Oh...that's bar 11 in my score. I do play the G. It goes by quickly though.
True, I suck at bar counting. Ok, all played correctly then. I still think you did not make the most of this delectable dissonance though.
He's right, you know, the G-Ab semitone clash is understated to the extent that I can't hear it either. Are you playing the (lower) Ab?
Since we're looking at this piece, here are another couple of things I've noticed. One of them is that you've essentially changed the rhythm into 6/8 (2+1 groups not 3+1 as printed), but I guess this is deliberate, inspired by some recording by the man himself.
The other thing is the two identical arpeggiated chords in bars 7 and 8 (if I can count right). The way you play the first of these makes the lower G# (in the upper staff but indicated to be played with the left hand) sound after
the top of the chord. I don't think this is what he intended, and it may well not even be what you intended (although the effect is delightful and enhances the mood of the piece). It seems to me that this G# is printed a little to the right of the rest of the chord only to make sure it has a distinct stem, to make clear that the m.g.
instruction applies only to it; the note is otherwise to be treated as part of a nine-note chord, arpeggiated as normal all the way from bottom to top, it is not a separate note to come after an eight-note chord. I notice you play the next one differently, but I can't tell whether this is because you got it right or the G# failed to sound.
I presume you agree that the unbroken wavy line across both staves means that all nine notes should be attacked at slightly different times, the first note of the RH after the last note of the LH; that these are not two chords (one of 5 notes, the other of 4) which can be played concurrently and arpeggiated independently of each other.
Don't read too much into these petty criticisms. You certainly do these pieces justice, and I reckon "haunting" is a pretty suitable word with which to describe them.