[vid]

This thread plays out like most others - blind rejection of evidence counter to beliefs. What's different is that the rejection is followed by a plea for objectivity.
 
In my view, the classification is a case study of a single person, and in effect something that would normally go in the "rank your headphones" thread.

 

[levap]

  This thread plays out like most others - blind rejection of evidence counter to beliefs. What's different is that the rejection is followed by a plea for objectivity.
 
In my view, the classification is a case study of a single person, and in effect something that would normally go in the "rank your headphones" thread.

I don't know where do you find "blind rejection". I was interested in discussion on details/speed headpnones classification and as we see:
 - No practical method to make objective details/speed headphones classification was proposed in this thread. 
 - I agree, that "the classification is a case study of a single person, and in effect something that would normally go in the "rank your headphones" thread". And here is this thread http://www.head-fi.org/t/715478/headphones-for-metal-music-ultimate-solution

 

[vid]

The question of speed is easy, and I already suggested a method for that as well. Take a waterfall plot, see how fast the sound decays, and that's your speed. Then see that as the response becomes more neutral, the waterfall plot shows a faster decay. Thus speed = neutrality of response. That, then, brings us to the question of HRTF since the physical ear is an acoustic filter and thus the response varies depending on which ear you combine the headphones with.

 

[levap]

  The question of speed is easy, and I already suggested a method for that as well. Take a waterfall plot, see how fast the sound decays, and that's your speed. Then see that as the response becomes more neutral, the waterfall plot shows a faster decay. Thus speed = neutrality of response. That, then, brings us to the question of HRTF since the physical ear is an acoustic filter and thus the response varies depending on which ear you combine the headphones with.

This does not take us any far from definition of "speed" from my first post. Again, you advice to develop the method. Not to "use practical method" I'm asking about.
Your advice have obvoius gaps from practical use. Why do you consider only decay, attack does not mean here? Does treble influence on "detail" recognition? How to classify "how fast the sound decays" for fast/average/slow? And most important - does your method have any proof on real headphones classification/comparison?
And you make final statement with open question, that it "brings us to the question of HRTF since the physical ear is an acoustic filter and thus the response varies depending on which ear you combine the headphones with"
 
Really, "The question of speed is easy"?
 
You are joking or you are troling I ask again?

 

[Head Injury]

  Your advice have obvoius gaps from practical use. Why do you consider only decay, attack does not mean here? Does treble influence on "detail" recognition? How to classify "how fast the sound decays" for fast/average/slow? And most important - does your method have any proof on real headphones classification/comparison?

Attack is the result of treble. Take a look at any square wave response and you'll see that it's actually made up of every frequency. The leading edge of the square wave, how high it rises, is the result of the highest frequencies. That's why Grados sound "fast", they have lots of treble and peaks that result in a sharper attack. But generally speaking, the rise time isn't any faster on these "fast" headphones, it just peaks higher. Go through some of Inner Fidelity's other graphs to see how frequency response translates to square wave response.
 
Decay's a similar story. Sure, all headphones have a different waterfall plot that shows slower or faster decay (Golden-Ears.net measures them), but most of it won't make a difference. Headphones all generally decay to -36 dB within 3 milliseconds in the treble and upper midrange, and waterfall plots are too inaccurate for anything below that. We're not going to hear the difference between a 2 ms and 3 ms decay in treble, at least not while music is playing and constantly masking those frequencies. The decay we hear has more to do with how much the low frequencies mask others, and the louder those frequencies are and the slower they decay the more easily they mask others. It's no coincidence that dark headphones tend to sound "slow" (or if the reviewer likes them, "lush"

).
 
Treble influences detail because it's not as likely to mask other frequencies. Bright headphones sound more detailed because there's more treble compared to bass, so there's less masking and lower overall decay which makes it easier to hear small details. Detail is also, obviously, affected by how neutral the frequency response is; if there's a big dip in the response, you're not going to hear much detail there. Keep in mind that headphones rarely actually produce more details than other headphones, they just make some easier to hear. I've never had a headphone reveal "details I've never heard" that don't immediately become audible on every other headphone. Once we hear it and know it's there, we'll hear it on everything.
 
You should spend some time down in good ol' Sound Science.

 

[vid]

I brought up the question of speed over at the sound science forum, too. I think the argument from that side was indeed that it's impossible to hear the difference in decay when it's quick anyway. Yet if decay and frequency response are linked, as they appear to be, you'd also need to specify how much an audible change in response would change the decay. I dare suggest a 3 dB increase in response is audible, won't change the decay massively, and will nonetheless change the decay, which loops back to the question I asked there, ie. are you hearing the decay or the response in that case, and does it matter which one?

 

[levap]

  Attack is the result of treble. Take a look at any square wave response and you'll see that it's actually made up of every frequency. The leading edge of the square wave, how high it rises, is the result of the highest frequencies. That's why Grados sound "fast", they have lots of treble and peaks that result in a sharper attack. But generally speaking, the rise time isn't any faster on these "fast" headphones, it just peaks higher. Go through some of Inner Fidelity's other graphs to see how frequency response translates to square wave response.
 
Decay's a similar story. Sure, all headphones have a different waterfall plot that shows slower or faster decay (Golden-Ears.net measures them), but most of it won't make a difference. Headphones all generally decay to -36 dB within 3 milliseconds in the treble and upper midrange, and waterfall plots are too inaccurate for anything below that. We're not going to hear the difference between a 2 ms and 3 ms decay in treble, at least not while music is playing and constantly masking those frequencies. The decay we hear has more to do with how much the low frequencies mask others, and the louder those frequencies are and the slower they decay the more easily they mask others. It's no coincidence that dark headphones tend to sound "slow" (or if the reviewer likes them, "lush"

).
 
Treble influences detail because it's not as likely to mask other frequencies. Bright headphones sound more detailed because there's more treble compared to bass, so there's less masking and lower overall decay which makes it easier to hear small details. Detail is also, obviously, affected by how neutral the frequency response is; if there's a big dip in the response, you're not going to hear much detail there. Keep in mind that headphones rarely actually produce more details than other headphones, they just make some easier to hear. I've never had a headphone reveal "details I've never heard" that don't immediately become audible on every other headphone. Once we hear it and know it's there, we'll hear it on everything.
 
You should spend some time down in good ol' Sound Science.

You are consistent in refusal of the gap from good ol' Sound Science to good ol' Sound Practice.
I'll try to express myself in IT example (we're in Internet anyway). UNIX OS professional can spend years polishing skills to use command line to control different OS aspects. And using GUI anyone can repeat 90% of operations in much more intuitive way. 
We do not have real GUI to translate measurments into qualitative parameters. Let's look at Golden-Ears review. http://en.goldenears.net/index.php?mid=GR_Headphones&page=1&document_srl=55120
They have a lot of charts with no comments and a chapter "Reviewer's opinion" with qualitative marks and (surprise, surprise!) no link to measurments results and a disclaimer "

The following chart is an expression of the reviewer's opinion on this product's measured data and perceived sound, and may include bias from using the Golden Ears reference unit as a standard."
 
It looks like we have a lot of professional R&D ahead to feel this gap. And sound industry is not going to apply resources. Or to publish results for internal use.

 

[vid]

We have research linking headphones, their frequency responses, and subjective impressions of those headphones. Your tear for no measurements-to-words is misplaced if it assumes headphones-to-words to be the alternate route; the obvious issue there being obvious.
 
By the way, no one in the sound science forum could answer me the decay question - in a way that would've stuck in my mind, anyway - so it's still open.