Adding to the cable discussion. I made my cables just for comfort and elimination of microphonics. However I do notice that I get better channel separation and wider soundstage. It probably has to do with crosstalk difference between the two cables.
I can't say I understand it but I think it has to do with the signal from one side of the output amp bleeding into the other side's output amp path to ground through the ground wire. So in theory if we have separate ground wires to both drivers, it should reduce that. Then after adding the 100ohm resistors it further reduced the crosstalk.
http://www.head-fi.org/t/538615/if-you-still-love-etymotic-er4-this-is-the-thread-for-you/5610#post_11411686
I tried to get my head around that several times and I just keep on failing. ^_^
by having a bigger load (cable or IEM) the amp will proportionally use less current and that in turn helps reducing crosstalk inside the amp(more so if the channels aren't physically well separated) and in fact it reduces all sorts of troubles, as most of our amps are voltage amp, not current amps so they would rather go higher with volts.
so in a way we could say that a higher impedance headphone(again cable included as it's one and the same for the amp, it's just a load) is improving crosstalk.
now the problem is that adding impedance in the cable can tend to increase crosstalk in the cable itself (arrrghhhhhhhhh), I remember reading something about the difference from adding the resistor to the ground or one resistor to each side on the positive cable, and I think the result was that 1 resistor on the ground sucked.
also shielding can play a part. if ideal, you can end up reducing crosstalk. if not it can do the opposite and increase both crosstalk and impedance by some margin.
of course I have no idea how to do those stuff best or the magnitude of what I'm talking about. add to that the quality of insulation for each wire in the cable and crosstalk values become something you can either measure and be sure off, or should just forget if you ever want to sleep again ^_^.
now the icing on the cake: try to get yourself some VST adding crosstalk. if you have no advanced audio stuff, you can try crosstalk2 vst (to use vst in something like foobar you first need a vst wrapper). you can get crosstalk2 in this free cool bundle http://sleepytimedsp.com/downloads
when used at 100% setting in consumer mode(the worst and most obvious) you end up with -30db crosstalk (average audio sources will usually have better than -60db crosstalk if they're not plugged into really low impedance IEMs(my first point in this post). to put it in perspective, listening to vinyls will give that kind of crosstalk value, it's pretty massive, yet it doesn't stop us from perceiving some great soundstage with vinyls. and that's when my head explodes.
it seems intuitive to think that less crosstalk will improve stereo perception simply because it does, adding crosstalk is nothing more than the sound slowly coming to mono, but only with the quiet parts of the music. yet in practice, you can hear the change with -30db, so it's not transparent, but does it ruin the soundstage? not at all, in fact with some musics, it kind of feels better, more like a big mass of music instead of exclusive instruments, but the feeling of space is still very real(with a slight forwad push a little similar to what you get with a very very soft crossfeed).
all that to say that I lost my mind trying to associate what I clearly felt as being better space/stereo on some sources, and crosstalk. without real success unless the values become crazy bad, I stand there like an idiot with no conclusion at all to all those informations(and that's only the surface of what I actually read on the subject).
still, can we call a cable good is it makes crosstalk up to -30db or maybe higher? I'm guessing that's not too hard to measure but I have a very very poor soundcard and the input has more crosstalk than anything I tried plugging into it
. so I can't say by myself.
I thought crosstalk was my nemesis because it was the only measurement that was going real bad on a sansa clip with low impedance IEMs, and I never appreciated the clips on really low impedance IEMs for some reason, so I ignorantly associated both, but in practice the clip is still at almost -50db crosstalk into 15ohm(from memory so the value might not be exactly it). so I just feel the the small soundstage/headstage I get with low impedance IEMs is disproportionate to what crosstalk at -50db does. I'm missing something but what? ...
anyway crosstalk alone really need to be bad to be audible. that's the only rational conclusion on the subject.
ahahah Ive done it again, sorry for that guys, as you see I go crazy over nothing all day long and I'm loving it in a kind of S&M way.
I want to know everything about audio, but it's kind of a challenging project.