Preamp inverts signal.

Not every listener is sensitive to absolute polarity.  And even if a listener prefers one polarity over the other it doesn't mean the preferred setting is the "correct" polarity.

In a typical recording session the engineers pay attention to phase issues, but not to absolute polarity.  In fact, nearly every mic preamp and mixing console has a polarity switch for each input channel.  These switches are freely used which can result in a multitrack recording where the drums have one polarity and the piano another.

In a purist, true stereo recording absolute polarity takes on more importance, but for most modern recordings I don't think it's a major factor.  Someone may have a definite preference for how they have set up their system, but I think Almarg's comments about circuitry is probably a better explanation for one's preference.

Onhwy61 wrote,

"Not every listener is sensitive to absolute polarity."

>> I never said they were. Some people hear better than others, no doubt about it. And there are exceptions to every rule. I suspect it’s probably true that people need to be trained to listen for the difference between N and R, otherwise they tend to not know what to listen for.  Just like distortion or say transparency or grain or whatever. Everyone should go out and and find a guru. ;-)

Onhwy61 also wrote,

"And even if a listener prefers one polarity over the other it doesn’t mean the preferred setting is the "correct" polarity.""

>>I never said it did mean that. People frequently don’t know what the heck they’re listening to half the time, don’t you think? No one actually thinks about polarity when he’s listening. Only if he’s testing for polarity. And then he has to do the test correctly, no? It’s a little more complex than you let on. Most likely 99% of audiophiles do not think about polarity EVER in the course of a year. But that doesn’t mean polarity is not an issue or is a non-trivial one.

"An ordinary man has no means of deliverance." ~ Old audiophile expression

GeoffK, very possibly. It's a fairly straightforward path. I'll go with what sounds best though.

Another factor is the design of the loudspeakers playing any given recording. Very few are phase-coherent; planars are, of course (one reason for their generally high-quality sound), but most multi-driver dynamics are not. Vandersteens yes (at the x/o frequency, at least), Wilsons no. The drivers’ polarities may be inverted relative to one another, in order to optimize the speakers frequency response and/or waterfall plot performance. If that requires running one driver with inverted polarity, so be it. Dave Wilson, for one, does that regularly. The Eminent Technology LFT-8b magnetic-planar and Vandersteens models 1, 2, and 3 use 1st order (6dB/octave) cross-overs between drivers, and are phase-coherant. If you are listening to a speaker which has it’s drivers working in opposite polarity, good luck listening for a change in absolute polarity upstream.

I must say that I was never troubled by polarity of a recording. Have inverted polarity at the speaker or amp end of their terminals at times when amp or pre were inverted. Guess i am not that trained in listening. Also mayby thankful for that.