deanlambrecht, I'm convinced that knowing what other people think is a huge hazard in this business. Half the time, people hear what they think they should hear, instead of what they actually do hear.
I will admit: On those "preamp shootout" CDs I can't hear any difference among the preamps that I can put my finger on. Nor am I sure that things sound the same, either. Each track just sounds to me the way it sounds, and I honestly can't tell what's different or the same, or better or worse.
Other people listen to those same CDs and they think, "Wow, that's great, this makes everything so clear for the first time." Which of us is right? Damned if I know.
Look, as far as bright and dark are concerned, that's pretty much a matter of frequency response and distortion, and most professional microphones have low enough distortion that it's not going to make them sound bright if they're not bright sounding to begin with. So that axis of the graph, if it's accurate (which it's not), at least makes some fundamental sense as a design choice.
But brightness and darkness are both deviations from neutral (flat) frequency response, specifically in the upper midrange and high frequencies. Is "coloration" vs. neutrality really an independent variable, then? I don't see how it can be. So right away, the whole structure of the graph makes no sense to me. You can graph a person's height as a function of their age, and chart their growth or shrinkage, or you can graph the distribution, say, of people 6 feet or taller in various ZIP codes. But these two axes aren't "orthogonal"--they're really different ways of expressing one thing that has more than one aspect to it, rather than being two independent quantities or one quantity that depends on the other. It's just a goofy design.
--best regards