[Head Injury]

Quote:

 


 

No, the speaker is not still turned on 

 
Crosstalk is undesirable and electronic, and is not delayed. Crossfeed is the intentional mixing of delayed signals to eliminate channel separation. If these headphones are designed to work like this, it is indeed crossfeed.
 
My point is that the effect is 1. Too weak to be meaningful, and to build an entire theory of soundstage upon, and 2. Too dependent on playback volume and ambient volume to be associated directly to soundstage performance, or compared between headphones.
 
I have pointed out flaws in your reasoning and logic. The effect is too weak to be meaningful, from my perspective. It is far, far weaker than the effect the recording itself has, and seems to pale in comparison to the effect that driver placement has. And because there is not necessarily a correlation between headphone isolation and sound bleeding between channels (as I indicated with the Grados and LCD-2) there isn't necessarily a correlation between this proposed crossfeed effect and isolation, as I believe you originally implied.
 
And of course if it were significant, it is a severely unnatural effect and I doubt would improve imaging in any way. Speaker crossfeed is natural. Crossfeed plugins are designed to be as natural as possible. Headphone crossfeed in this example would have too low volume and too long delay to be able to accurately mimic real life.
 
At this point I expect proof that it is significant and "perceptible by a perfectly healthy individual"

 

[NA Blur]

I have been wondering the same thing and in my brief investigation I found that it is dependant upon the mid to treble frequency curve.  My Brainwavz M2's have a large soundstage so the size of the driver has nothing to do with it.  My DT990's also have a very large soundstage so it has nothing to do with sealing the ear / ear canal.  Somewhere between 1kHz and well into the 8kHz range there appears to be frequencies that allow the brain to process location and distance.
 
With this in mind there is no doubt in my mind that you can measure it especially if it occurs well above listening threshold.
 
I found this very helpful:
http://www.kemt.fei.tuke.sk/Predmety/KEMT320_EA/_web/Online_Course_on_Acoustics/hearing.html

 

[SP Wild]

Quote:

No, the speaker is not still turned on 

 
Crosstalk is undesirable and electronic, and is not delayed. Crossfeed is the intentional mixing of delayed signals to eliminate channel separation. If these headphones are designed to work like this, it is indeed crossfeed.
 
My point is that the effect is 1. Too weak to be meaningful, and to build an entire theory of soundstage upon (Show me where I built an entire theory of soundstage) , and 2. Too dependent on playback volume and ambient volume (as are all issues regarding audio fidelity) to be associated directly to (subjective) soundstage performance, or compared between headphones.
 
I have pointed out flaws in your reasoning and logic (with flawed counterpoints). The effect is too weak to be meaningful, from my perspective (a subjective one). It is far, far weaker than the effect the recording itself has (agreed), and seems to pale in comparison to the effect that driver placement has (debatable). And because there is not necessarily a correlation between headphone isolation and sound bleeding between channels (as I indicated with the Grados and LCD-2) there isn't necessarily a correlation between this proposed crossfeed effect and isolation, as I believe you originally implied.
 
And of course if it were significant, it is a severely unnatural effect and I doubt would improve imaging in any way. Speaker crossfeed is natural. Crossfeed plugins are designed to be as natural as possible. Headphone crossfeed in this example would have too low volume and too long delay to be able to accurately mimic real life.
 
I agree with you on your last point.  As for the rest of your argument you can contact AKG for clarification on design principles of their discontinued K1000 headphone.
 
At this point I expect proof that it is significant and "perceptible by a perfectly healthy individual"
 
Seek and you shall find.  I suggest starting at AKG, for which I have supplied you with their contact details.

 

 

[Head Injury]

The K1000 is an exception because it basically is a speaker.
 
Please provide evidence for a normal headphone, like the ones you suggested would be affected (HD800 and LCD-2 for example).

 

[SP Wild]

"too long delay"    check
"sound leakage"    check
"poor isolation"    check
 
The moment there is a headband it is no longer a speaker, it is a headphone.
 
 
 
 
 
Edit:  Checkmate.
 

 

[Head Injury]

What?
 
"too long delay" - No, not when the speakers are positioned forward. Unlike headphones, the cups aren't pointing out horizontally, so the sound doesn't have to cross the whole head, just on a diagonal.
 
"sound leakage" - I'm not sure what your point here is. The more sound leakage, the more effective your proposed crossfeed. Sound leakage is a good thing. Speakers are good specifically because they have tons of sound leakage. They are sound leakage.
 
"poor isolation" - Again, this is a good thing. You said it yourself. It's only a bad thing when ambient volume is so high that the apparent difference in volume between one channel and the background noise shrinks to nothing. Speakers don't have this problem because both speakers play at each ear at nearly the same volume. Headphones do have this problem because they don't. The K1000, because of their sound leakage and position, will sound louder at the opposite ear than a normal headphone.
 
You seem to be missing my point, or you're not trying to understand it.

 

[deadlylover]

I just cut one of the channels of my O2's and lambda's, both having fairly significant differences in soundstage size and I just can't hear any of that leaky isolation crossfeed effect with my other ear. The effect is pretty much inaudible for me.

 

[Head Injury]

I can hear it with the LCD-2 (like I said, they play LOUD on the outside), but only when I mute the right channel and cover my right ear. The pads are lifted up somewhat, so I'm hearing leaking from both sides, which is increasing the effect. I don't expect it to be making a big difference during playback when both ears are hearing things. High stereo separation is as hellish as with Grados or IEMs.
 
The issue isn't the audibility, which under some conditions is definitely possible. The issue is what effect, if any, it has on imaging and soundstage.
 
SP Wild, I drew you some awesome images to illustrate my point:
 
Speakers

 
Headphones

 
K1000

 
Notice how that because the K1000's drivers are so far forward, the difference between y and x is much smaller than with headphones, where the sound has to travel not only across the whole head but around it as well. Obviously it's not perfectly to scale.

 

[deadlylover]

Hmm, how about varying our listening volume to see what effect this thing has? If I set my volume where the leakage is very low, it's not like the soundstage collapses or opens up, it's pretty much stable as a rock throughout any volume on my cans. And I just noticed the dispersion pattern on my headphone drivers, it's difficult to hear the effect as it's very directed(?), especially the high frequencies.
 
I didn't catch it from the last few posts but what are we saying with this effect? More = less soundstage 'width' like any other crossfeed?

 

[Head Injury]

Quote:

Hmm, how about varying our listening volume to see what effect this thing has? If I set my volume where the leakage is very low, it's not like the soundstage collapses or opens up, it's pretty much stable as a rock throughout any volume on my cans. And I just noticed the dispersion pattern on my headphone drivers, it's difficult to hear the effect as it's very directed(?), especially the high frequencies.
 
I didn't catch it from the last few posts but what are we saying with this effect? More = less soundstage 'width' like any other crossfeed?

SP Wild seems to be suggesting that more leakage from one ear to the other improves imaging, by simulating a (weak) version of stereo crossfeed.

 

[kiteki]

 
I think first you would need a true 3D soundstage x/y/z graph, something like this video except in 3D, you'd also need microphones which can pick up the way we perceive headphone or IEM soundstage, microphones which don't exist, afaik.
 
Just my theory.
 

 
 

 

[Lorspeaker]

Tot I read somewhere the shape ofyour outer ear also affect soundstage sensing.....
a hyaena hears differently from an elephant...for example...I can proof it. :rolleyes:

 

[kiteki]

 
Well, I think the important part is your can't measure the soundspace of the UE18Pro, which will sound considerably different to the Etymotic ER-4.
 
So, which one is more accurate?