Square-1 Move-count

[pjk]

I've been working on speedsolving the square-1 lately, and have been thinking about move-counts on various solves. I'd like to compare various move-counts of various speedsolvers to see what an avg move-count is.

Take an average of 5 or 10 solves and post the move-count here in this format:
Method Used: [brief description of method used]
Average solve time: [what time you normally are, don't need to actually take an average]
Number of moves:
1.
2.
3.
....
[the number of moves used will be the addition of the number of twists and the number of turns. A twist will occur everytime you perform a "/", while a turn will be any rotation of the U or D faces (for example, (3,-3)/ is 3 moves, while (3,0)/ is 2 moves). When you post your move-count, please post the number of twists and number of turns separately. Keep in mind this is different than Lars V. defines them (twist is a "/", while a turn is any rotation plus any twist]

 

[masterofthebass]

hmm... ok:

Method used: vandenbergh (about 10 EPs)
Average solve time: just sub20
# of moves:
1. (-3,-3)/(-3,3)/(3,6)/(0,3)/(3,-3)/(0,1)/(3,-4)/(3,-3)/(-1,3)/(0,6)/(6,2)/(-5,4)/(6,0)/(-4,0)/(1,-4)/
53 with adj parity left.
2. (-3,0)/(6,0)/(0,-3)/(-4,1)/(0,6)/(4,-3)/(6,4)/(6,-2)/(4,6)/(-4,6)/(3,6)/(2,6)/(0,-4)/(4,0)/(-4,-2)/(6,4)/
68
3. (-2,6)/(2,-1)/(4,-3)/(-3,6)/(-3,-4)/(1,4)/(3,0)/(-4,5)/(3,1)/(0,-3)/(6,6)/(3,3)/(3,6)/(6,0)/(-4,0)/
65

 

[Dene]

Method Used: Vandenbergh with almost 20 EPs and 1 adjustments in CP
Average solve time: 22-23
Number of moves:
1. (6,-4) (4,3) (3,3) (-4,3) (0,4) (-2,3) (-5,0) (0,5) (6,4) (6,2) (-3,2) (6,0) (6,0) (-5,0) (0,5) (0,2)
Turns: 33. Twists: 21. Adj parity left, then one more turn to AUF.
2. (-5,5) (3,6) (0,5) (0,1) (3,1) (-1,2) (0,3) (6,4) (6,0) (0,3) (-4,4) (0,3) (0,4) (1,0) (0,2) (-4,5) (6,0)
Turns: 42. Twists: 26.
3. (-2,3) (-1,-3) (3,0) (6,2) (4,3) (4,0) (2,0) (2,1) (-4,2) (1,4) (2,0) (1,4) (0,5) (-3,0) (4,0) (6,0)
Turns: 37. Twists: 22.
4. (4,-3) (0,6) (-4,0) (6,0) (6,3) (3,3) (6,0) (-3,0) (-3,5) (0,3) (0,1) (0,4) (0,1) (-2,2) (0,2) (0,4) (2,0) (4,0)
Turns: 33. Twists: 27.
5. (0,-4) (4,1) (-1,0) (6,3) (6,0) (0,3) (0,1) (-3,5) (0,3) (-3,3) (6,0) (0,3) (-3,3) (6,0) (3,0) (-3,3) (0,5)
Turns: 28. Twists: 22.

It was very hard to keep track of twists and turns separately. I counted the turns in my head, and kept a tally of the twists every time I did one. I just put "1" every time I did a twist, then counted them up afterwards. Therefore the numbers might not be completely accurate, but close enough.

 

[Swordsman Kirby]

Method Used: LBL (not Jason's)
Average: 20-25
Number of Moves: (twist metric plz (out of laziness))
21, 24, (20), 31, 22, 23, 26, 25, 24, 25, (33), 23 => 24.4 twists

As you can see, it varies a lot, especially when you don't know full parity PLLs.

 

[DavidWoner]

Method Used: Vandenbergh (10 regular EPs, 4 parities) and a few tricks
Average solve time: 21-22
Move count:

1. (0,-3) (6,0) (3,3) (3,3) (-3,3) (-3,0) (-3,5) (6,2) (-2,4) (0,4) (2,2) (6,2) (-2,4) (-2,0) (0,4) (4,0)
time: 16.06 turns: 34 twists: 22 one look parity EP

2. (0,-1) (4,1) (-4,3) (-3,3) (-3,0) (3,0) (0,3) (0,3) (-2,2) (0,2) (6,0) (0,2) (3,4) (-4,3) (0,3) (0,1) (4,0)
time: 22.56 turns: 43 twists: 29

3. (0,3) (0,-3) (4,5) (6,0) (-4,4) (-3,0) (6,0) (0,2) (2,0) (-4,0) (-2,4) (-4,2) (0,2) (0,2) (0,4) (-2,5) (6,0) (4,0)
time: 26.86 moves: 78 O perm on both layer D:

4. (1,0) (-4,6) (0,3) (0,3) (0,3) (-3,3) (0,3) (4,3) (0,3) (0,2) (4,3) (-5,0) (-4,0) (2,1) (-1,2) (1,3) (5,0)
time 25.62 moves: 83 parity then 2 look

5. (0,5) (0,-3) (4,4) (-3,5) (6,3) (-2,0) (-4,5) (-5,3) (-5,3) (0,5) (0,5) (2,0) (-5,0) (0,4) (1,2) (0,2) (2,0)
time 18.75 moves: 59

I got bored with separating twists and turns, so w/e. I don't think this definition of "moves" is a very good measurement. Looking at the 4th one, a 4 second parity alg is 22 of those moves. I think you'd have to look at the actual turn metric to get a good idea of whats going on.

 

[pjk]

I got bored with separating twists and turns, so w/e. I don't think this definition of "moves" is a very good measurement. Looking at the 4th one, a 4 second parity alg is 22 of those moves. I think you'd have to look at the actual turn metric to get a good idea of whats going on.

I'm just going by what the WCA calls a Square-1 "move", although I see that it is pretty hard to keep track of when solving.

Does anyone know the upper and lower bounds for an optimal square-1 solution, and if so, the definition of a "move" associated with it?

 

[blade740]

Method used: Vandenbergh/CP parity/about 50 EPs
average solve time: 17-18
# of moves:
[45/17] [67/30] [73/31] [56/25] [69/29]

You were kinda vague in your move description. You said (-3, 0)/ was 2 moves, but also said that a turn is only U or D, not /. My numbers are counting /s as well as U and D moves. Subtract to get the other one.

 

[jazzthief81]

Does anyone know the upper and lower bounds for an optimal square-1 solution, and if so, the definition of a "move" associated with it?

From Jaap's Puzzle Page:

God's Algorithm:
In 2005 Mike Masonjones calculated God's Algorithm for the Square-1, in the twist metric. This shows that every position can be solved in at most 13 twists (12 if the middle layer is ignored). The full results for each shape are on a separate page, but the totals can be seen in the table below.
Twists Positions
0 1
1 64
2 1153
3 17050
4 235144
5 3091458
6 38893230
7 452031138
8 4459167504
9 33671064770
10 149502310936
11 183662070768
12 63945120032
13 157452752
Total: 435891456000

[...]
Mike Godfrey has optimally solved over 300,000 positions, and found two positions at depth 30 (LV: this is move metric).

 

[Neutrals01]

Method used : Own method(variation from jb's)
average solve time : 32~35
# of moves:
61(4 steps no parity),67(4 steps no parity),67(3 steps with parity),60(3 steps no parity),62(3 steps no parity),69(3 steps no parity),70(4 steps with parity),83(4 steps with parity),84(4 steps with parity),84(4 steps with parity),84(4 steps with parity),71(3 steps with parity)..

61+67+67+60+62+69+70+83+84+84+84+71 / 12 = 71.83 moves on average

My moves I count by (2,3)/ = 3 moves, (0,2)/(1,2) = 4 moves..I sacrificed around 5~8 unnecessary moves just to make my look ahead and recognition better..

it is hard to keep track my twist and turns seperately..so I just count it as a whole...

 

[Mike G]

Does anyone know the upper and lower bounds for an optimal square-1 solution, and if so, the definition of a "move" associated with it?

<Mike Masonjones results for twist metric>

Mike Godfrey has optimally solved over 300,000 positions, and found two positions at depth 30 (LV: this is move metric).

moves = twists + face turns. Solving a further 250,000 random positions didn't throw up any more at depth 30 or above. However, a search of all the odd permutations of the cube shape found 6 distinct positions at depth 31. As far as I know, 31 is best known lower bound on the maximum number of moves needed to solve the Square-1.

[Sorry for bumping this thread, but I don't often visit this forum.]

 

[fanwuq]

I'm getting into the square 1 for up coming competitions. I suck at it a lot.

Method: my variation of beginner LBL.
1. cube shape intuitive non-optimal random turning.
2. Block building for 1/2 of bottom layer.
3. Get other 1/2 of bottom layer corners in correct permutation.
4. set up to EPLL to finish bottom layer edges.
5. correct parity if it exist.
6. CPLL (I use J and 2x(double layer J U2) as N)
7. EPLL (H, Z, Ucw, Uccw)

Average: I don't know, maybe ~1:40? Look ahead is horrible and turning speed is low.

Move counts:
1. 70
2. 73 (P)
3. 71
4. 93 (P)
5. 69
6. 89 (P)
...
more to come later.

Observation: (1,0) is one move, yet being off by (-1,0) is solved, not +2. Inconsistency in the turn metric? Or just the rule to make it practical to speed solve?

 

[Swordsman Kirby]

4. set up to EPLL to finish bottom layer edges.

Come on, just learn the proper algs. >_>

 

[DavidWoner]

4. set up to EPLL to finish bottom layer edges.

Come on, just learn the proper algs. >_>

6. CPLL (I use J and 2x(double layer J U2) as N)

What SK said.