Originally Posted By: VideoTrack
Yes, double flats and double sharp notes are dependent on the key signature, and written as accidentals so that they don't alter the notes used by the key signature (and we are never allowed to do that, are we!)

A classic example that comes to mind is Moonlight Sonata, Opus 27 by Beethoven. Several bars have notes such a F double-sharp and B sharp because it is incorrect to change the notes of the key signature (C#min)
Trev

Hmm, not quite sure I agree with you wink
I would suggest instead that it is less the key signature in control here and more the current chord.

In your example, the left hand is playing octave G#'s. This suggests that either a G#m chord, or much more likely a C#m/G# chord is being played.
Now, despite the key sig, a G nat. cannot exist in either a G#m or C#m chord because you can't have a G AND a G# - the chord ends up being mispelled, so you hve an F## instead because you CAN have an F in the chord.
Then there's the A#, the G#m and C#m chords have a B nat. in them, they can't have a Bb too, so an A# must be used.

In the second bar, it looks to me like the chord is definitely a C#m on a G# bass, so the chord spellings still need to follow the same rules, Can't have a C AND a C#, so we have a B# instead, and the A nat. is fine.

Of course, because those 2 bars are both likely in C#m and thus match the key sig, it looks like we're trying to preserve key sig spellings, when it's actually the chord spelling that we're maintaining.

You see this a little more clearly in jazz numbers where the key centre can change every couple of bars without a corresponding key signature change.

In that case you keep an eye out for V7 chords 'cos they help identify the current key centre. Then you can end up with V of V chords...

Getting far too complex for this reply and there are better theorists on this forum than me anyhow.

--=-- My credo: If it's worth doing, it's worth overdoing - just ask my missus, she'll tell ya laugh --=--
You're only paranoid if you're wrong!