The FMC thread

[kinch2002]

Firstly, try not to do rotations. You'll give yourself problems if you try to do insertions/NISS and things like that.
After 2x2x3 I look at EO. You should see that 4 edges aren't oriented (pretty standard). i.e. they can't go into place without moving parts of the 2x2x3 block. So that's the BR, WG, OB and WO edges. I look at the 3 movers that orient 2 of the edges first. So, if you're in standard WG scrambling orientation, L' B L or L' B' L or L U' L' would orient 2 of those edges. Then I'd see whether any blocks/pairs are created by doing those moves, and work with those. Or try orienting the last 2 edges in different ways too.

EDIT: Also, pseudo blocks and NISS give you a lot more options here too. But I'm not going to type an explanation of those out right now - they're probably mentioned somewhere else in this thread

 

[Zane_C]

Thanks for the advice, I finished the solution.
I got rid of the rotation at the start. Regardless of the high move count, I'm happy to have incorporated some FMC techniques.
After I look at more FMC solutions in this thread and the weekly comp I should be able to set out my solutions neater.

Scramble: D' F U F2 R2 U2 R' B2 F2 D F' L2 F' L' U F' D' U' B2 D B2 L2 R' D2 U'

2x2x2 block: B2 U L2 F2
2x2x3 block: R2 U2 B' U R' * Insert [R F2 R', B'] - cancels to R2 U2 B' U [F2 R' B' R F2 R' B]

Orient 2 edges: L' B' L

Add 2x2 block: F R' U R U' F' U

Orient+permute remaining edges: U B' U' L' B' L B2 U B' U'

Solve remaining corners:
[L', U R' U']

= 42 moves STM, that was fun.

 

[kinch2002]

A couple of thoughts I just had using your 2x2x3
Scramble: D' F U F2 R2 U2 R' B2 F2 D F' L2 F' L' U F' D' U' B2 D B2 L2 R' D2 U'
2x2x3 block: B2 U L2 F2 R2 U2 B' U R' (9)

1. Same EO: L' B' L
Same block using less moves: B2 U L U' L'
Cancels into awesome last pair case: L' B L' B' L2 (also cancels into OLL if you want to try noob LL)
Haven't really tried to use this for a full solution. OLL PLL gives like 38 though

2. Focus on making a block and hope to do some EO during it: U B' R B' R' U B2 U2 (note the R B' R' orients 2 bad edges)
Another block: B L' B2 L
Insert last pair of yellow F2L with sledgehammer to orient edges: U R B' R' B U'
3 corners left with 27 moves. Note the GRW block eventually becomes part of the last layer and helps to force 3 corners left. Should be 33/34 moves

 

[Zane_C]

I have a question on insertions. When you're left with some corners after the 'skeleton', what's the common way to find commutators that can be inserted to cancel moves?

The only way I know is:
- Stick tape on the stickers and indicate which corners need to go where.
- Then start solving from the very start one move at a time. After each turn, analyse the new corner cycles, looking for cancellations.

For all I know this is the only way to go about it, I have absolutely no clue. Thanks.

 

[guusrs]

I have a question on insertions. When you're left with some corners after the 'skeleton', what's the common way to find commutators that can be inserted to cancel moves?

The only way I know is:
- Stick tape on the stickers and indicate which corners need to go where.
- Then start solving from the very start one move at a time. After each turn, analyse the new corner cycles, looking for cancellations.

For all I know this is the only way to go about it, I have absolutely no clue. Thanks.

Or revert your solution while searching for indertions. Saves time (no scrambling) but needs some practice

 

[irontwig]

I have a question on insertions. When you're left with some corners after the 'skeleton', what's the common way to find commutators that can be inserted to cancel moves?

The only way I know is:
- Stick tape on the stickers and indicate which corners need to go where.
- Then start solving from the very start one move at a time. After each turn, analyse the new corner cycles, looking for cancellations.

For all I know this is the only way to go about it, I have absolutely no clue. Thanks.

I don't know what to say; that's it.

 

[Zane_C]

Thanks guys, good to know.

Finally, something I'm happy with, despite a bit of luck.
Took about 90 minutes, spent the majority of time getting the insertions right. I kept on messing them up, because I assumed they behaved the same way as BLD cycles...

Scramble: U L' U2 R F2 B2 D2 F' D F' D' R2 U2 R' D' F' L' R2 U' D R2 L2 F R F'
Solution: R D B D' F2 D B' D' F2 F2 U' B D' F2 B B L' F2 L B' L' F2 L L2 B U2 B2 L U' B' L' B D L D' = 31 moves

2x2x2: R . F2 U' B D' F2 (5)
F2L-1: B : L2 B U2 B2 L U' (7/13)

Orient+permute edges: B' L' B D L D' (6/19)

Insert at dot: [D B D' F2 D B' D' F2]
Insert at colon: [B L' F2 L B' L' F2 L]

Comment: I tried to set it out neat. I was only expanding to a 2x2x3 block (green/red/white/yellow), the red/blue/white block was a complete accident.

Back up:
2x2x2: R F2 U' B D' F2 (6)
2x2x3: B L2 B U2 B2 L U' (7/13)

Orient+permute edges: B' L' B D L D' (6/19)

Last 4 corners:
U B U' F U B' U' F' (8/27)
B2 R F' R' B2 R F R' (8/35)

Constructive criticism is always welcome.

 

[Brest]

I really like the F2L-1, very nicely built. :tu

I noticed that during the start of your L5E that the last F2L pair connected. What I usually do in these cases is to play around with different insertions to see what kind of LL is left. In this case, a standard F2L insertion leave a SuneLL for a PLL skip!

Scramble: U L' U2 R F2 B2 D2 F' D F' D' R2 U2 R' D' F' L' R2 U' D R2 L2 F R F'

R F2 U' B D' F2 // 2x2x2
B L2 B U2 B2 L U' // F2L-1
B' L' B L2 D L' D' // F2L
F' L2 F L F' L F // SuneLL

Although a skip is not common, it is often better to spend the time looking for L3C rather than L4C / L5C. Insertions can take a lot of time to execute, so looking for one is better than looking for two.

 

[Zane_C]

Woah! I'll make certain to experiment with the LL next time. Thanks.

EDIT: Also, do most FMC solvers keep green on front and white on top while solving?
The other option is to perform rotations, just making sure to write the solution with respect to W/G.

 

[irontwig]

You can rotate at much as you want; you just write down F when you turn the green side rather than the front one.

 

[cubernya]

You can write down rotations. As long as it's within the limit of the competition (written # of moves) then it makes no difference (counts as 0 moves)

 

[irontwig]

You can write down rotations. As long as it's within the limit of the competition (written # of moves) then it makes no difference (counts as 0 moves)

You can, but it's a stupid idea.

 

[cubernya]

I personally find it easier to just write them down. Then I don't have to perform an OLL after x z2

 

[Zane_C]

Just did weekly comp 44 FMC:

FMC = DNF

Spoiler

Nooooooo! Ran out of time and DNF'd my back up solution by a U2. :fp
If only I had a couple more minutes I would've finished my final solution. I couldn't find any good skeletons so I did LL for both solutions.


Scramble: D F' L' B' L2 D2 U F' D B2 U2 B2 F' R' B2 U2 B2 U'
Final solution (after exceeded 60 minutes): L' D' L2 U2 L2 F2 U2 B' R B F' U' R F U2 F2 U' R' F R R U R' F' U2 F2 R' F' R L F' L' F U F U' F' U2 = 37 moves

2x2x2: L' D' L2 U2 L2
2x2x3: F2 U2 B' R B F' U' R
F2l-1: F U2 F2 U' R' F R
Solve 1x2x2 block+finish F2L: R U R' F' U2 F2 R' F' R
LL: L F' L' F U F U' F' U2

Fail back up:
Scramble: D F' L' B' L2 D2 U F' D B2 U2 B2 F' R' B2 U2 B2 U'
Solution: L' D' L2 U2 L2 F2 U2 B' R B F' U' R F U2 F2 U' R' F R U R U' R' U F' U F F R2 B' R' B R' F' B2 U R L' B2 L R' U B2 U = 44 moves...

2x2x2: L' D' L2 U2 L2
2x2x3: F2 U2 B' R B F' U' R
F2l-1: F U2 F2 U' R' F R
Finish F2L: U R U' R' U F' U F
OLL: F R2 B' R' B R' F'
PLL: B2 U R L' B2 L R' U B2 U

 

[Cubenovice]

You can write down rotations. As long as it's within the limit of the competition (written # of moves) then it makes no difference (counts as 0 moves)

Yesterday at the Dutch open I have seen a great example of why not to do this.
One CFOP solve of 60+ could hqve been a 50-something solve because of missed cancellations due to rotations and AUF's in the solve.

It is really not hard to just write the correct faces during your solve.
In the std WCA orientation two faces even begin with the correct letter: Back and Red

 

[kinch2002]

Also, do most FMC solvers keep green on front and white on top while solving?
The other option is to perform rotations, just making sure to write the solution with respect to W/G.

You can rotate at much as you want; you just write down F when you turn the green side rather than the front one.

That's pretty much what I do. I rotate all the time, because it's almost impossible to see what's going on without doing so. But make sure I write down the 'correct' faces. You could do FMC in your bld orientation, because then you'll already know which colour corresponds to which face.

 

[Zane_C]

That's pretty much what I do. I rotate all the time, because it's almost impossible to see what's going on without doing so. But make sure I write down the 'correct' faces. You could do FMC in your bld orientation, because then you'll already know which colour corresponds to which face.

Solving in BLD orientation seems nice, but I've decided to solve in any orientation. It's pretty easy since you just need to know that red=R, orange=L, yellow=D... etc
Also, I no longer restrict myself to building F2L-1 on only the green or blue face.

 

[Attila]

I found this solution for Dutch Open’s FMC scramble (about 2 hours) :
Scramble: D2 B D2 B F' U2 L R'U B U R U2 B2 D L' F2 D2 L
Solution:B2 F’ L2 D’ R L’ F D2 U2 L2 B F’ U F B’ R U L U2 L R F2 U D R ( 25 moves)
Explanation:
Simply inverse solution, without premoves and NISS.
R’ D’ U’ F2 R’ L’ U2 L’ all corners – 2 moves, and 3 edges,
U’ R’ B F’ U’ F B’ L2 more 3 edges,
U2 D2 F’ L R’ D L2 F B2 L6E.

 

[irontwig]

An important skill when using NISS is being able to perform the inverse scramble accurately and fast, so decided to do a average of 12; these were my 12 first times:

11.22, 10.04, 10.90, (9.54), 10.11, (DNF), 10.35, 11.16, 12.52, 10.41, 10.11, 10.46 = 10.73

This is several seconds faster than my PB average (15.05). So, what about you other NISS users?

 

[Mike Hughey]

An important skill when using NISS is being able to perform the inverse scramble accurately and fast, so decided to do a average of 12; these were my 12 first times:

11.22, 10.04, 10.90, (9.54), 10.11, (DNF), 10.35, 11.16, 12.52, 10.41, 10.11, 10.46 = 10.73

This is several seconds faster than my PB average (15.05). So, what about you other NISS users?

Mine:
11.69, (9.37), 11.41, (DNF), 10.24, 12.33, 11.27, 12.02, 10.43, 10.59, 11.65, 12.35 = 11.40

Considering my regular 3x3x3 speed, I'd say that's pretty good. Basically there's a very small difference for me between applying a regular and an inverse scramble - no more than a couple of seconds. But then, in addition to fewest moves, I also unscramble my cubes every week for "match the scramble" by doing the inverse scramble. I figure that:
a. it's faster that way, and
b. it's good practice for NISS.