About the Schoeps vs DPA thing, and DPA being less fatiguing, sittng side-by-side, same recording etc. etc. Why not take a 30-second sample of the recording from both sources, make sure the volume level is the same, and then use a modelling EQ like the software one byVoxengo. You could model the EQ curve of one recording, then slap that on the other and visa versa. Not exactly scientific, but it might be interesting to superimpose the DPA curve on the Schoeps recording.