Quote:

skylab said:

I think the whole concept of describing "speed" of a headphone is an incredible misnomer.  It is a poorly applied substitute term for other real phenomenon.
 
Just to be clear - if a driver actually moved too slowly, or too quickly, it would make music that was off-pitch.  Sound is caused driver movements creating the compression and rarefaction of air molecules.  The exact frequencies of the music depend on those waves or air molecule motion being exact.  The movement of the driver cannot be too fast, or too slow, as if this occurred, music would not have the notes reproduced at the correct pitch.  I have not personally heard this phenomenon in a headphone, and I highly doubt that anyone else has, either.
 
You can talk about a headphone's ability to reproduce transients, sure.  But that has absolutely nothing to do with speed of the driver's movements.

Click to expand...

I think we're having a communication problem here.  You have to stop thinking sine waves, and start thinking transfer function.
Speed, or acceleration factor is exactly what we're talking about here.  We can liken this to driving a race course.  Two cars on the road.  One car is a Porsche Caymen, the other is Dodge SRT8 wagon.
 
Both cars can drive the course up to a given speed, but as speed increases, the Dodge, as fast as it is, will eventually begin to miss some of the sharper corners, overshooting some, not able to turn in quickly enough, etc, until eventually, the Dodge is left in the dust.  Neither car is bad, but one has twice the mass of the other, and it's physically impossible for the heavier vehicle to successfully negotiate the sharper turns above a certain speed.
 
So it is with headphone drivers as they reproduce complex musical waveforms.  They can all play a nice sine wave, but when called upon to accurately reproduce a transient attack of a hammer hitting a bell, the striking of a piano key, etc, you are going to hear a difference.  The lower mass driver has an easier time of moving exactly when it's told to move and stopping motion when it's told to stop.
 

My whole point was exactly that there is a terminology problem. And "speed" just isn't the right term. A Kia, and a Porsche, as you point out, can both drive at the same exact SPEED.

Quote:

odigg said:

The graphs don't tell me about the audibility of the "speed."  There must be some threshold as which we can no longer differentiate between the speed of a particular headphone and a speed that is faster than that.  Yes, the LCD2 measures much faster than most any other headphone, including the HD800.  But is the LCD2 audibly faster than the HD800?  I really doubt it.  Keep in mind that I listened to both a lot at CanJam - but I'll concede we are talking about meet conditions.
 
Is the HD800 then audibly faster than the other headphones?  I really don't know but I'm inclined to say no.  It's hard for me to believe that another decently engineered headphone (e.g. the HD555) is so slow that it would result in a audible change to the sound because the headphone cannot react to changes in the input signal when that signal is real music.
 
If the driver was so slow that it couldn't keep up with the signal, wouldn't that result in measurable, if not audible, distortion?  Yet there are a lot of headphones (very cheap ones too) that have distortion at a level that is inaudible.

Click to expand...

I promised I would pick this up again, so let's give it a go...
 
[size=13pt]In Tyll's measurements, he uses two different frequency square waves, one at 30 Hz and the other at 300 Hz.  In any measurement system, distortion occurs when the output waveform shape deviates in any way from input waveform.[/size]
 
[size=13pt]It's easy to observe sine waves on a scope and to see any distortions or deviations from that simple waveform, but a sine wave bears little resemblance to a complex musical waveform, and is very non demanding of driver performance.  Trouble is, complex musical waveforms are, well, complex.  Too much going on to be able to clearly identify distortion products.  [/size]
 
[size=13pt]This is where the square wave comes in.  The shape of a square wave is simple and easy to observe, yet brutally difficult for a speaker / headphone driver to reproduce accurately.  Let's talk about why a square wave is so difficult for a headphone driver to reproduce.[/size]
 
[size=13pt]The signal/waveform straddles 0 output, and instantly changes from 0 to either a positive or negative value, held for 50% of the time, and then instantly transitions to the opposite value, passing again through 0 on the way. Just imagine the stress on the driver... The amplifier demands the driver go from an off state to instantly extended in one direction or the other. With virtually massless electrons it's pretty easy for an amp to do this at audio frequencies, but for the transducer that has real mass and has to move the mass of the air out of the way also. Then, just as the driver is moving on it way to the positive state, the amp tells it to stop instantly in it's tracks and hold that position for 50% of the time and just as it settles into that position the amp tells it to instantly move to it's full negative position, and back again, and so forth.[/size]
 
[size=13pt]What happens to mass at rest? I tends to stay at rest. Just as it gets moving, it's told to stop and soon as it stops, it's told to move all the way to the negative position. A mass in motion, tends to stay in motion. The greater the mass, relative to the motor strength which moves the driver, the harder time it has of responding to the control signals from the amplifier. [/size]
 
[size=13pt]So the perfect audio driver has no mass and infinite motor strength. This allows it to navigate reproduction of the complex musical waveform with perfect faithfulness.[/size]
 
[size=13pt]A square wave is a unique way to represent the extreme challenges of complex musical waveforms in a simple way, so we can observe what the driver under test is doing.[/size]
 
[size=13pt]Consider this graph from the LCD2;[/size]
 
As you can see, it resembles a square wave as it should. When the amplified tells the driver to go from 0 to 1, it does so pretty quickly, but as you can see, not perfectly.
[size=medium][/size]
 
You'll notice at the vertical leading edge of all the positive and negative going wave, there is some overshoot, that is, the driver actually move past the position it has been told t go to, and then quickly corrects itself.  
 
Let's look at another from the ADH-AT2k;

Still recognizable, but more overshoot, more ringing, etc.
 
Here's another from Koss KDE 250
 

 
As you can see, the original perfectly square and symmetrical waveform is becoming less and less recognizable.  Imagine what this does to your music.
 
Here's the same plot from the HD800;

This plot looks much, much better than the previous ones except for that of the LCD-2.  At least the square waves look like square waves in the 300Hz plot.  In the 30 Hz plot, you can see goo fast rise times, some over shoot, but then rather than a flat top on the wave, it slopes down.  The is from the headphone's lack of ability to maintain positive pressure on the ear when called to do so.  It correlates to impact and punch that you would hear in your music.  The flatter the 30Hz top, the better the impact and punch.
 
Check out the Qualia plot;

I'm sorry, this is a bad headphone.  Completely uncontrolled.
 
More later...

Quote:

kwkarth said:

They can all play a nice sine wave, but when called upon to accurately reproduce a transient attack of a hammer hitting a bell, the striking of a piano key, etc, you are going to hear a difference.  The lower mass driver has an easier time of moving exactly when it's told to move and stopping motion when it's told to stop.
 

Click to expand...

I'm mostly a layman who learns as he goes along, mostly from seeing others' posts and Wikiing (that's based on the form of "skiing") things. I've been assuming that this is what distortion measurements tell us, right -- like how much the headphones get in the way of the source signal?
 
Also, I've assumed that the continuity in the square waves tells the same story --- if it doesn't wiggle much it implies good control, or "speed"?
 
--Haha, I posted just after you, KwKarth, so my statement is rendered unnecessary and thoroughly outclassed. It's wonderful to have my suspicions confirmed. Looking forward to "later"

Quote:

skylab said:

My whole point was exactly that there is a terminology problem. And "speed" just isn't the right term. A Kia, and a Porsche, as you point out, can both drive at the same exact SPEED.

Click to expand...

Only up to a point, then the lower mass, better suspension, better tired car will be able to negotiate the corners better.

Quote:

sambones said:

I'm mostly a layman who learns as he goes along, mostly from seeing others' posts and Wikiing (that's based on the form of "skiing") things. I've been assuming that this is what distortion measurements tell us, right -- like how much the headphones get in the way of the source signal?
 
Also, I've assumed that the continuity in the square waves tells the same story --- if it doesn't wiggle much it implies good control, or "speed"?

Click to expand...

Basically, yes, it implies a lot of things, bottom line, faithfulness to your music.
 

Quote:

kwkarth said:

Only up to a point, then the lower mass, better suspension, better tired car will be able to negotiate the corners better.

Click to expand...

That's steering and handling, buddy - not speed.  The truth is, the speed of a car is most analogous to the frequency response.  The Kia has a rolled-off top end compared to the Porsche   But think of each MPH as being some frequency in Hz.  It HAS to be exact - 50 MPH either is, or isn't.  50 Hz either is or isn't.  Slightly off, and you are off pitch.
 
Impulse response - that's handling, in the car analogy.
 
Anyway, we're arguing (in a fun way BTW) semantics and linguistics, not science.  So I will stop.  But I still maintain "Speed" is the wrong term for the phenomenon people are trying to ascribe to it,

Nonetheless, speed is an accepted audio term, and while it definitely doesn't mean the same thing it does for cars, it has a standard meaning when it comes to audio...but of course, I waited till you said you wouldn't argue it anymore to chime in.
 
You are still right, though, that audio speed is closest to car handling. But car handling can turn into faster track times, and therefore overall speed, as per KwKarth's example.

I don't own a Qualia and never plan to, but I'm pretty sure that it wasn't fitted optimally on the dummy head.  The Qualia is the only headphone I've ever auditioned that is as finicky about fit as the Stax O2; it can sound drastically different depending on angle and placement of the drivers relative to the ears.  The Qualia, when properly fitted, is also one of the better performing dynamic headphones to these ears.

Quote:

n3rdling said:

I don't own a Qualia and never plan to, but I'm pretty sure that it wasn't fitted optimally on the dummy head.  The Qualia is the only headphone I've ever auditioned that is as finicky about fit as the Stax O2; it can sound drastically different depending on angle and placement of the drivers relative to the ears.  The Qualia, when properly fitted, is also one of the better performing dynamic headphones to these ears.

Click to expand...

Nope, I don't think so. I've seen other measurements on the web that looked similar.
 
 
 
edit:
About square waves, while those measurements are fine, square waves are not what real music looks like if you know what I mean.

When I wore the Qualia in a "traditional" sense (headband running vertical and on the top of the head) there was an annoying echo that made them sound a bit like playing music in a cave - it wasn't until I tilted the headband forward that everything snapped into place.  I'd imagine most people place headphones in the traditional position when they measure, but Tyll will have to comment on how the Qualia was placed if he remembers.

Quote:

skylab said:

That's steering and handling, buddy - not speed.  The truth is, the speed of a car is most analogous to the frequency response.  The Kia has a rolled-off top end compared to the Porsche

  But think of each MPH as being some frequency in Hz.  It HAS to be exact - 50 MPH either is, or isn't.  50 Hz either is or isn't.  Slightly off, and you are off pitch.
 
Impulse response - that's handling, in the car analogy.
 
Anyway, we're arguing (in a fun way BTW) semantics and linguistics, not science.  So I will stop.  But I still maintain "Speed" is the wrong term for the phenomenon people are trying to ascribe to it,

Click to expand...

I take it you've never been on the track.  Mass has everything to do with things along with steering geometry, tire grip, yada, yada.   Impulse response is the ability to respond to the impulse, moving mass and motor strength are the enablers and inhibitors.
 
Sky, how about "nimbleness?"  Does that work for you?
 

Quote:

n3rdling said:

I don't own a Qualia and never plan to, but I'm pretty sure that it wasn't fitted optimally on the dummy head.  The Qualia is the only headphone I've ever auditioned that is as finicky about fit as the Stax O2; it can sound drastically different depending on angle and placement of the drivers relative to the ears.  The Qualia, when properly fitted, is also one of the better performing dynamic headphones to these ears.

Click to expand...

You may be absolutely right.  The most important thing to know and understand about these tests is the inherent and possible limitations thereof.  Yup, I agree the Qualia needs a do-over.

Quote:

xnor said:

Nope, I don't think so. I've seen other measurements on the web that looked similar.
 
 
 
edit:
About square waves, while those measurements are fine, square waves are not what real music looks like if you know what I mean.

Click to expand...

Please read my explanation about how square waves are relevant.
 

I read it kwkarth, looks fine. But to expand on that car analogy it's like driving a car with max. speed into a 90° curve - an extreme challenge.