]

it would be helpful to know which measurements you guys are arguing over

No arguments here, just lively debate Apparently, based on the Clear's FR chart, it 'measures' better than the LCD-3F because it conforms more closely to some Holy Grail of neutrality that's meant to be the standard bearer for headphone fidelity. Some are perfectly happy with that yardstick, while others (myself included) think the yardstick shouldn't be set in stone, and that indeed some variance of the 'curve' could actually be advantageous. I've never really been able to make heads or tails from FR charts, unless there's a glaring issue (like the treble mountain on the Beyer DT990), but if you think you can pick between the two headphones on the chart below and conclusively say which one measures 'better', be my guest. (Hint: LCD-3F on the left).

PS. If I'm not allowed to use this graphic here, please let me know. Credit: Innerfidelity.

 

[nanaholic]

I think it’s fair to say that a neutral headphone (again, according to Harman or another target curve) is OBJECTIVELY better, since after OBJECTIVE experiments we’ve found that that’s what most people think sounds the best.

That actually makes it subjective as the result is decided by a popular vote, which is the exact opposite of objective. It's a majority subjective preference.

Harmon was pretty specific in how they came to their famous curve, the analogy is like they give people to choose between Coke and Pepsi, and because more people choose Coke they then declare Coke as the way to make cola where most people will happen to like it. That is NOT objective, no matter how controlled they make the experiment like giving everyone the same cup at exactly the same temperature and only 100ml to taste etc.

 

[Luke Thomas]

So to simply. Clears = little more detail.
Lcd3. = little more, overall warmth.
I’ve owned both.

 

[makne]

That actually makes it subjective as the result is decided by a popular vote, which is the exact opposite of objective. It's a majority subjective preference.

Harmon was pretty specific in how they came to their famous curve, the analogy is like they give people to choose between Coke and Pepsi, and because more people choose Coke they then declare Coke as the way to make cola where most people will happen to like it. That is NOT objective, no matter how controlled they make the experiment like giving everyone the same cup at exactly the same temperature and only 100ml to taste etc.

In my eyes, it’s the closest thing we know of an «optimal FR» with the data we have available, and I don’t know of any other yardstick that would fit better. The objectivity-part is comparing all headphones to the same yardstick without personal bias from the reviewer, in my eyes that makes it objective (as objective as we can make it).

Anyway, my point is that I think we need to have a yardstick to measure different headphones up against, or else one could say that all FR’s are equally good (like the argument made above that FR is like colours...), which then would imply that FR doesn’t matter on an objective plane (obviously mighty important for the listener, still).
And at least in my eyes, the closer you get to that yardstick the more objectively good the headphone will be. The reason I chose Harman is because we don’t have a «one FR to rule them all» and afaik the Harman curve is the closest thing we know of, the best yardstick available.

Anyway, I think Tyll said it very well: Big deviations from neutrality should be considered a flaw, but small deviations create character. (Not a word for word quote, but the meaning’s the same).

So I guess we’re basically arguing if «character» is equal to or inferior to «dead neutral». In my eyes the line between «character» and «flawed» can often be blurry, so I say make «dead neutral» the objective winner to keep it clearer/more understandable.

And just to add, this was never to bash the lcd-3 in any way if that’s how it came out. I do think it’s a great sounding headphone, but throw all objectivity and measurements away and the Clear still has stolen my heart.

 

[up late]

No arguments here, just lively debate Apparently, based on the Clear's FR chart, it 'measures' better than the LCD-3F because it conforms more closely to some Holy Grail of neutrality that's meant to be the standard bearer for headphone fidelity. Some are perfectly happy with that yardstick, while others (myself included) think the yardstick shouldn't be set in stone, and that indeed some variance of the 'curve' could actually be advantageous. I've never really been able to make heads or tails from FR charts, unless there's a glaring issue (like the treble mountain on the Beyer DT990), but if you think you can pick between the two headphones on the chart below and conclusively say which one measures 'better', be my guest. (Hint: LCD-3F on the left).

PS. If I'm not allowed to use this graphic here, please let me know. Credit: Innerfidelity.

thanks. at least we now know the source of the frequency response measurements. like all headphone measurements, they should be treated as indicative rather than absolute. your ears and personal preference should be the final arbiter when comparing these headphones and any other headphones for that matter.

 

[up late]

In my eyes, it’s the closest thing we know of an «optimal FR» with the data we have available, and I don’t know of any other yardstick that would fit better. The objectivity-part is comparing all headphones to the same yardstick without personal bias from the reviewer, in my eyes that makes it objective (as objective as we can make it).

Anyway, my point is that I think we need to have a yardstick to measure different headphones up against, or else one could say that all FR’s are equally good (like the argument made above that FR is like colours...), which then would imply that FR doesn’t matter on an objective plane (obviously mighty important for the listener, still).
And at least in my eyes, the closer you get to that yardstick the more objectively good the headphone will be. The reason I chose Harman is because we don’t have a «one FR to rule them all» and afaik the Harman curve is the closest thing we know of, the best yardstick available.

Anyway, I think Tyll said it very well: Big deviations from neutrality should be considered a flaw, but small deviations create character. (Not a word for word quote, but the meaning’s the same).

So I guess we’re basically arguing if «character» is equal to or inferior to «dead neutral». In my eyes the line between «character» and «flawed» can often be blurry, so I say make «dead neutral» the objective winner to keep it clearer/more understandable.

And just to add, this was never to bash the lcd-3 in any way if that’s how it came out. I do think it’s a great sounding headphone, but throw all objectivity and measurements away and the Clear still has stolen my heart.

are you sure that tyll was using the harmon target response curve for both the lcd-3 and clear fr measurements because i suspect that he wasn't

 

[makne]

are you sure that tyll was using the harmon target response curve for both the lcd-3 and clear fr measurements because i suspect that he wasn't

afaik Tyll used the harman curve when talking about «neutral», but when you say it I can’t remember reading what compensation went into the graphs

 

[Phronesis]

Tyll describes his process for measuring headphone FR here: https://www.innerfidelity.com/content/headphone-measurement-proceedures-frequency-response

IMO, a key point is that a head-related transfer function (HRTF) has to be assumed for the measurement process, but in reality we each have our own HRTF based on the geometry of our ears and how the headphones are geometrically oriented relative to our ears. So the measurement provides some objectivity relative to the assumed HRTF, but perfect neutrality based on that HRTF and a given measurement process doesn't mean that the sound will be neutral at our own eardrums.

Moreover, generating a FR curve by sweeping single test tones across the frequency range isn't the same as playing music through a headphone and getting modal interactions in the driver. In other words, even if we all had the exact same ear geometry and the HRTF captured it perfectly, the measured FR curve still wouldn't tell us the whole story about how the headphone will sound to us. I can change the EQ of the LCD-i4 quite a lot using the 10-band EQ that comes with its app, which means big changes in the FR curve, yet the i4 continues to retain its personality and sounds distinctly different from the LCD-3, Clear, etc.

That doesn't mean that these kinds of FR curves aren't useful, just that we shouldn't infer too much from them. I think their main value is in telling us if we're in the ballpark of neutrality - which I think is a desirable target - and I agree with the view that good headphones will have subtle deviations from neutrality, not big deviations. The substantial impact those subtle deviations can have on our perception is an indicator of the artistry involved in designing headphones, kind of like the differences in how top chefs will prepare the same dish.

 

[makne]

Tyll describes his process for measuring headphone FR here: https://www.innerfidelity.com/content/headphone-measurement-proceedures-frequency-response

IMO, a key point is that a head-related transfer function (HRTF) has to be assumed for the measurement process, but in reality we each have our own HRTF based on the geometry of our ears and how the headphones are geometrically oriented relative to our ears. So the measurement provides some objectivity relative to the assumed HRTF, but perfect neutrality based on that HRTF and a given measurement process doesn't mean that the sound will be neutral at our own eardrums.

Moreover, generating a FR curve by sweeping single test tones across the frequency range isn't the same as playing music through a headphone and getting modal interactions in the driver. In other words, even if we all had the exact same ear geometry and the HRTF captured it perfectly, the measured FR curve still wouldn't tell us the whole story about how the headphone will sound to us. I can change the EQ of the LCD-i4 quite a lot using the 10-band EQ that comes with its app, which means big changes in the FR curve, yet the i4 continues to retain its personality and sounds distinctly different from the LCD-3, Clear, etc.

That doesn't mean that these kinds of FR curves aren't useful, just that we shouldn't infer too much from them. I think their main value is in telling us if we're in the ballpark of neutrality - which I think is a desirable target - and I agree with the view that good headphones will have subtle deviations from neutrality, not big deviations. The substantial impact those subtle deviations can have on our perception is an indicator of the artistry involved in designing headphones, kind of like the differences in how top chefs will prepare the same dish.

you’re clearly more knowledgeable than me in this field, and yeah, this post I can 100% agree with.
Measurements can only get you so far, like you say our methods are far from perfect and leaves room for many variables. I never make a purchase decision based upon measurements, and I don’t think anyone else should either, it’s just a tool that makes it easier to know where to start.

 

[Phronesis]

Fancy DAC, amp, etc. not needed with the Clear, IMO. Plugged Clear into the computer and watched/listened to this: .

OMG, mind blown, these guys are amazing. I feel like I was there with them. Note everyone listening with headphones, maybe it was mixed for headphones.