See my post from earlier here:
http://www.guitarforum.co.za/general-discussion/advice-required-playing-acoustic-guitar-through-the-deskmixer/msg226288/#msg226288 - some common ground.
You can't always generalise, but the "smiley face" graphic EQ is a common sign of someone who has no idea. Unless it's a room with major problems, usually the closer everything on the EQs is set to flat, the better the engineer. Mids are where the human voice and guitars live, so knock 'em out at your peril.
The most important thing to remember is: There are no "magic" settings that work with every system, every venue and every example of an instrument type. There are a few rules you can apply in general, but generally you have to use your ears. Hell, I EQ on a song-by song basis - as needed (I also ride the faders, even with just a singer/guitarist).
The only rules I apply as fixed are these:
- Use your high pass filters on everything but kick, bass and keys to get rid of 90% of the low end mud.
- Graphic EQs are for room tuning on a system (or "ringing out" a monitor system), not for EQing a mix.
- Use decent mics and you will need a lot less EQ.
- A cut is better than a boost (need more mids? Cut highs and lows and boost the overall level).
- The closer to flat an EQ is set, the better it works. A combination of small boosts and cuts is more effective than large scale single frequency adjustments.
Get your EQ right and the levels follow naturally. But yeah, if there is a voice in the mix it almost always takes precedence. With the good acoustic players and singers, more-or-less equal level and minimal fiddling with it is required - let them control their own levels and dynamics or you'll end up fighting them at some point (Dave Ledbetter and Tina Schouw - easiest sound gig ever. Set basic levels and they do everything else with mic control and playing dynamics).
Oh another thing I do as a matter of course: Kill the reverb between songs to make spoken stuff more intelligible.