[John In Cali]

I think the answer to this this is pretty obvious.  The sounds we hear out of a headphone are because of movement of a driver.  Bass notes and treble notes are the same to a driver, just electrons flowing through the driver creating a magnetic force that either is repelled from or brought towards another magnetic field, usually a magnet. How good a driver is at copying a sine wave is how easily it will change its direction or speed without resistance.  Naturally lighter drivers are better at this(why electrostats are detailed), a driver doesn't know if it is getting a treble note or bass, it is just moving however the magnetic field wants it to.

Sorry for the crude MS Paint drawing.
A driver getting the start of the green line does not care about the rest of the line whether it follows the blue or green, it is just being pushed away from the magnet, if it can follow that path then it can follow the treble path or bass path with the same detail.  Also yes the srh 940 bass detail is quite spectactular, it had me going over the same 10 seconds of a song, amazed at what seemed to be a simple lone bass note with all my other headphones, now a complex melody of a changing bass note.

 

[Tilpo]

That makes me thinking. What if someone tried to make a headphone with an electromagnet instead of a permanent magnet?
It might take some trouble to get an as uniform and strong field as with using neodymium, but the rewards could be lower driver weight. Power requirements of the headphone could rise dramatically on the other hand, and a separate high current power supply for the electromagnets could be needed.

 

[John In Cali]

Quote:

That makes me thinking. What if someone tried to make a headphone with an electromagnet instead of a permanent magnet?
It might take some trouble to get an as uniform and strong field as with using neodymium, but the rewards could be lower driver weight. Power requirements of the headphone could rise dramatically on the other hand, and a separate high current power supply for the electromagnets could be needed.

That is what electrostats are...
 

 

[Tilpo]

That is what electrostats are...
 

I'm talking about dynamic headphones using electromagnets instead of neodynium.

 

[k00zk0]

Semantics...
 
Yes technically the fast transients in a "detailed" bass tone is treble. The early posters meant bass such as the low organ note posted above, something that subjectively sounds bassy but of course has elements of higher frequencies extending to inaudibility. As square waves don't appear in nature, rarely there are plain sine waves also. Most everything has higher freq elements which contribute to the way the "low bassy noise" (subjective description) sounds like.
 
There is a growing number of worthy, completely organic sounding, yet synthesized works. Square waves DO exist in nature. They come out of our naturally man made machines, a natural product of our natural advancement. The question is whether a 40hz square wave is considered a 40hz sound, or a number of 22.5khz (or whatever the sampling rate allows) quick treble tones with pressurized silence in between. Again, semantics. Well, a spectrogram shows that it's actually a bunch of sine waves added together, as an earlier post showed. That's because spectrograms display the presence of sines within a signal. If it were looking for squares, and somehow squares were what we defined everything by, such as in a digital reality, a plain sine wave would be termed a most complicated waveform.
 
The treble quality of the headphones most definitely defines the quality of these complicated "bass tones" (again, the earlier implied description)
 
Look at the waveform of popular electronic tracks and you will see year to year, there are more "artificial" impulses, triangles and intentional zero-crossings added in specific patterns, sometimes hidden on top of normal recordings. There's definitely a whole world of experiences to be discovered hidden within specially-crafted synthesized sound not made yet.
 
Edit: In terms of the electromagnet idea, wouldn't the diaphragm's coil induce changes in the electromagnet's field, requiring active correction of some form anyway?
 
Correct me if I'm wrong but the only reason not to pump 10 watts for microseconds into a driver (with wire that can handle it and enough heat dissipation) to emulate these transients perfectly is because it will overshoot the mark and ring horribly. Where are the actively corrected headphones? It only needs a little sensor by the edge of each diaphragm.

 

[Joe Bloggs]

I would just note that human hearing is much more refined in the frequency domain than the time domain (as evidenced by the imperceptible effect of accurate or inaccurate phasing in things from loudspeaker crossovers to EQ), and overemphasis of one part of the frequency spectrum tends to mask other frequencies in the spectrum, such that a balanced frequency response contributes much more towards the perception of "detail" than time domain performance such as transient response (as evidenced by the messed up transient response of the "detailed" LCD-3)

 
That said, a sharp square wave or dirac impulse is composed as much of bass frequencies as treble frequencies, so you are just as apt to lose that sharp edge whether you take bass or treble frequencies out of the picture.  Or if any part of the spectrum is overemphasized (such as the 18kHz in the LCD-3).  So a balanced FR is again necessary (if not sufficient) for good transient response.  Sufficient (or "good", whatever that means) treble is certainly part of that necessity, but so is bass, and everything in between.
 
I notice that, once I even out the humps and dips in the bass frequency response of my loudspeaker setup with a parametric equalizer, the bass sounds much tighter, even though the atrocious resonance acoustics of my untreated (and cluttered) room are still there.  And many phones sound tighter and more detailed when you equalize down a common midbass hump from 100-200Hz, even though the resonances that created that hump are still there.  (that, or this part of the spectrum just seems to cloud over other parts of the spectrum more easily to my ears)
 
And when you said the SRH940 has better treble despite equalizing the HD650 and SRH940 to sound similar, do you mean with a graphic EQ (with however many bands) or a parametric EQ, evening out the many resonant high frequencies as detailed in this tutorial?
http://www.head-fi.org/t/413900/how-to-equalize-your-headphones-a-tutorial
 
After following the tutorial I find myself preferring a pair of $10 canalphones (Philips SHE3580) to $200 ones (Etymotic ER-4P) for everything from bass to treble to my *perception* of "transient response", despite the $10 phones having many more flaws to be corrected in the process--its overwhelming advantage is that it (with its reasonably-sized dynamic drivers and non-ported design not sacrificing low bass for midbass) does everything from 20Hz to 20kHz, which the Etys (with its first generation BA drivers) sadly don't.  (well that, and the overwhelming comfort and cheapness of the $10 canalphones, such that I can take them anywhere and everywhere without fear or losing or breaking them or making my ears hurt)

 

[ac500]

I really doubt equalization is everything. Although certainly a flat line would be a good thing, there are characteristics along the frequency spectrum more than just the amplitude, and they matter a lot.
 
To be fair I haven't equalized my HD650 with any special equalizer but my sound card's built-in one (which works pretty well) since I never got the other software to work. But, when I make it sound very close to the SRH940, it still lacks the 940's incredible crispness - there are still subtleties in complex electronic music the HD650 misses. I like to use electronic music as a test because it's usually quite complex, versus more organic things like classical music. Sure, a detailed headphone makes classical music and female vocals sound incredible (as they do on the SRH940), but it's much easier to judge on electronic music because you can objectively see if you can hear a certain effect or layer of the music (which slower / less detailed headphones tend to muddle over).
 
Also, despite the SRH940 being more detailed, the HD650 is much superior in other ways. For example if you equalize the bass of the SRH940 up to crazy levels, the waveform saturates (in other words, it starts to get muddy and loses half the sound quality while bass is playing). The HD650 on the other hand can be turned up as far as possible at extreme volumes, rattling your head, and the bass is still perfectly defined and smooth.
 
So basically I think there are many details to headphones beyond FR even though FR helps. While I think detail is inherently entangled with treble quality, I do not think recessed treble quality necessitates less detail.

 

[Joe Bloggs]

I never said that equalization (or rather, frequency response) is everything.  What I'm saying is that it matters a lot.  Let's say I take a pair of muddy headphones with bad FR and equalize them (with parametric EQ, not graphic EQ) to have good FR and take a good detailed headphone and equalize them to have the FR of the muddy headphones.  I bet the equalized muddy headphones would sound more detailed than the "de-equalized" good headphones, even though the latter should still have superior time domain characteristics.

 

[awinn17]

(I was going to PM this to ac500 but decided to just post it for review, contemplation, or humiliation)
 
 
 
 
Hi- I was skimming and found this thread. I read the first page, and then it started drifting off to equilization and stuff I can't help with. BUT...
 
As an engineering major I've taken math classes like this and had mechatronics labs specifically using frequency and MATLAB and representing harmonics.
 
But is not treble just the ability to move a driver at a higher frequency? The clarity of sound is the response rate for playback- not the highest frequency. A recording at 4 kbps will sound worse than a recording at 16 kbps. Likewise, a recording recorded at 32 kbps but played at 4 kbps will also sound bad. Neither of these require treble.
 
The Nyquist Citeria can be used to determine the correct sampling rate (recording kbps) so as to catch the smallest detail- the smallest cycle, the highest frequency accurately.
 
To return- Treble is the representation of this highest frequency. I believe you are thinking of the speaker's maximum playback response rate- how fast it can read bit to bit (which will indeed influence treble but not require it). The most this can do for bass is change how quickly a bass note can change- not change the quality of the bass note. So- you're right, but it's not really treble that's influencing bass so much as playback response rate (I think is what it's called..)
 
In short-short: To play high frequency representation (treble) requires a speaker with high playback response rate. A high frequency response rate does not dictate what it's playing. but assuming the sound was recorded above the Nyquist Criterion, the higher the playback response rate, the more accurate sound reproduction. WHEW
 
I am no expert. I just regurgitated this from my least favorite lab last year.
 
Thoughts?

 

[Joe Bloggs]

Just one--you confused recording bitrate (eg 128kbps vs 320kbps mp3) with sampling rate (which can be any arbitrary value for the 128kbps or 320kbps mp3 respectively, be it 44.1kHz, 96kHz...

 

[awinn17]

Quote:

Just one--you confused recording bitrate (eg 128kbps vs 320kbps mp3) with sampling rate (which can be any arbitrary value for the 128kbps or 320kbps mp3 respectively, be it 44.1kHz, 96kHz...

True, I did. But the argument is still valid if you correct the terminology because my point was that quality of playback is based on how quickly (and accurately of course, but just assume 100% for now) data can be interpreted and played back by a certain speaker. Using kHz that means how high a frequency can it play without over extending range. Using kbps it is saying how many data points it can play without aliasing. They're intimately related of course. But aliasing is where quality comes from- no aliasing means music is played as recorded (again, assuming the speaker is 100% accurate across all frequencies).
 
I believe the OP said that treble is where quality came from, or at least had an influence on it. Totally different thing- treble just defines how high a speaker can go. Clarity of treble, however, at least data wise, means that it can play the "recording bitrate" (thanks Joe) bit for bit assuming the speaker can range high enough (which is a function of it's mechanics). I wish I had a whiteboard.
 
It's too early for this. Just woke up. Late night...
 
Here's another thought- I would say that bad mechanical design can muddy and frequency or internal natural frequencies can distort or possibly mute or enhance certain ranges, but that's a totally different topic.

 

[Joe Bloggs]

The way a low lossy compression bitrate screws up the audio data and the way a bad headphone driver design screws up the audio data are very different and not comparable...

 

[ac500]

Yeah bitrate really has nothing to do with sampling rate, at all. Bitrate is all about the encoding method used, and I've never really seen it mentioned outside of lossy compression formats like MP3/OGG. In theory, the bitrate tells you something about the fundamental information conveyed in the sound. This is a kind of complicated topic and there are entire careers devoted to the study of information theory, entropic coding, data compression algorithms, perception based lossy compression, etc. etc. I know more about photo/video compression than I do audio, but as I understand it they're very very similar. In short, bitrate has nothing to do with sampling rate or treble.
 
In any case, let's assume for the sake of this argument you have an infinitely perfect input signal, that is, infinite number of samples per second. This is so we can talk exclusively about the headphone itself, and not your source.
 
You say treble "defines how high a speaker can go". I disagree completely. If you define the quality or quantity of treble as the quoted frequency range, you'll see how ridiculous and wrong that definition is. Treble is simply a word referring to a specific frequency range of a headphone, typically above 9khz. The actual qualities or quantities of the treble is a complex matter of its own.
 
Saying a speaker "has" treble tell you nothing. Instead, you can analyze the FR response, and that tells you something about the magnitudes of various treble frequencies. This still isn't the point of my theory. My theory is that treble QUALITY = detailed sound. As far as my example, I showed (and can prove mathematically) that removing treble entirely results in a undetailed blobby waveform, and therefore from this fact you may make the reasonable guess that if treble presence is required for detailed lower frequencies, then it follows that the quality of such treble also effects the quality of the lower frequency sounding waveforms.
 
In theory high frequency doesn't effect low frequency sound, but rather what we think of low frequencies aren't always mathematically correct low frequencies. A deep bass guitar believe it or not has many octaves and overtones that are VITAL to how it sounds -- in fact these overtones precisely define its very character that makes it sound like the detailed vibrating string you hear in real life.
 
Like I said, this is actually not so hard to prove mathematically as long as we define mathematically what it means to be detailed and what it means to have good treble. My conjecture is that once we define "quality treble" and "detailed sound reproduction", it will be relatively easy to show that both mathematical descriptions are of the same exact thing.

 

[Armaegis]

Repost from the last page because it might have been missed... here's a low organ note:
 
(click to enlarge)

 
notice the "treble" within the larger bass waveform

 

[ac500]

Also note that in the case of the organ note, if you have a lot of quantity of treble without quality, the fine details would be scrambled or incorrect, and it wouldn't sound really detailed. It would result in "bad" or "grainy" or "undetailed" treble sound.
 
For quality treble, even if slightly recessed, all the little details would be accurately represented, and never skipped or blurred over. Even if treble is recessed, the fine details seen in that waveform might be lesser in magnitude, but they'd still all be there.