CS4398 capacitor vs. op-amp out

[Ciorny]

Personally, I think that the distortion of the OpAmp is low enough as it is and it is not worth messing with biasing in class A. It reduces the feedback capacity of the OpAmp on one alternance and limits the speed response on that alternance. It might work well for headphone buffer/amp, but for other stages, that rely heavily on a high open-loop feedback value, is a sub-optimal solution.
Hoever, more about bias in class A: Biasing Op-Amps into Class A

Maybe you are right:

Have switched both OPA627 in Class A, each with a 3K (see picture) and describe my impression as follows - without wonderful (double)-blind test stories:

Anyway, the bass remarkably decreased: I hear it very clearly through my 2x sub-woofers: opposed with it, the middles and the treble became more prominent than if I would turn back without Class A the bass control knob.

In sum, a the music sounds tiny else no significant improvement over standard with my circuit or Gigawork first step op-amp: finally have taken out the resistors..