I think it's fair to say that quill knobs are the preferred
biomechanical solution for transferring forces from wing feathers to the
skeleton in extant flying birds.
I'm not so sure it's the "preferred" solution as much as it may be the "best
available solution." Keep in mind that two different solutions to this
problem can seen in most any bird today. The flight feathers that are most
perpendicular to the extended wing (the first and secondary remiges) are the
ones embedded directly into the ulna via a "quill knob", while the shafts of
the more distal primary remiges (those that tend to be most parallel to the
extended wing) are secured to the carpometacarpus by resting in a groove and
anchored by tendons (sans quill knobs). It may be that evolution eventually
settled upon different solutions for transfering to the skeleton the forces
generated by the secondaries (which mostly generate lift) versus those of
the primaries (which mostly generate thrust), based on their inherited
biomechanical orientation rather than just aerodynamic optimization. After
all, evolution had to work with what it had. There may well have been other
solutions too, earlier in the process, which we haven't yet seen.