Hi Langel,
It depends on what you want!!
if you want the ball to react more on the bounce, ie for a topspin shot kick forward more, then a surface with higher friction is better (last photo, right-hand side)
If you want less reaction from the ball at the bounce with the spin on the ball being carried through to your opponent, then a lower friction surface is better (last photo, left-hand side) I think I've got that correct from the last photo!!
As you say it may end up being a trade off with the 'middle' ground being chosen by the ITTF
It would also be interesting to see if there are any papers on the difference in spin levels produced by a stroke for the differing balls - plastic, celluloid, 38mm 40 and 40+ etc. They are hinting that the spin levels are pretty similar when the ball is initially struck, and that it's the differing materials used to manufacture the ball contacting with the table that results in loss of spin carried by the ball.
So removing the bounce.
If the rotational spin on a 38mm celluloid ball from a set topspin stroke was lets say 600 RPS, and for the same stroke a 40+ plastic ball was 600 RPS (I'm not saying this is actually the case, just an example), which ball is carrying more 'spin energy' ? and would have more 'reaction' when contacted directly with a rubber (before the ball bounces) ?
Is there a correspondence between number of rotations & circumference of the ball ? when RPS are the same and circumference varies?