My best time is under 1:00. It’s been almost 30 years since I picked up one, though.
I learned it from a book. I’ve never been one for putting more work on my shoulders than absolutely necessary. 
I can’t quite figure out how those instructions work. At the moment I’m thinking…
“R” - take hold of the right slice and rotate it … up? Does that mean that “RU” is actually one instruction? Or is it take the right slice and move it some direction (may not matter if you’re consistent) and then do the same with the top slice?
“F[sup]2[/sup]” presumably spin the front by 180
“D’” bottom slice to the…left? right? opposite of the direction you normally would?
“R[SUB]s[/SUB]” - this one’s the killer. What’s the Right Middle slice? Is that the slice that runs through the cube horizontally, turn that to the right?
feverishly experimenting…
Me, too!
More Rubik’s goodness from a very good and fun brain store.
High School.
Foods and Nutrition class.
We had a cake decorating contest. My partner and I made …
… a Rubik’s Cube cake.
I’ve contacted my lawyer.
R means rotate the Right side 1/4 turn clockwise, as viewed from the right. D means rotate the bottom face clockwise, as viewed from below. U means rotate the top face clockwise, as viewed from above.
R[SUB]s[/SUB] means rotate the middle slice (the one between the right and left slices) 1/4 turn clockwise, as viewed from the right. (so L’[SUB]s[/SUB] would actually be the same move.)
I’ve tried a couple different interpretations of [sub]a[/sub], with no luck figuring out what it means.
No idea, but could you modify your Virtual Cube with loops to do an exhaustive search N moves deep, saving any moves where the cubes are in the right poition, and only two are rotated (or flipped)? Maybe be able to limit the possible moves to only three or four sides, to be able to search deeper in the number of moves.
For any value of N large enough to be useful, that would take a prohibitively large amount of computational resources. And my program isn’t implemented in such a way that I could easily build a smart solver, either: It tracks squares, not cubelets.
One thing that I was pleasantly surprised that it simulated accurately: The first time I demonstrated it to the teacher, one of the stickers peeled off. That is to say, some unknown computer glitch zeroed out one of the memory elements, causing one of the little squares to turn black, but the cube continued to operate correctly, moving that black square around exactly as if the sticker had peeled off of a real cube.
Does he like cake? 
He probably wouldn’t have liked the one we made; it sucked.
But it’s the principle of the thing. Which sucks too.