[warpdriver]

Quote:

Originally Posted by flecom human hearing looks more like this:

Interesting that if you take the K701 graph and numerically sum it to the ear sensivitiy chart you posted, you would actually come up with a pretty flat line except for the lowest of bass frequencies.

The human ear chart shows sensitivity emphasis in the mids which is compensated by the droop in the K701 mids

The human ear chart shows a sensitivity droop in the mid treble which is compensated by the spike in the K701 mid treble.

The human ear chart shows a sensitivity emphasis in the upper treble which is ompensated by the droop in the K701 upper treble.

This could explain why so many people think the K701 is the most neutral of many cans they've heard in the midrange and upwards.

If you did the same thing with the HD650, you'll find that the upper treble is rolled off.

As far as this chart is concerned given this ear chart, I'm not surprised at all at this graph, I find the K701 sounds almost exactly like what the chart shows.

 

[flecom]

Quote:

Originally Posted by warpdriver Interesting that if you take the K701 graph and numerically sum it to the ear sensivitiy chart you posted, you would actually come up with a pretty flat line except for the lowest of bass frequencies. The human ear chart shows sensitivity emphasis in the mids which is compensated by the droop in the K701 mids The human ear chart shows a sensitivity droop in the mid treble which is compensated by the spike in the K701 mid treble. The human ear chart shows a sensitivity emphasis in the upper treble which is ompensated by the droop in the K701 upper treble. This could explain why so many people think the K701 is the most neutral of many cans they've heard in the midrange and upwards. If you did the same thing with the HD650, you'll find that the upper treble is rolled off. As far as this chart is concerned given this ear chart, I'm not surprised at all at this graph, I find the K701 sounds almost exactly like what the chart shows.

exactly!

which is why everyone who commented that maybe headrooms testing setup was defective, its that headphone manufacturers use these graphs as refrences (a truly perfectly neutral headphone would be the exact inverse of that graph, or whatever your hearing looks like)

thats why they all have those bass humps and peaks in the highs... because your ears have those humps/peaks

 

[amartignano]

I can't understand why in hifi sometimes someone points out these uman sensitivity graphs and compare them with response of headphones.

You have to reproduce (more or less) the same frequency power across the spectrum of the reality. Stop.

To explain:

In live:
REALITY --> UMAN EAR SENSITIVITY --> BRAIN

In reproduction:
HEADPHONES --> UMAN EAR SENSITIVITY --> BRAIN

So to decide the quality of reproduction you compare headphones vs reality, and the ear sensitivity is out of the game, it comes after, it's the same in both situation, you don't have to correct this, because you are this..

So there's no reason to "equalize" the uman sensitivity, as we hear the reality with it!!! (imho, of course, but it's seem so obvious to me...)

 

[flecom]

Quote:

Originally Posted by amartignano I can't understand why in hifi sometimes someone points out these uman sensitivity graphs and compare them with response of headphones. You have to reproduce (more or less) the same frequency power across the spectrum of the reality. Stop. To explain: In live: REALITY --> UMAN EAR SENSITIVITY --> BRAIN In reproduction: HEADPHONES --> UMAN EAR SENSITIVITY --> BRAIN So to decide the quality of reproduction you compare headphones vs reality, and the ear sensitivity is out of the game, it comes after, it's the same in both situation, you don't have to correct this, because you are this.. So there's no reason to "equalize" the uman sensitivity, as we hear the reality with it!!! (imho, of course, but it's seem so obvious to me...)

I dont agree, everyone claims that the K701 and DT880s are "neutral" headphones, and if you notice their response is almost the exact opposite of those graphs... so obviously if every headphone manufacturer is doing it, its for a reason...

 

[slwiser]

Being exact opposite of those graphs is exactly the point. To field in where the actual ear has issues. In my way of thinking exact opposite is very much a positive thing.

Quote:

Originally Posted by flecom I dont agree, everyone claims that the K701 and DT880s are "neutral" headphones, and if you notice their response is almost the exact opposite of those graphs... so obviously if every headphone manufacturer is doing it, its for a reason...

 

[amartignano]

Quote:

Originally Posted by flecom I dont agree, everyone claims that the K701 and DT880s are "neutral" headphones, and if you notice their response is almost the exact opposite of those graphs... so obviously if every headphone manufacturer is doing it, its for a reason...

I expected this.

I'm not saying that every headphones has to be totally flat (for the same reason posted some posts before).

Quote:

Originally Posted by amartignano Every manufacturer has it "purpose" in the sound, so different headphones, so different principles to develop in engineering, so different measures, etc...

But I think that you don't have to correct your own ear sensitivity, because you hear also the reality with it.

Anyway, the perceiving of headphones freq resp is hard to deduce from measure. So I give measures a "very relative" importance.

Better listen, and speak about listening...

bye
Andrew

Ps. look better at the graphs: for example HD650 has a valley where ear sensitivity has a valley too (about 4 kHz), and K701 has a peak when uman sensitivity begin to raise again...

 

[JunktionFET]

Often times, the sense of "presence" from music comes from the 2-4 kHz band, which is also where our ears are generally most sensitive. If a headphone has a sharp or deep valley at those frequencies, I'd wager they probably don't sound quite as "forward" or perhaps as real as a pair with less of a valley.

Those Fletcher-Munson curves posted earlier are compensation curves. If the human ear was a microphone, its response graph would theoretically look like those curves but inverted. This BTW is why loudness controls at low volume boost the bass and treble--it is a sloppy way to compensate for this curve.

 

[Andrea]

Quote:

Originally Posted by amartignano Ps. look better at the graphs: for example HD650 has a valley where ear sensitivity has a valley too (about 4 kHz), and K701 has a peak when uman sensitivity begin to raise again...

You would want to take a look at Headroom's instructions for interpreting their measurements,

http://www.headphone.com/technical/p...eadphone-data/


Especially,

Quote:

You’ll notice all headphones have a lot of ups and downs in the high frequencies; this is normal and mostly due to reflection cancellations in the folds and ridges in the outer part of the ear. Ideally however, the ups and downs of the frequency response should average out to a flat line. Large peaks or valleys over 3kHz in width usually indicate poor headphone response, and should be viewed as a coloring of the sound. Some small dips in the highs may actually be desirable and should exist in the 2kHz to 8kHz region.

 

[amartignano]

Quote:

Originally Posted by Andrea You would want to take a look at Headroom's instructions for interpreting their measurements, http://www.headphone.com/technical/p...eadphone-data/

Always very interested in what I write, don't you?

Anyway, man, I was arguing with flecom considerations, not headroom one. Wrong address.

And I prefer to interprete my listening.

bye
Andrew

 

[Steve999]

Interesting point! If your point is that K701s are not for bassheads

, despite the as-measured (correct or not) elevated bass response, it's a point worth making and a point well-taken!

Notice that the bass peak is about 2 db higher than the treble peak in the K701 curve, and also for the HD595 curve. Thus the bass relative to the treble might be perceived as about the same -- i.e., close to perfect -- for both phones. The main difference seems to be that the HD595 midrange is more upfront, so to get more bass and treble impact you'd have to crank the volume a little more on the HD595s.

This thread is very interesting. Perhaps you could view the K701 as a headphone that would be more fulfilling at low to moderate volumes than the HD595, given the ear's lower sensitivy to these refequencies. Or, alternatively, you could view the HD595 as a headphone that would sound a little less shrill at higher volumes. Does this ring true, based on anyone's personal experience?

Anyway, this appears to be state-of-the-art design -- the manufacturers appear to be accomplishing very nearly exactly what they want to do.

Great thread!

Quote:

Originally Posted by spike33 I see your point, but take a look at this: According to my ears, the graph is accurate in the mids/highs, but bass is totally off. I'd say there's slightly punchier and deeper bass on K701 but not by that much. I find the K701 graph misleading for potential buyers. Bassheads will be disappointed with lack of bass once they hear it, and the bass averse will be turned off just by the sight of the graph (K701 is more suited for this crowd).

 

[JunktionFET]

Quote:

Originally Posted by amartignano But I think that you don't have to correct your own ear sensitivity, because you hear also the reality with it.

This is very true--a headphone or speaker system should really try to produce an acoustically flat frequency response. We can't adjust an EQ on reality, so our headphones and speakers should be true to their source. The only time a "pre-EQ'd" headphone or speaker is practical is if you only listen to one piece of music--because it will probably not be right for other pieces.

Also, the only time those Fletcher-Munson curves have meaning in this context is in comparisons of midrange and upper midrange. Our ears are most sensitive to the frequencies on that graph that dip the most.

In other words (generally speaking), we are most sensitive to 4kHz, but least sensitive to 20 Hz, 8kHz, 15kHz+, etc... Therefore, a 3db peak or dip at 4Khz is a much bigger deal sonically than a 3db peak or dip at 8kHz.

 

[Tyll Hertsens]

Quote:

Originally Posted by amartignano there's not a standard in headphones measurements, and a dummy head ear has not the same reflections properties of our "cartilagenous" ears

That's not really true. There are standards in headphone measurements, just not one single standard. There are a number of "standardized" ways of measuring headphones and the come out with different spectral results. And the Head Acoustice dummy head we use has a very specific outer ear "pinna" that is designed to have a very specific average set of reflections and acoustic absorption coefficients. Please be carefull not to just assume you know things.

 

[NotJeffBuckley]

Quote:

Originally Posted by Tyll Hertsens That's not really true. There are standards in headphone measurements, just not one single standard. There are a number of "standardized" ways of measuring headphones and the come out with different spectral results. And the Head Acoustice dummy head we use has a very specific outer ear "pinna" that is designed to have a very specific average set of reflections and acoustic absorption coefficients. Please be carefull not to just assume you know things.

I know that the earlier rounds of measurements were done to, ah, tenuous standards. In the more recent testing, have you guys been able to more accurately and consistantly compare them? In other words, are these graphs more "final word" than the others?

 

[Andrea]

Quote:

Originally Posted by NotJeffBuckley I know that the earlier rounds of measurements were done to, ah, tenuous standards.

I don't think so. They don't look like that, to me. And I've been studying them well

 

[NotJeffBuckley]

Quote:

Originally Posted by Andrea I don't think so. They don't look like that, to me. And I've been studying them well

I remember him saying as much earlier.