Since Jigwally's submission, I've been working on my own movie. This movie has all the puzzles solved:
User movie #65083085390942935
It's actually slower than my
previous movie, because the puzzle solves were entirely generated from scratch. There's actually a bit of subtlety to how fast a puzzle solve is, so it may be a little while before I'm satisfied with the level of optimization and end up submitting.
EDIT: Here's a level-by-level comparison:
Easy
0 0 0 0 0 0 0 0
0 0 0 0 -1 0 0 0
0 0 0 0 0 -8 0 0
0 0 0 0 0 -1 0 0
0 0 0 8 0 -7 0 0
0 2 -2 0 0 0 0 0
0 0 0 0 0 0 -1 0
0 0 0 0 0 0 0 0
Kinoko
3 0 0 0 0 0 0 0
0 -1 0 0 8 0 0 0
0 -8 0 0 0 0 0 0
-1 0 8 -2 0 -5 0 0
0 0 0 -2 0 -3 0 1
2 0 0 0 1 3 0 0
-6 0 0 6 0 0 -6 0
-2 0 -2 2 4 -2 0 -2
Star
0 -1 -4 0 0 2 -2 0
9 -1 0 0 0 0 0 0
0 0 0 1 0 0 0 0
0 0 0 -1 0 5 0 0
-1 2 -6 6 5 0 0 0
1 0 0 0 -2 -1 3 -2
-1 4 2 0 3 -1 0 0
2 0 0 1 6 5 6 0
Gains: -85, losses: +111, net: +26
The differences in level times comes down to (at least) three factors: whether the first two A-presses are adjacent, timer lag, and the time it takes to chisel the final square. The third factor is the most significant, making a difference of up to 7 frames in the final product (at least from this comparison.) All the solutions are optimal.
Also, all differences could be accounted for by in-level comparisons, so each level can probably be treated individually without worrying too much about later consequences.
The most I expect to save over my most recent WIP movie upload at this point is
00:38.58, and that's a pretty optimistic projection (assumes saving 9 frames on each level).
