[Trae]

Well, this is more of a discovery on my part, and I wanted to share my findings.
 
I'm using Foobar2K with a polarity inversion dsp, and the resulting sound is pretty different. Headphones are AD2000Xs.
 

Spoiler: Illustrations

 
Overall, the sound is wider and more out of your head sounding. On some songs, there are some sounds that sound the same regardless of the polarity (mainly treble-centric atmospheric sounds). It sounds really unique, and I think I like it more than a normal polarity, especially in the bass. It also emphasizes the separation between the channels. For example, if a particular sound is closer to the right channel than the left, it will sound like it's on the right channel even more with the polarity inverted. With the polarity normal, there is less distinction, and actually sounds more in the center on said sounds.
 
 
VST Plugin is called Polarity
 

 
Off does nothing. L inverts the left channel, and R inverts the right channel. Both L and R sound the same, but L+R sounds the same as off.
 
Download Here (bottom of page where DualPanner is at).
 
You'll need a vst plugin component for foobar in order to use it. I use this one.
 
Give it a try! It sounds pretty neat, imo.

 

[Speedskater]

Polarity is a better word choice than Phase.  Phase can vary with frequency but with polarity there are only two choices. Yes flipping the Polarity of either channel should make a vast change in sound.  Long ago at Radio Shack, speakers often had that strange sound, they were not very good at hooking them up. Most of the time, flipping both channels, should sound the same either way.  But a few recordings and/or a few headphones are so asymmetrical that there could be a interesting difference.

 

[castleofargh]

to be clear, it's about inverted polarity on one side. as you mention, changing the polarity of both sides doesn't really matter and manufacturers often do it on headphones, that's how much they find it unimportant. 
now on only one side it indeed makes for some weird positioning feelings. I tried that when I first got custom IEMs by turning around the cable plug on one side. a vst is a lot better for testing thanks for the link I'll try it later.
I remember that I loved bass inverted but not the rest, and thought that multi BA IEMs should have the low frequency driver inverted for funky soundstage feeling(but maybe some already do that?).

 

[Trae]

Right now, I love the bass of a reversed polarity, but the rest of the frequency range does take some getting used to. I wouldn't call either one better than the other, just different. I tried this with my K7XXs that just came in, and it too behaves the same.
 
I guess a simpler way of explaining the sound is with a normal polarity, it sounds like the microphone is inside your head, whereas with the polarity reversed, it sounds like the mic is divided in two and are positioned beside your ears. So that got me thinking. I wonder what binaural music sounds like now? Not too good. It sounds like there is some extra reverb introduced and it messes up the sound.

 

[ProtegeManiac]

Anyone have any test CDs? It doesn't matter if it's car audio, home audio, or a binaural headphone demo disc, there's bound to be some kind of phase test there that goes:

"My voice is in phase, and should be coming from a clear point in the center of the soundstage;
My voice is now out of phase, the imaging should be vague/hard to pinpoint in the soundstage." (this part usually sounds left-towards the floor on my set-ups)

For the heck of it I tried this polarity plug in. Reversing left pulled the image towards the left: vocals are off center and seemed to be coming only from the left channel (like, I can hear the trailing end of lyrics coming from the left driver), electric guitars normally imaged in a wide swath along center coming only from the left, keyboard on the right coming from somwhere behind me. Similar story when right channel is inverted, but what went left went right. 

 

[Trae]

I have some ultrasone test CD somewhere, but I don't think it did any polarity tests. I guess you could use audacity to flip one of the channels and A/B it that way.
 
It doesn't matter what channel is flipped. If the sound was made to be on the left channel, then it will stay on the left channel regardless of the phase. The only difference is that now the channel differences will be more distinct when the polarity is inverted. It'll sound like it's on the edge of the channel, if that makes sense. Now, if you were to reverse the L/R channels (as in, the left channel will play right channel and vice-versa), then yeah, but polarity has nothing to do with it. Then again, maybe some headphones behave differently to polarity changes.
 
Maybe someone could help me out with this:
 
Planar headphones have a pretty flat phase through the frequencies, whereas on your typical dynamic driver, the phase is up a couple degrees in the bass, down a couple degrees in the mids, then swings back up exponentially, resonances aside. Here are some pics describing what I'm talking about:
 
Source

Source

 
 
I wonder how an inverted polarity would affect it, or are the things I'm talking about not related to each other?

 

[ProtegeManiac]

  It doesn't matter what channel is flipped. If the sound was made to be on the left channel, then it will stay on the left channel regardless of the phase. The only difference is that now the channel differences will be more distinct when the polarity is inverted. It'll sound like it's on the edge of the channel, if that makes sense. Now, if you were to reverse the L/R channels (as in, the left channel will play right channel and vice-versa), then yeah, but polarity has nothing to do with it.
 
 
 

 
No I didn't reverse the L-R channels, I clicked L then R in the interface you linked (quoted below) and claimed that L inverts the polarity in L and R inverts polarity in R. Effect is just as I described. Kind of like "My voice is out of phase, and should be harder to locate in the soundstage" on a test disc, but on actual music. Vocals moved off-center, just not as badly as on the test track, but other instruments shifted left or right but not in a more distinct stereo separation, but somewhat confused. The electric guitar that sounds like it's a wide swath in the center moved to the left, but on some parts it still sounds like it's in the center, but with a weird tone; when I do hear it "on the left" it's not like a usual recording where it's placed on the left, but as a note fades in or out it moves towards the left.
 
Quote:

   

 
Off does nothing. L inverts the left channel, and R inverts the right channel. Both L and R sound the same, but L+R sounds the same as off.

 

[Trae]

   
No I didn't reverse the L-R channels, I clicked L then R in the interface you linked (quoted below) and claimed that L inverts the polarity in L and R inverts polarity in R. Effect is just as I described. Kind of like "My voice is out of phase, and should be harder to locate in the soundstage" on a test disc, but on actual music.

 
I understand what you're saying, but clicking L and then R won't change the sound between the two. You said that when you inverted the right channel after inverting the left channel, what went left went right. What I was saying was that there shouldn't be a difference between L and R when selecting the two.
 
Edit: your results from the polarity inversion do match mine, though.

 

[ProtegeManiac]

   
I understand what you're saying, but clicking L and then R won't change the sound between the two. You said that when you inverted the right channel after inverting the left channel, what went left went right. What I was saying was that there shouldn't be a difference between L and R when selecting the two.

 
Just tested it again - some tracks move left, some tracks move right. In other tracks some instruments moved left when left is inverted, in other tracks it seemed like it's coming more from the right but it's too indistinct (and vice versa). It could be on those tracks that I thought it was moving, but likely clicking L or R had some suggestive powers on my brain in dealing with hard to locate sound sources.

 

[Trae]

   
Just tested it again - some tracks move left, some tracks move right. In other tracks some instruments moved left when left is inverted, in other tracks it seemed like it's coming more from the right but it's too indistinct (and vice versa). It could be on those tracks that I thought it was moving, but likely clicking L or R had some suggestive powers on my brain in dealing with hard to locate sound sources.

 
Suggestive powers it is

I was in the same boat when I first found out about this, like why is everything being shifted to the right channel? Then, I swapped the channels around and found out that nothing has changed, channel-wise, but the inverted polarity makes it much more distinct. If you're running Foobar2k, use the reverse stereo channels DSP while playing around with the polarity and you'll see (well, hear) what I'm talking about. 

 

[jnorris]

Inverting the polarity of one channel is like swapping + and - on a speaker connection. Bass becomes diffuse and placement of instruments within the soundfield becomes less specific. Also, some people (like me) find the different pressures on my left and right ears uncomfortable. Inverting the polarity of both channels is helpful if one of the components in the audio chain also inverts the signal. I can't imagine how inverting one channel would be an improvement.

 

[Trae]

Well, I guess I'm done playing with polarity now. An inverted polarity doesn't sound right. Who'd have thunk it

Maybe I'll snatch my friend's beats headphones and swap the wires around. Then again, I wouldn't be surprised if they were wired backwards to begin with

 

 

[SilverEars]

  I have some ultrasone test CD somewhere, but I don't think it did any polarity tests. I guess you could use audacity to flip one of the channels and A/B it that way.
 
It doesn't matter what channel is flipped. If the sound was made to be on the left channel, then it will stay on the left channel regardless of the phase. The only difference is that now the channel differences will be more distinct when the polarity is inverted. It'll sound like it's on the edge of the channel, if that makes sense. Now, if you were to reverse the L/R channels (as in, the left channel will play right channel and vice-versa), then yeah, but polarity has nothing to do with it. Then again, maybe some headphones behave differently to polarity changes.
 
Maybe someone could help me out with this:
 
Planar headphones have a pretty flat phase through the frequencies, whereas on your typical dynamic driver, the phase is up a couple degrees in the bass, down a couple degrees in the mids, then swings back up exponentially, resonances aside. Here are some pics describing what I'm talking about:
 
Source

Source

 
 
I wonder how an inverted polarity would affect it, or are the things I'm talking about not related to each other?

Just tried the plugin.  If I reverse one of the channels, it makes the sound more muffled.  With both reversed, not much of a difference.  I don't find it to cause perception of wide environmental affect.
 
The reverse in polarity is like inverting the music signal wave in time domain about the 0 axis.  Phase is time shift of a sinusoid.  Top graph has phase deviations compared to bottom that is nearly 0 phase.  The phase on the graphs are either the current or voltage lagging each other  in time depending on the if the phase is positive or negative quantity.  It is caused by non-linear impedance which means there is a reactive component in the impedance that causes phase difference.  Power is the product of current and voltage, so if you have the current and voltage out of phase, the product would be distorted.
 
So reversing polarity and phase response of headphones are not related.  By that I mean reversing polarity doesn't affect the phase. The graph is showing the headphone's impedance and phase response.  It's showing how the headphone responds interms of impedance and phase to each frequency component fed into the headphone.  I would think if you do a frequency sweep of the headphone with reverse polarity you would still get the same impedance response.
 
So, in summary polarity swap is reversing the highest and lowest potential attached to the headphones which should reverse the signal about the 0 axis, and phase has to do with time shift between current and voltage.

 

[castleofargh]

Originally Posted by SilverEars /img/forum/go_quote.gif
 
Just tried the plugin.  If I reverse one of the channels, it makes the sound more muffled.  With both reversed, not much of a difference.  I don't find it to cause perception of wide environmental affect.
 
The reverse in polarity is like inverting the music signal wave in time domain about the 0 axis.  Phase is time shift of a sinusoid.  Top graph has phase deviations compared to bottom that is nearly 0 phase.  The phase on the graphs are either the current or voltage lagging each other  in time depending on the if the phase is positive or negative quantity.  It is caused by non-linear impedance which means there is a reactive component in the impedance that causes phase difference.  Power is the product of current and voltage, so if you have the current and voltage out of phase, the product would be distorted.
 
So reversing polarity and phase response of headphones are not related.  By that I mean reversing polarity doesn't affect the phase. The graph is showing the headphone's impedance and phase response.  It's showing how the headphone responds interms of impedance and phase to each frequency component fed into the headphone.  I would think if you do a frequency sweep of the headphone with reverse polarity you would still get the same impedance response.
 
So, in summary polarity swap is reversing the highest and lowest potential attached to the headphones which should reverse the signal about the 0 axis, and phase has to do with time shift between current and voltage.

I remember some trend where people were saying they could hear inverted polarity(on both side). and with some really weird hand picked signals it seems that you indeed can.
I always thought it was a pointless problem given that it would actually change very little to the music, and seeing how many IEMs have inverted polarity, I feel like I'm not alone no caring ^_^.
 
about the soundstage, it doesn't really expend anything obviously, but it makes some sounds harder to place in space. that in effect can, at least for me, give a feeling that the bass surrounds me even more. but it's so bad with higher frequencies that it's not an interesting thing to use. 
only the sub bass, I think I might like that(hoping that it wouldn't give me a headache after 1hour ^_^.

 

[Trae]

  Just tried the plugin.  If I reverse one of the channels, it makes the sound more muffled.  With both reversed, not much of a difference.  I don't find it to cause perception of wide environmental affect.
 
The reverse in polarity is like inverting the music signal wave in time domain about the 0 axis.  Phase is time shift of a sinusoid.  Top graph has phase deviations compared to bottom that is nearly 0 phase.  The phase on the graphs are either the current or voltage lagging each other  in time depending on the if the phase is positive or negative quantity.  It is caused by non-linear impedance which means there is a reactive component in the impedance that causes phase difference.  Power is the product of current and voltage, so if you have the current and voltage out of phase, the product would be distorted.
 
So reversing polarity and phase response of headphones are not related.  By that I mean reversing polarity doesn't affect the phase. The graph is showing the headphone's impedance and phase response.  It's showing how the headphone responds interms of impedance and phase to each frequency component fed into the headphone.  I would think if you do a frequency sweep of the headphone with reverse polarity you would still get the same impedance response.
 
So, in summary polarity swap is reversing the highest and lowest potential attached to the headphones which should reverse the signal about the 0 axis, and phase has to do with time shift between current and voltage.

 
Thanks for the info. I bolded the things I don't completely understand. I get that the phase is dependent on how resistive the driver is at a particular frequency (right?), but I don't understand how the phase affects the sound, and how the voltage/current differences change the phase. If you or any others could help me out, I'd really appreciate it. This kind of stuff is really interesting to me.