Edit: nevermind the stuff I deleted, if you saw it--bad at doing math on phone.

anyway ignore this rambling and read

this if you're interested in this stuff

This is confusing, because the power loss due to impedance is due to mismatches--namely, since the LX has an impedance of 14 ohms, you'd have the least power loss if your source also had an impedance of 14 ohms (resulting in a loss of 6.02dB). With a 10 ohm source impedance, you have a loss of 6.14dB, basically equal (less than a 3% difference in power loss, which results in a volume reduction of 4x). With, say, the ES100 in balanced mode, the source only has an output impedance of .5 ohms, which results in a power loss of 14.8dB, or a 30x reduction in volume.

So it's strange, because when you were losing (considerably) less power, you felt like it sounded worse! Now I've got some reading to do...

in the meantime, note:

or, with log-log scaling:

“power loss in dB”, here, is such that every 3dB results in a halving of the volume. but that makes you wonder—are any of your sources really not able to provide the extra dB when we’re talking about power scales around—and generally less than—a mW? and how does that result in a change in sound? I guess sensitivity and impedance are non-constant with frequency response.

I get that this is all pretty elementary for people involved in audio; is there a good source (a…aha) where I could read more about this? Specifically, about impedance curves—since the LX has an impedance of 14 ohms and a sensitivity of 102dB/mW

*at 1kHz*, I’m curious how the FR changes as a function of source impedance if we normalize either for total volume or volume at some target frequency.

edit: just to take this rambling a little further—is the frequency response curve, in a sense, just inversely related to the impedance curve? I tried the first stuff that came to mind, but didn't really get reasonable results.

Click to expand...