To me, the new version is a totally different sounding speaker from paper already, shifting 70hz (400 vs 330hz)more to the more “directional” midrange should be obviously audible in real world in terms of sound stage accuracy, even though measurements on axis won’t show a difference. But off axis, I bet the new one should have a more slanting slope starting now from 330hz.
Even if it didn’t had meta material, this crossover change alone would sound different as physically the midrange and woofers don’t lie in the horizontal axis. Either the old driver had this capability or they should have done the needed improvement to handle this Change.
You are referring to the R11, correct?
Typically all things being 'normal' a smaller diver is LESS directional at a lower frequency not more.
Based on that premise, shifting the crossover to the coaxial midrange to a lower 330hrz frequency from 400hrz, will broaden the dispersion. (if it does anything notable)
The difference between 330hrz and 400hrz is pretty small. It might even just be a by-product of some other tweaks and the semantics of deciding exactly what number to nominally call the crossover point.
Both a 6.5" and 5" driver will have a very wide dispersion here.
Especially on a very narrow baffle. Using a wider baffle could of course affect dispersion here and simultaneously screw up the design's stellar looks.
At any rate not much stereo directional information is in the 330-400hrz zone to begin with but certainly there is some.
These wavelenghts of 330&400hrz are about 3feet and so easily wrap around many objects including the speaker to about the same degree.
KEF has changed the published crossover points for most of the line.
It will be interesting to see what is new in the measurements and the subjective take.