how to count moves

[Tommy]

I see lots of conversation about "move count". If I {R R} is that counted as two? If you do a Sune (or another algor) do you count each move of the algor or the algor as a single move?

I'm seeing that people are stating that they can do a solve in sub-20 seconds with sub-45 moves. This seems to be pretty amazing to me and I'm a little puzzled on how this happens.

BTW: My shipment of 7 cubes are reportedly at my area code's distribution center for EMS. I already have all of my tiles in. I picked up steel wool and CRC heavy duty silicone yesterday. I'm so excited about getting back to cubing since I dropped my puzzle the other day and had it smashed into little bits.

 

[ShadenSmith]

Well it depends what standard you're using. In HTM (half-turn metric) R2 (R R) is counted as 1 turn. However, every turn you make in an algorithm is counted.

 

[pjk]

When you count moves, every time you rotate a face, it is a move. People will talk about cancellations, such as this:
R U F F' R2 F D2
If you performed that, it would be 7 moves. However, with cancellations, you can perform that in 5 moves because the F and F' cancel:
R U R2 F D2

I'd suggest you check out the weekly competitions in the Competition Area forum and check out the fewest move entries that are posted.

Solving sub-20 in 45 moves is amazing? Which part is amazing, 20 seconds, or 45 moves?

For your information, the current 3x3 world record is 8.72 seconds with 49 moves I think, which comes out to 5.6 TPS (turns per second). 20 seconds for 45 moves is 2.25 TPS, which is much slower, and not very fast.

 

[Tommy]

pjk: Thanks for the statistics and the perspective. I've never heard of "cancellations" and will need to search the forum a little to understand that more. My guess is that you are trying to align the cube so it's columns and rows match better for the next move.

Personally, I think that sub-20 is amazing by itself. Then, I think that the 45 move solve is amazing too. Of course, my POV is from a 2:08 solver (best time) and I just got my cubes yesterday from c4y (uh, lots of assembly required) and am starting out.

 

[pjk]

Personally, I think that sub-20 is amazing by itself. Then, I think that the 45 move solve is amazing too. Of course, my POV is from a 2:08 solver (best time) and I just got my cubes yesterday from c4y (uh, lots of assembly required) and am starting out.

A lot of us on the forum have been there, and know what it is like to see people so fast. I remember a couple years back when I started, I never thought I'd get sub-20, all I wanted was sub-60. Then after sub-60, I really really thought about my times, I just practiced for fun. Now 20 seconds is actually pretty horrible. Keep practicing, if it interests you.... you will get there.

 

[rjohnson_8ball]

... I've never heard of "cancellations" and will need to search the forum a little to understand that more. My guess is that you are trying to align the cube so it's columns and rows match better for the next move.

It's simpler than that. A set of moves that cancel simply bring you back exactly as you started, as if you had done no moves at all. So (F F') does nothing. (R U)*105 does nothing (because every 15 repetitions the corners return to normal and every 7 repetitions the edges return to normal so after 15*7=105 repetitions everything returns back to normal).

Here is a practical application... developing a "J" permutation. The sequence (R' U L U' R U L') swaps the ULF and ULB cubies and also a couple edge cubies. But that sequence messes up the orientations of 3 corners, so the Sune can be used directly afterward to fix them: (L U L' U L U2 L'). Note that the L' of the 1st sequence is undone by the L of the 2nd sequence. When those are removed, we will see U U together, which is simplified to U2. So the result is (R' U L U' R U2 L' U L U2 L').