[cdma2k]

   
 My reason for caring about neutrality (as an ideal to pursue) is because I want to get the most accurate and realistic sound possible. I want no more warmth and lushness than is present in the actual recording.

Yes, that's also my understanding of the definition of neutrality.

 

[Music Alchemist]

  This is roughly in line with my experience.  HD800 was neutral with just a bit too much treble, and Audeze is warm/dark though I did like the LCD2 and LCDX in general.  No experience with TH-900, but the SR-009 was just stupidly detailed and "right" out of a BHSE.  I really enjoyed it.  Definitely my choice if I had about 10K to put into just headphones.
 
Regarding if TOTL are worth it?  Some are some not so much.  Depends on you want out of your system.  You can probably get what you want without spending nearly as much.  Also, by the time you are looking in to an SR-009 with a BHSE you can start considering pretty nice speaker systems.

 
hehe, I debated with myself for a long time and decided to just go straight for the SR-009 and BHSE (with SR-007 for good measure), instead of collecting stuff that I would only upgrade from (and may not keep thereafter) anyway. As for speakers, that can get very complicated and expensive. I for one am not even considering them until at least a few years from now.

 

[Head Injury]

   
Interesting. This is the first time I've seen warm and lush described as neutral. Most people differentiate between them definitively, even regarding them as opposites, such as saying, "Do you want a neutral, warm/dark, or cold/bright sound?" Maybe you're using different meanings here, because one meaning of warm involves excessive bass. My reason for caring about neutrality (as an ideal to pursue) is because I want to get the most accurate and realistic sound possible. I want no more warmth and lushness than is present in the actual recording. Don't headphones that are more warm/lush add more than is actually there? Most people also regard Audeze and Fostex as very colored headphones, so your perspective is confusing to me. I appreciate any explanations you can give.

I just mean what other people call "warm" is more neutral than what people normally call neutral, like the LCDs and TH-900. Warm and bright are relative to neutral, so I simply wouldn't call these ones warm. Anything much bassier would be warm.
 
The question is, what's on the recording? Who's to say the bright headphone is reproducing it correctly but the warm on isn't? Nature is more warm than Head-Fiers tend to think. So are recordings. I think the tracks on Michael Jackson's "Thriller" are pretty bright, but look at the spectrum analysis. There's still more bass energy than treble energy.

 
 
Here's some harpsichord:

 
And Stevie Ray Vaughan's Little Wing, which despite a little clipping here sounds fantastic:

 
So shouldn't these sound a little warm? If it sounds a little warm it's reproducing what's on the track. It's the bright headphones that aren't doing it right.
 
Remember that YouTube video I linked of the TH-900 review, and how close it sounds to the reference track? The same group reviewed some other high-end headphones in the same way, including the HD800, and they aren't nearly as close. I think that's a good indication of neutrality, if the goal is to reproduce the recording without coloration. Is the HD800 still an excellent headphone? Yes, but it needs a little EQ like virtually all headphones. People need to stop shying away from that. The best speaker systems in the world need EQ too.
 
When someone spends $1500 on a headphone, they don't want to think it's colored. But they're all colored, that's why we keep buying more

 

 

[SunshineReggae]

Isn't it normal for recordings to have more bass emphasis than treble emphasis though?

 

[Music Alchemist]

  I just mean what other people call "warm" is more neutral than what people normally call neutral, like the LCDs and TH-900. Warm and bright are relative to neutral, so I simply wouldn't call these ones warm. Anything much bassier would be warm.
 
The question is, what's on the recording? Who's to say the bright headphone is reproducing it correctly but the warm on isn't? Nature is more warm than Head-Fiers tend to think. So are recordings. I think the tracks on Michael Jackson's "Thriller" are pretty bright, but look at the spectrum analysis. There's still more bass energy than treble energy.
 
Here's some harpsichord:
 
And Stevie Ray Vaughan's Little Wing, which despite a little clipping here sounds fantastic:
 
So shouldn't these sound a little warm? If it sounds a little warm it's reproducing what's on the track. It's the bright headphones that aren't doing it right.
 
Remember that YouTube video I linked of the TH-900 review, and how close it sounds to the reference track? The same group reviewed some other high-end headphones in the same way, including the HD800, and they aren't nearly as close. I think that's a good indication of neutrality, if the goal is to reproduce the recording without coloration. Is the HD800 still an excellent headphone? Yes, but it needs a little EQ like virtually all headphones. People need to stop shying away from that. The best speaker systems in the world need EQ too.
 
When someone spends $1500 on a headphone, they don't want to think it's colored. But they're all colored, that's why we keep buying more

 

 
That's reassuring, then, since I was going to get the TH-900 anyway. Regardless, what I want to go after is not too warm, not too bright, which ideally would be neutral. I have no problem with headphones reproducing warmth...but when you call a headphone warm, that tends to imply that it is adding extra warmth that isn't there. If what you are saying is true, then the headphones often thought to be neutral are in fact too bright (which is the subjective impression of many listeners anyway), and the neutral* headphone are the ones thought to be on the warmer side, when in fact, they are merely reproducing the warmth in the recordings.
 
I told a friend about some of the things you said and he replied with this:
 

And what.... the LCD-2 is one of THE MOST COLOURED headphones EVER. I don't even....

 
Assuming again that you are correct, perhaps this is another example of mistaking bright headphones as neutral* and neutral* headphones as dark.
 
*As you said, all headphones have some coloration, so when I say neutral, I really mean "more" neutral.

 

[Head Injury]

  Isn't it normal for recordings to have more bass emphasis than treble emphasis though?

Yes it is, and that's sort of my point. I would expect the perceived response from a neutral headphone to follow a downward slope a bit like these (not quite so extreme because we're naturally more sensitive to upper midrange and lower treble). I would think, in theory, if a speaker that tested flat in a treated room played these tracks back, the overall response should look the same. The trouble is finding the right headphone response that achieves the same at the ear drum, and to my ears the "warm" headphones achieve it (and that slope is why they're called warm, but it's actually inherent to the recordings). But I understand why more treble is more acceptable at Head-Fi than the alternative; treble is where the detail is, and doesn't mask other frequencies as easily as bass. It's very easy to make a "warm" headphone that's too warm and sounds slow, but even overly bright headphones sound fast and detailed. So ultimately the best headphones to test these things would be the ones that scored high on the flagship analysis, due to their smooth extended frequency response and low distortion being easier to EQ without causing unwanted masking in the bass.
 
Probably two years ago now I set up my first LCD-2 EQ using online graphs and a sine wave generator to try to even out perceived frequency response. What I got is almost the same as the Harman-based EQ I'm using now, just with a little less bass. So I think what I'm hearing is pretty neutral. I could lower the lower midrange more, but that's part of what makes the LCD-2 fun.
 
I want to continue to emphasize that everyone has different priorities and ears. Coming from Grados, the "dark" LCD-2 was a breath of fresh air, so no doubt I'm now biased toward warmth. I also want to continue to emphasize how awesome EQ is, and encourage everyone to use one to find what's neutral or preferable to them. You can save yourself a lot of money in the long run.

 

[derbigpr]

   
On the other hand, some headphones get A+ while they have serious problems. The HD800 has annoying sibilance for example.

 
Recordings have sibilance, not HD800's.

 

[derbigpr]

 
/cut
 

 
 
 
I agree with you that headphones that sound slightly "warm" sound more realistic and more neutral than those that have flat response and look neutral on paper. But, it's hard to judge that from recordings, simply because, as you put it here, no recording is neutral and flat, they all have an emphasis on one or the other area of the sound. On top of that, most sound engineers nowadays don't know what the hell they're doing when mixing, and end up totally ruining any realism and naturalness in the sound, so the final recordings have data on them that sounds nothing like the vocal or instrument sounded live when it was recorded.
 
I mean, take these two recordings. Same song and singer, same guitar, same microphone. The first video is a totally raw recording, no processing at all. The second one is edited and mastered for final release. Sure the first one is a bit sibilant, but it sounds so much more realistic, textured, dynamic, simply alive, listening to that with T1's makes me feel like someone is singing 1 meter away from me. The mastered one sounds like a compressed, thin, hollow mess, infinitely worse in every way. Notice how vocal sounds much thinner, but at the same time boomy compared to a raw recording, both the vocal and guitar have poor tonality and lack detail, despite being more airy. Both recording are colored, both are far from being flat if we take a look at their spectrum analysis.
 
Raw: http://youtu.be/4PCne5BZyeU?t=52s
Mastered: http://youtu.be/b1fT7MQujVI?t=6s
 
 
This is why I think it's totally pointless to try and get a "neutral" or " balanced" headphones in hope that they will sound better than colored headphones do. Everything we listen to is colored anyway, there's no such thing as neutral, and the whole point of hi-fi (despite the definition of high fidelity) is to find the coloration that we like the most, or the one that sounds closest to the live sounds. We're mixing the lack of neutrality and precision of a recording AND the lack of neutrality and precision of the headphone every time we listen to music. Sure, buying a headphone with a flat frequency response does mean that we eliminated a big part of the equation and it means one less variable to worry about, but again, it's not a guarantee of better or more neutral, natural, realistic, whatever...sound.

 

[Music Alchemist]

  I agree with you that headphones that sound slightly "warm" sound more realistic and more neutral than those that have flat response and look neutral on paper. But, it's hard to judge that from recordings, simply because, as you put it here, no recording is neutral and flat, they all have an emphasis on one or the other area of the sound. On top of that, most sound engineers nowadays don't know what the hell they're doing when mixing, and end up totally ruining any realism and naturalness in the sound, so the final recordings have data on them that sounds nothing like the vocal or instrument sounded live when it was recorded.
 
I mean, take these two recordings. Same song and singer, same guitar, same microphone. The first video is a totally raw recording, no processing at all. The second one is edited and mastered for final release. Sure the first one is a bit sibilant, but it sounds so much more realistic, textured, dynamic, simply alive, listening to that with T1's makes me feel like someone is singing 1 meter away from me. The mastered one sounds like a compressed, thin, hollow mess, infinitely worse in every way.  Both recording are colored, both are far from being flat if we take a look at their spectrum analysis.
 
Raw: http://youtu.be/4PCne5BZyeU?t=52s
Mastered: http://youtu.be/b1fT7MQujVI?t=6s
 
This is why I think it's totally pointless to try and get a "neutral" or " balanced" headphones in hope that they will sound better than colored headphones do. Everything we listen to is colored anyway, there's no such thing as neutral, and the whole point of hi-fi (despite the definition of high fidelity) is to find the coloration that we like the most, or the one that sounds closest to the live sounds. We're mixing the lack of neutrality and precision of a recording AND the lack of neutrality and precision of the headphone every time we listen to music. Sure, buying a headphone with a flat frequency response does mean that we eliminated a big part of the equation and it means one less variable to worry about, but again, it's not a guarantee of better or more neutral, natural, realistic, whatever...sound.

 
High fidelity is the reproduction of audio signals with minimal distortion or coloration, so yes, this is in spite of the definition. Perhaps we should come up with new terms to reflect this: that some people are interested in not only high quality sound, but coloring/altering it in a way that is more pleasing (and potentially more realistic) to them.
 
I'm still very confused about Audeze and Fostex supposedly being more neutral than Sennheiser and STAX, though...

 

[SunshineReggae]

 
Yes it is, and that's sort of my point. I would expect the perceived response from a neutral headphone to follow a downward slope a bit like these (not quite so extreme because we're naturally more sensitive to upper midrange and lower treble). I would think, in theory, if a speaker that tested flat in a treated room played these tracks back, the overall response should look the same. The trouble is finding the right headphone response that achieves the same at the ear drum, and to my ears the "warm" headphones achieve it (and that slope is why they're called warm, but it's actually inherent to the recordings). But I understand why more treble is more acceptable at Head-Fi than the alternative; treble is where the detail is, and doesn't mask other frequencies as easily as bass. It's very easy to make a "warm" headphone that's too warm and sounds slow, but even overly bright headphones sound fast and detailed. So ultimately the best headphones to test these things would be the ones that scored high on the flagship analysis, due to their smooth extended frequency response and low distortion being easier to EQ without causing unwanted masking in the bass.
 
Probably two years ago now I set up my first LCD-2 EQ using online graphs and a sine wave generator to try to even out perceived frequency response. What I got is almost the same as the Harman-based EQ I'm using now, just with a little less bass. So I think what I'm hearing is pretty neutral. I could lower the lower midrange more, but that's part of what makes the LCD-2 fun.
 
I want to continue to emphasize that everyone has different priorities and ears. Coming from Grados, the "dark" LCD-2 was a breath of fresh air, so no doubt I'm now biased toward warmth. I also want to continue to emphasize how awesome EQ is, and encourage everyone to use one to find what's neutral or preferable to them. You can save yourself a lot of money in the long run.

 
I was with you when you posted those examples, but now I'm not sure what you are posting haha.
 
All I was gonna say was, if recordings generally have tendency to emphasize lower frequencies, we need bigger sample size to see if Thriller album really is inherently bright or dark recording. This album surely sound on the bright side on my DT880. Honestly I like it better with HD598, which is exception for me.

 

[Music Alchemist]

  I was with you when you posted those examples, but now I'm not sure what you are posting haha.
 
All I was gonna say was, if recordings generally have tendency to emphasize lower frequencies, we need bigger sample size to see if Thriller album really is inherently bright or dark recording. This album surely sound on the bright side on my DT880. Honestly I like it better with HD598, which is exception for me.

 
Having a higher sample size and/or sample rate will do nothing, since all the extra information is outside the range of human hearing. The master is what matters.

 

[Head Injury]

   
Having a higher sample size and/or sample rate will do nothing, since all the extra information is outside the range of human hearing. The master is what matters.

He means more examples, not higher sampling rate

 
 
Thriller is still definitely a bright album (a lot of 80s albums were, when digital audio was new). Even in the examples I posted you can tell it has more treble energy (10 kHz is at -48 dB compared to -60 and -54). But bright is relative. When reproduced by a neutral system, there should still be more bass produced than treble.
 
All I'm trying to say (words are hard) is a downward slope is neutral when playing a recording with a downward slope. A perfectly flat system (one that reproduces all pure tones at the same volume, as we perceive it at our ear drum) will sound like it has less treble when playing back a recording because the recording itself has less treble. The problem is, what kind of headphone response is flattest at the ear drum? That's what these compensation curves are trying to figure out. Harman hasn't got it right either, I'm sure, and I'm still tweaking my EQ to see if I can make it better for me.

 

[Music Alchemist]

  He means more examples, not higher sampling rate

 
Oh. lol.
 
So, just to clarify...when you call the LCD-2 / TH-900 warm and lush, you really mean neutral, accurately reproducing the warmth and lushness in recordings...right?

 

[SunshineReggae]

 
He means more examples, not higher sampling rate

 
 
Thriller is still definitely a bright album (a lot of 80s albums were, when digital audio was new). Even in the examples I posted you can tell it has more treble energy (10 kHz is at -48 dB compared to -60 and -54). But bright is relative. When reproduced by a neutral system, there should still be more bass produced than treble.
 
All I'm trying to say (words are hard) is a downward slope is neutral when playing a recording with a downward slope. A perfectly flat system (one that reproduces all pure tones at the same volume, as we perceive it at our ear drum) will sound like it has less treble when playing back a recording because the recording itself has less treble. The problem is, what kind of headphone response is flattest at the ear drum? That's what these compensation curves are trying to figure out. Harman hasn't got it right either, I'm sure, and I'm still tweaking my EQ to see if I can make it better for me.

 
Yes that is true that is what I meant. And this is clear what you mean.
 
I will say, when you use headphone in a vacuum, you will forget eventually if it is bright or dark headphone. I used to own HD650 for a year, and listen to it exclusively. At that point you can ask me is that a bright headphone? I will say yes, some of the time. Other times, sure, it is dark. And for the DT880, same. I only notice it is bright headphone when I refer back to a HD598 or other typical Sennheiser. Otherwise, I don't think about it, it is just a background I don't notice after some time. (I think this is the true burnin).
 
This is recognizable?

 

[Head Injury]

   
Oh. lol.
 
So, just to clarify...when you call the LCD-2 / TH-900 warm and lush, you really mean neutral, accurately reproducing the warmth and lushness in recordings...right?

I mean the LCD-2 is pretty neutral. Remember, I did EQ the LCD-2 lower midrange down a bit, so it's not neutral out of the box. Just closer than bright headphones, IMO. And yes, I do think it's perfectly normal and neutral to have a little more bass than treble.
 
I don't have the TH-900, and I'll never have one. It seems very neutral from how similar to the reference track it sounds in the video I've linked, and how relatively close it is to the curve I like.
 
Just to stir the pot a little more, the Beats Solo2 is very close to that same curve. The bass boost extends a bit too high, it's supposed to slope down closer to 100 Hz but it doesn't have the dip at 500-600 Hz. So yes, I dare to say its frequency response may be more neutral than the flagships being discussed.