[YTCrazyTieGuy]

lol what an honor, I didn't expect Tyll to comment himself ^^. Sure is interesting.
 
Tyll do you find the full size FR plots compensated with Sean Olive curve match your subjective opinion of them better than the IEM plots do?
 
Edit: btw right now i'm listening to my Ety MC3 with -6dB at 1500 Hz (Q = 0.8) and +3dB at 50 Hz (low shelf) using Equalizer APO, which should get them close to Sean Olive curve (I used headroom MC3 and NAD Viso HP50 graphs and the Sean Olive compensated HP50 graph from Innerfidelity), and they sound better than ever! although, I don't have much experience with high end sound.

 

[Tyll Hertsens]

   
Tyll do you find the full size FR plots compensated with Sean Olive curve match your subjective opinion of them better than the IEM plots do?
 

 
Not sure it answers your question exactly, but I did an article showing the Harman curve compensating a variety of cans.  I think the Harman curve had the treble too depressed, and the bass possible a tad high. But it was a preliminary curve, I don't think they've published a final curve yet.

 

[money4me247]

  It would have been nice if Tyll could elucidate a bit on what he's thinking.
I agree that the transfer function should be the same, but I also agree on that there are some discrepancies between the measurements and my own experience. I don't know what's going on.
 
How is it we call him forth; Tyll Tyll Tyll, then click our heels thrice?

woww... i am just really surprised that summoning spell worked!!! =P

 

[YTCrazyTieGuy]

@Tyll
 
Actually I was exactly talking about that post - when you look at those graphs does it seem the full size graphs match your subjective opinion better than the IEMs?
 
for example, the SE535 graph shows a quite recessed treble response, and a very slight focus on mid-bass over mids and sub-bass. does this match your subjective opinion of them? the LCD-3 look quite similar but with a slightly less recessed treble and the mid-bass focus extending into the lower mids, but a lot more 10k-20k. does this match your subjective opinion of them? what would you say are the subjective differences between the SE535 and the LCD-3?
 
ty

 

[castleofargh]

  @Tyll
 
Actually I was exactly talking about that post - when you look at those graphs does it seem the full size graphs match your subjective opinion better than the IEMs?
 
for example, the SE535 graph shows a quite recessed treble response, and a very slight focus on mid-bass over mids and sub-bass. does this match your subjective opinion of them? the LCD-3 look quite similar but with a slightly less recessed treble and the mid-bass focus extending into the lower mids, but a lot more 10k-20k. does this match your subjective opinion of them? what would you say are the subjective differences between the SE535 and the LCD-3?
 
ty

the LCD3 is heavy and has really poor isolation. what it wasn't the question? but that will clearly change my audio experience too

 

[YTCrazyTieGuy]

Assuming you are in a perfectly quite environment isolation shouldn't make a difference, but it's true that maybe the weight of the headphones changes the subjective evaluation of the sound, at this point I have a feeling that these psychological differences are what we are looking for.

 

[bigshot]

There really isn't much point in measuring subjective preferences. It's going to be different for everyone. The thing to measure are the objective aspects, and then just give a precise description so people can decide whether they "like" it or not.

 

[YTCrazyTieGuy]

In an ideal world recordings would be made to be played with a flat response, but that ideal is far from the truth. Therefore, since the objective of music is to please us, the best response should be the one that makes as many recordings as possible sound as good as possible, and this can only be done subjectively.
 
(IMHO)

 

[Claritas]

There really isn't much point in measuring subjective preferences. It's going to be different for everyone. The thing to measure are the objective aspects, and then just give a precise description so people can decide whether they "like" it or not.

That's a very good way of putting it and makes good sense.

I used to think that. But here's a current counter-example: I used to find discussions of target curves irritating. The problems as I saw it included: it's a sophisticated poll, not an eternal truth; and it's a tyranny of the majority. I still think those concerns are valid, but I've gone out and tried some of the headphones that are broadly within range of the proposed curve and the proof, so to speak, is in the listening. It just sounds quite good. So I'm considerably more at ease now with the study of subjective preferences.

 

[bigshot]

My target curve is audibly flat. Then, if I want to adjust from there, I can. The point is to achieve the closest thing to accuracy that you can as a baseline. Talking about individual tastes in response curves is a waste of time. Talking about accurate response curves isn't. Accuracy is the first goal. You don't have any control over your presentation if you can't achieve that. Pursuing personal taste and pleasing subjective impressions is like a dog chasing its tail. Feelings change like the wind. Accuracy is an anchor.

 

[Claritas]

The proposed response curve isn't far from accurate. One of the basic ideas is that headphones need a mild bass boost (not mid-bass only) to reproduce the recordings accurately as they would sound on speakers. Makes sense. Another is some roll off in the treble to reflect equal loudness (Fletcher-Munson, &c.). Again, makes sense. Last, a slight upward notch in the upper treble to give it some air. Well, maybe. What we're seeing is that people prefer what they perceive as flat. Not something to complain about, I think.

 

[money4me247]

  My target curve is audibly flat. Then, if I want to adjust from there, I can. The point is to achieve the closest thing to accuracy that you can as a baseline. Talking about individual tastes in response curves is a waste of time. Talking about accurate response curves isn't. Accuracy is the first goal. You don't have any control over your presentation if you can't achieve that. Pursuing personal taste and pleasing subjective impressions is like a dog chasing its tail. Feelings change like the wind. Accuracy is an anchor.

What is considered to be perfectly accuracy is a bit variable & debatable though. All frequency response graphs use certain weighing to compensate for our hearing. There are also many variations of target frequency response curves. Also, due to the limitations of headphone drivers, a linear non-emphasized bass may appear deficient compared to the same frequency response on a set of speakers. Finally, we naturally hear the mid-range to be louder, so a linear uncompensated frequency response will actually be mid-centric to our ears.
 
I agree with the approach to target something that is "audibly neutral" for best results, but I think there is some debate on what exactly neutral is. There is also some variation in sound signature in the TOTL neutrally-tuned flagships, so it seems like there quite a bit of wiggle room.

 

[bigshot]

  What is considered to be perfectly accuracy is a bit variable & debatable though. All frequency response graphs use certain weighing to compensate for our hearing.

 
The key word is *audibly* flat... not flat on a chart or flat to a microphone... flat to human ears.

 

[castleofargh]

but many people want someone to hold their hand and say "that one is neutral, it's the real sound. if you buy it you're an expert audiophile and you hear the real sound like the artist intended" (and other usual marketing bullzz). we could have 50different kind of measurements and profiles with a scan of our own ears, you would still have that guy asking if it's "better" than X headphone. some people just need that, so for them it would be nice if the compensation curve didn't push most people into thinking that an ER4 sounds neutral. because then they're in for a bad joke.

 

[Strangelove424]

  In an ideal world recordings would be made to be played with a flat response, but that ideal is far from the truth. Therefore, since the objective of music is to please us, the best response should be the one that makes as many recordings as possible sound as good as possible, and this can only be done subjectively.
 
(IMHO)

I agree with your point to a certain extent. I recently picked up a used pair of HD600s because I needed something more forgiving. A large degree, perhaps most music is produced with no regard for the "ideal". But the ideal is not something complicated or mystifying, it's just the same old standards that have been around since day one. The people making music just need to respect those standards. Once the recordings start chasing non-neutral responses, and then we go chasing non-neutral responses to cover it up, we've entered a really dangerous (and expensive) cycle. I think we should judge a headphone not by how much it covers up in badly produced music, but how much it reveals in well produced music.