If you do not like post-processing/EQ at all, I suggest you stop reading now
So the more I think about this, the more it makes sense to be recording with high boost grids with DPA 406x (and similar mics) instead of soft boost grids; then finally EQ away the excess in post
after applying gain.
Like most folks, I prefer to set levels conservatively to avoid surprise overloads. Most of my recordings peak around -12 to -5 dB, sometimes less.
However, my recording peaks are always in the low and mid regions:
excess bass due to room gain, audience clapping/screaming, or even a mic impact (mine or the performer)
The "air" regions are barely present at all and are likely to get lost in the noise floor. So, there should be more than enough headroom to accomodate the 10dB boost centered at 12kHz (looks like 0.707 Q) from the DPA graphs.
If I want a flatter recording after post, I'd add gain to boost all frequencies first, then drop to counter the boost as much as I need.
So something like the following sox command:
sox $IN.flac $OUT.flac gain $DB equalizer 12000 0.707q -10
For removing the soft-boost, it might be:
sox $IN.flac $OUT.flac gain $DB equalizer 15000 0.707q -3
Of course, the exact levels and widths can be tweaked accordingly in case of obstructions and such, and recordings typically benefit from good EQ anyways IMHO.
SoX also has a nice --plot option to visualize equalizer adjustments (assuming you you're using a *nix system with gnuplot and sox installed):
sox --plot=gnuplot >/tmp/plot -n -n equalizer $FREQ $WIDTH $GAIN
gnuplot /tmp/plot
Anyways I'll test this theory at upcoming shows to see if things are acceptable after post with high boost grids.