HN Theater @HNTheaterMonth

⬐ niroze

Just takes a bit of practice! Anyone can do it.

http://www.speedcubing.com/

I, personally, love this book and have it on my nightstand https://www.amazon.com/Speedsolving-Easy---Follow-Step---Ste...

⬐ kelvin0

All I could hear in my head while the video was playing: "M M M M M M M M M o n s t e r Kill" (Anyone who has played Unreal Tournament will know what I mean :-) This is truly impressive if you only consider the aspect of solving the cube quickly, on top of that you have to admire the dexterity of the solver. Impressive (Quake 3 :-)

⬐ k_vi

I wonder if they have an algorithm solving the cube with 20 moves for any combination, if not Rubik's cube solution optimisation with a longer time limit could be a more interesting competition than speed solving.

⬐ iamgopal

My best is 30 sec and with the same technique I could not seem to reduce it further, whats the trick ?

-----

As the cube gets larger, the complexity will get reduced, ( by ratio of volume/surface area ) ? Am I right ?

⬐ blastrat

| As the cube gets larger, the complexity will get reduced, ( by ratio of volume/surface area ) ? Am I right ?

I know nothing about this except what a Rubik's cube looks like, but at first cut I'd say your comment is wrong but on the right track.

Rubik's cubes don't depend on volume at all, but the ratio of increasing the surface area is still an inverse square, so you'd be right if you focused on the ratio between area to linear dimension rather than the ratio of volume to area...

⬐ gregfjohnson

By "as the cube gets larger" do you have in mind 4x4x4 .. NxNxN cubes? The largest (physical) cube I have according to that metric is a 7x7x7 cube.

I have the following: 1x1x1 ;-) 2x2x2 3x3x3 4x4x4 5x5x5 7x7x7

Going from 3x3x3 to 4x4x4, there was one additional algebraically distinct sub-puzzle that I needed to solve.

Above 4x4x4, those same algorithms solve the 5x5x5 and 7x7x7. I have a marvelous proof (but don't have room in this margin) that these algorithms will work for all larger cubes. (Just kidding..)

I'm sure there are increasing numbers of moves that do multiple things simultaneously as N gets larger, but my approach is to ploddingly move 3 cubies at a time leaving the rest of the cube unchanged, and then around the corners and edges flipping two cubies at a time. (The new move I needed for the 4x4x4 was to simply rotate one of the middle slices to switch the parity of the edges if it was wrong.)

⬐ jozydapozy

The 4x4x4 is far more difficult then the 3x3x3. The 5x5x5 is almost the same as the 3x3x3 with just one extra step, though it takes more time to solve. The cubes with even numbers (4x4x4, 6x6x6) are more difficult because these have no fixed centers.

⬐ Someone

There are several ways to speed up:

- pick a way to solve it that requires fewer moves

- make fewer mistakes while working on the approach you picked

- make your moves faster (may require better accuracy in making close to 90 degree turns, or just speeding up your moves)

- spend less time pausing to think about future moves

- when possible (which is rarely) make more moves in parallel. A good cube allows some pipelining, though, in the sense that you can start turning a face before you completely finish turning one orthogonal to it.

To get better, pick the area you think you can improve on the easiest, train it, and iterate.

Once you find your cube keeps falling apart, get a better cube.

And complexity keeps going up, of course. It may go up slower, but I somewhat doubt it. My hunch (based on nothing) would be that god's number goes up slower than the number of cube positions. If so, finding the optimal solution gets harder the larger the cube.

⬐ sethbannon

Great video. You you enjoyed the above, you'll likely also appreciate this video of a Rubik's Cube being solved in 0.9 seconds by a robot: https://www.youtube.com/watch?v=ixTddQQ2Hs4