Probably the wrong thread for discussing this type of things, but I would actually like to see three different levels of EQ being applied in combination:True but the bass still needs to be corrected. I have been correcting over the range 30-400Hz.
When I used REW to determine independent channel PEQ values the result was too much bass when I measured L+sub and R+sub due I think to the correlated left and right signals in the subwoofer. How would you obtain a better result?
3) Adjustment to personal taste, different quality of programme material, night time listening and so on. Could be PEQ, but just GEQ would be OK.
2) Room correction, mostly limited to the frequency range where it's really the room making the difference (but an adjustable range like we already have it is fine, of course). Must be PEQ, great if the algorithm gets further tuning.
1) Speaker EQ. If you already have spinorama data for your speakers (and you want to correct for the little defects) or you know that your silub needs taming (or a boost) in a certain frequency range, why leave this job to room correction? Make speaker EQ (optional, of course) the basic layer. People could even correct e.g. the baffle step of their full range drivers with no passive crossover components! Requires measuring equipment or reliable data. Must be PEQ.
Hint: Lyngdorf amps happen to provide exactly that. Plus virtual bass and treble knobs with variable center frequcies, but let's not go over the top.