[Help Thread] FMC Discussion

With 4 corners which of one is twisted I mark the cycle [X] and [A1i->B2ii->C3iii->D4iv] where X is the twisted one and the A and D stickers is on the same corner. With one twisted corner you only really have two options: a. Insert twisted corner in a place where a oriented one is (off by one move) b. Insert at corner in the same layer as the twisted one. Looking for more than just 8-movers does rarely produce better results and just wastes time (and requires more thinking. KISS!). For me looking for one corner cycle typically takes about 5 minutes. After I've done the first cycle I typically just choose (one of the) best insertion and then continue with the second. And again here one might get a slightly better result by checking more possibilities, but the clock is ticking...

My try at your skeleton:

D R2 D' L D R2 L' D B' D B' U2 B D' B' U2 R2 B' R2 U2 R' U R2 U' B R B R' (28)

L.D L' D B2:R2 B' R2 U2 R' U R2 U' B R B R'

.=L' D R2 D' L D R2 D' Quite annoying for me since I work from the end.
:=B D B' U2 B D' B' U2 Finally they came into a 8 move position!
 
At work so no opportunity to go through the actual solve…

My stickers for this situation:
“X” or “O” for the twisted corner
“1” , “2” and “3” for the “open” cycle. I do not put 2 stickers on the same corner but make a note on my paper:
Cycle: 1 -2 -3 – 1 “target sticker color”
This tells me that 3 needs to go to that particular color on the “1” corner

My solving approach:
-1- place a single corner and just make sure the twisted corner is in the interchange part
-2- solve the remaining three corners

At the 1st stage there are a lot of opportunities as the orientation of the twisted corner does not matter.
Make sure to look further than the “sticker” position, look at the twisted corner as a “cubie” that needs to be interchanged.

In case of twisted corner(s) I actually re-sticker the cycle when switching to the 2nd stage to avoid errors.
 
Or you know the [wiki]L4C[/wiki] :

Alg here
R U2 R' L' U2 R [*] R' U2 L R U2 R' [AUF] ... ez case, I use it for COLL, 2x2x2 CLL and even big cube edge pairs (r,l and only U2 then)

* = U or U' depending on cycle direction, U2 makes a H case.
 
Last edited:
There was a 19-move solution (one move from optimal) submitted to Fewest Moves Challenge #338, with no explanation as to the derivation. The moves seem pretty random and don't seem to be solving much in the beginning half. It seems to me likely to have been found by computer. Anyone else have thoughts about this?

Scramble: F B2 R2 L' F' U' D' R2 L B2 F' D' L R2 F2 U F' U' L D' U' F' L2 R2 D' R2 U2 R F' L2
Solution: D R F' L B2 U2 D F' R' U2 D B U2 F2 L B2 R L2 D2
 
Hmm...

that is a steep improvement from his/her last blog:

http://howtosolvea3x3rubixcube.blogspot.com

Friday, 18 February 2011

F2L Method
A woman, obviously an expert on speedcubing (the art of solving a 3x3x3 fast - very fast), Jessica Fridrich is often associated with this method. I really don't know anything about her except she can solve it in less than seventeen seconds.
Posted by Aditya Srinivasulu at 01:01

F2L Method
Well, I'm not much of an expert in the F2L, but I like to use some of the algorithms from the F2L in my method, and look, I can solve it in under 2 minutes.
Posted by Aditya Srinivasulu at 00:57

Rubik's Cube
Hey guys! Have you been trying to solve your original 3x3x3 Rubik's Cube? Want Help? Ok.
 
cuBerBruce said:
There was a 19-move solution (one move from optimal) submitted to Fewest Moves Challenge #338, with no explanation as to the derivation. The moves seem pretty random and don't seem to be solving much in the beginning half. It seems to me likely to have been found by computer. Anyone else have thoughts about this?

Scramble: F B2 R2 L' F' U' D' R2 L B2 F' D' L R2 F2 U F' U' L D' U' F' L2 R2 D' R2 U2 R F' L2
Solution: D R F' L B2 U2 D F' R' U2 D B U2 F2 L B2 R L2 D2
Click to expand...

Imo obvious Kociemba since the ending is in {R,L,U2,D2,F2,B2}
 
cuBerBruce said:
There was a 19-move solution (one move from optimal) submitted to Fewest Moves Challenge #338, with no explanation as to the derivation. The moves seem pretty random and don't seem to be solving much in the beginning half. It seems to me likely to have been found by computer. Anyone else have thoughts about this?

Scramble: F B2 R2 L' F' U' D' R2 L B2 F' D' L R2 F2 U F' U' L D' U' F' L2 R2 D' R2 U2 R F' L2
Solution: D R F' L B2 U2 D F' R' U2 D B U2 F2 L B2 R L2 D2
Click to expand...

Obviously computer solution, if not explanation. I think, should be ignored.
 
cuBerBruce said:
There was a 19-move solution (one move from optimal) submitted to Fewest Moves Challenge #338, with no explanation as to the derivation. The moves seem pretty random and don't seem to be solving much in the beginning half. It seems to me likely to have been found by computer. Anyone else have thoughts about this?

Scramble: F B2 R2 L' F' U' D' R2 L B2 F' D' L R2 F2 U F' U' L D' U' F' L2 R2 D' R2 U2 R F' L2
Solution: D R F' L B2 U2 D F' R' U2 D B U2 F2 L B2 R L2 D2
Click to expand...

The solution can be obtained by using Cube Explorer (I used ver. 4.64).
The step to reproduce:

  1. Enter Maneuver and push "Apply"
  2. Push "Add and Solve" (then 20 HTM solution is displayed, F' D2 ... D L2)
  3. Push the arrow button next to "optimal"
He might have used Cube Explorer in this way.

I would not like to say that any solution by Kociemba-like method should be disqualified. Actually I tried to use such a method in FMC #310, explained here, and Per also used such a method and got 19 HTM in FMC #102. However, I believe that obtaining a short solution by such a method in 1-hour is too difficult for usual human beings.
 
Last edited:
Scramble: U' F' D R2 D' U' B' L' D2 U' B' F L' B2 L' D' U2 L' R2 U F2


U' B L' F' (x-cross)
U B2 U' R [B'] ( U F' U' B' U F [U']) (double x-cross + insertion)
U' B U2 B' R' (third f2l)
U B U B' U' R' U2 R (fixing some LL blocks)

27 HTM

quite simple scramble... finally I got a sub30 solution!

what do you think about it?
 
okayama said:
I would not like to say that any solution by Kociemba-like method should be disqualified.
Click to expand...
You’re right, i also found sometimes like solution, which ending is similar to Kociemba two-phase method (for example this weekly comp.) But i (as everyone) try clearly explain, how i was thinking. His explanation (wohoo!) :D is not acceptable. Of course for Aditya it is possible explain here his solution, if he knows. :P
 
Ye, that's a cheater, no doubt...

-------
Optimal scrambles for LS + LL (50% are inverses). Nice if you like to test how optimal you are for the last parts.

U2 B R' F2 D2 L2 D2 F2 R B' U (11f*)
U' F R2 B2 R F R' B2 R F2 R (11f*)
U2 B R' F2 D2 L2 D2 F2 R B' U (11f*)
R' F2 R' B2 R F' R' B2 R2 F' U (11f*)

U L' U2 L F' U' F U' L F' L' F (12f*)
U' R U L R' D F D' L' F U' F' (12f*)
F U F' L D F' D' R L' U' R' U (12f*)
F' L F L' U F' U F L' U2 L U' (12f*)

U2 F U2 F' U' F R' U' F2 U F R U' (13f*)
U' F' U F U' F' U2 F U2 R U R' U (13f*)
U' F' U B' R2 D' F' D R2 B U' F U2 (13f*)
U R' F' U' F2 U R F' U F U2 F' U2 (13f*)
U' R U' R' U2 F' U2 F U F' U' F U (13f*)
U2 F' U B' R2 D' F D R2 B U' F U (13f*)

R2 B' R' B R U2 R2 U' R U R U2 R2 U' (14f*)
F2 L' U' L U F L' U' L U L' U' L F (14f*)
U' R2 F' R B' R D2 F' B' L F2 L D2 B2 (14f*)
F2 L F' B2 R2 D' R2 F B2 L' F U2 F U (14f*)
R L U2 R2 F' L D F2 D' F R F2 L2 U' (14f*)
R U L2 D B2 D' R' B L U B2 U B L (14f*)
U2 F2 L F D U F U' F' D' L2 U2 L F (14f*)
U2 R U L' B L U2 B' U' R' B' U B U2 (14f*)
U R2 U2 R' U' R' U R2 U2 R' B' R B R2 (14f*)
R' F R D F' U F2 D' L' F U' F' L F2 (14f*)
F' U' F2 U2 L F2 L' U2 F2 U' F R U R' (14f*)
F' L' U L U' L' U L F' U' L' U L F2 (14f*)
U L2 F2 R' F' D F2 D' L' F R2 U2 L' R' (14f*)
R' F D' F D F2 R' F2 U B U' F2 B' R2 (14f*)
U2 B' U' B R U B U2 L' B' L U' R' U2 (14f*)
B2 D2 L' F2 L' B F D2 R' B R' F R2 U (14f*)
U' F' U' F U B' R' F' R B R' F2 R F' (14f*)
F2 L' F U F' L D F2 U' F D' R' F' R (14f*)
L' B' U' B2 U' L' B' R D B2 D' L2 U' R' (14f*)
R2 B F2 U B' U' F2 R F2 D' F' D F' R (14f*)
U' F' U2 F' L B2 F' R2 D R2 B2 F L' F2 (14f*)
F R' F2 R B' R' F R B U' F' U F U (14f*)
R U' R' F' U F2 U2 L F2 L' U2 F2 U F (14f*)
F' L' U2 L2 D F U F' U' D' F' L' F2 U2 (14f*)

U' R U R' B L' B' L B' U2 B U L U L' (15f*)
B' L2 F' R B2 F' U2 B L' B2 F R2 D2 U F (15f*)
B2 R' B U2 L' U R' U' L R2 U2 B' R B2 U (15f*)
F' U B U' B' F U' R2 U2 R F' U2 F R2 U2 (15f*)
U F U' B' U F U' L U F U' L' B F U2 (15f*)
F' U L' U L2 F2 U L' F L U' F2 L' F U2 (15f*)
F R' F' R2 U2 R' U' F' U L' U2 L U2 F U (15f*)
B F' U' F U B' F' L' B' U2 B U' L U2 F (15f*)
U F' R' D F2 D' R U2 L' U L' U2 L2 U F (15f*)
F' U2 B' R' F2 R' F2 R2 B F U R U R' U2 (15f*)
U R U' B U L U2 L' B2 U' R U B U' R2 (15f*)
U' B2 R' B U2 R2 L' U R U' L U2 B' R B2 (15f*)
U2 R2 F' U2 F R' U2 R2 U F' B U B' U' F (15f*)
U' F' U2 L' U2 L U' F U R U2 R2 F R F' (15f*)
F' U' D2 R2 F' B2 L B' U2 F B2 R' F L2 B (15f*)
U2 R U' R' U' F' B' R2 F2 R F2 R B U2 F (15f*)
F' U2 L' U B' U2 B L F B U' F' U F B' (15f*)
U F' L' B' R' U' R U B U2 L2 F' L' F2 U2 (15f*)
R' F2 R2 U' R' F' L' R U' R' U2 L2 F L' U (15f*)
R U2 B2 U2 B2 U2 B R B' U B2 U' B2 R2 U2 (15f*)
L U' L' U' B' U2 B L' B L B' R U' R' U (15f*)
U' L F' L2 U2 R U R' L F R U R2 F2 R (15f*)
F' U' L2 U2 L U' L U2 R' D F2 D' R F U' (15f*)
U2 F' B' L U F' U' L' U F' U' B U F' U' (15f*)
F2 L2 D2 B2 L' F L2 B L F2 R' F' U2 R' U (15f*)
U2 F' L F2 U L' F' L U' F2 L2 U' L U' F (15f*)
U' R U2 F R F2 L' B' L2 F' L B2 D2 L2 F2 (15f*)
U F' U2 B F L2 F2 L' F2 L' B' R U' R' U' (15f*)
R2 U B' U' R' U B2 L U2 L' U' B' U R' U' (15f*)
U2 R2 B2 U B2 U' B R' B' U2 B2 U2 B2 U2 R' (15f*)
U2 F2 L F L2 U2 B' U' R' U R B L F U' (15f*)
U R U R' B L F2 L F2 L2 F' B' U2 F U' (15f*)

I'm collecting the cases because I like to know the average number of moves for this part, currently it is ~14 and it won't change much even if I add more cases. Getting the optimal scrambles is just a bonus.

-------
Mini challange, I randomly picked two scrambles from the list.

5 minute linear : U2 B' U' B R U B U2 L' B' L U' R' U2 (14f*)
30 minute traditional : F' U' D2 R2 F' B2 L B' U2 F B2 R' F L2 B (15f*)

5 minute linear: (took less than 1 minute)
EO: U' R' F R F' (5)
LS + EP: U F' U' F (4, 9)
A-PLL: L F' L B2 L' F L B2 L2 (9, 18)

30 min: (I used ~20)
Solution : R U' R' U F' U' L' U2 F U2 B' U2 B U2 F' U2 F R F' L F R' (22)

Using premoves: L F (I tried R' and F before this, all sets up to pseudo XXX cross)

Pair-EO: R U' R' U F' U' L' (7)
LS-EP-1 corner: U2 F U2 B' U2 B U2 F' (8, 15) ... a shorter LS-EP is possible but this com saves a corner.
AUF-premoves: U2 . L F (3, 18)

Leaves L3C on F (Niklas).

Insert: . = F R F' L F R' (6-2, 22) ... Nice to have the premoves in the com.

This was like the worst scramble, I did not even try to solve it without premoves :p
 
Last edited:
Tim Reynolds said:
All right, can someone explain to me how to do an insertion of more than 3 corners? Specifically, can you find something for the below? (Warning: this is a skeleton to the in-progress weekly competition.)

Scramble: U2 B D' B R U2 R U' R2 B D R2 U L B R D' R
Skeleton: L D L' D / B2 R2 B' R2 U2 / R' U R2 U' / B R B R' (18)

I'll post my full solution in the weekly comp thread (2011-12) when I finish some BLD stuff.
Click to expand...

My attempt:
L D L' D / B2 R2 B' R2 U2, / R' U R2. U' / B R B R'

,= U L2 U' R U L2 U' R'
.= R' D2 R U' R' D2 R U

Results in a 28 move solution of:

L D L' D B2 R2 B' R2 U' L2 U' R U L2 U' R2 U R D2 R U' R' D2 R B R B R'

With the cycle of A B C A+, (Z/+/-) I just went though and checked A B (Z/+/-), B C (Z/+/-), C A+ (Z/+/-) for each of the positions, finding the <RDU> cycle first. Then it is just a standard 3cycle insertion.
 
Tim Reynolds said:
All right, can someone explain to me how to do an insertion of more than 3 corners?
...
Click to expand...

Kenneth said:
Or you know the [wiki]L4C[/wiki] :

Alg here
R U2 R' L' U2 R [*] R' U2 L R U2 R' [AUF] ... ez case, I use it for COLL, 2x2x2 CLL and even big cube edge pairs (r,l and only U2 then)

* = U or U' depending on cycle direction, U2 makes a H case.
Click to expand...
Hmm... Kenneth's alg is not optimal.
L' U2 L F' U2 F' R F U2 F' R' F2 U2 (13f*)

For a twisted corner and a "twisted 3-cycle," the best case direct solve is 12 moves. I get the following distribution of moves for these cases (for a fixed direction of the twisted corner):
Code: 
12f*:     33
13f*:    369
14f*:    222
15f*:      6
The 33 12-move cases are:
(These are generator sequences for cases with a clockwise-twisted corner, or solver sequences for cases with a counterclockwise-twisted corner.
U2 L F' R B L2 D2 B R' F L B2 (12f*)
B2 U F D' B R2 U2 B D F' U L2 (12f*)
R B2 D2 R U L' D F2 R2 D L U' (12f*)
B2 D2 R F' L B R2 U2 B L' F R (12f*)
F U2 R2 F D' B U F2 L2 U B' D (12f*)
F U2 R2 F L B' R D2 F2 R B L' (12f*)
L' U R D2 F2 R U' L D R2 B2 D (12f*)
D L2 F2 D B U' F R2 D2 F U B' (12f*)
L2 B R F' L D2 B2 L F R' B U2 (12f*)
U2 R2 F L B' R D2 F2 R B L' F (12f*)
R B' L F R2 D2 F L' B R F2 U2 (12f*)
F U' B L2 D2 B U F' D R2 B2 D (12f*)
L2 B D' F U B2 R2 U F' D B U2 (12f*)
U2 L D R' U F2 L2 U R D' L B2 (12f*)
D L2 F2 D R' U L D2 B2 L U' R (12f*)
U2 R2 F D' B U F2 L2 U B' D F (12f*)
D L' U B2 R2 U L D' R F2 U2 R (12f*)
F R B' L U2 F2 L B R' F D2 L2 (12f*)
U' B D F2 R2 D B' U F D2 L2 F (12f*)
B2 U R' D L U2 F2 L D' R U L2 (12f*)
B' L F R2 D2 F L' B R F2 U2 R (12f*)
B' L U2 F2 L B R' F D2 L2 F R (12f*)
L2 F2 D R' U L D2 B2 L U' R D (12f*)
R D L2 F2 D R' U L D2 B2 L U' (12f*)
U' B L2 D2 B U F' D R2 B2 D F (12f*)
R B2 D2 R F' L B R2 U2 B L' F (12f*)
B2 D2 R U L' D F2 R2 D L U' R (12f*)
R D L' U B2 R2 U L D' R F2 U2 (12f*)
D F U2 R2 F D' B U F2 L2 U B' (12f*)
F R B2 D2 R F' L B R2 U2 B L' (12f*)
D F U' B L2 D2 B U F' D R2 B2 (12f*)
L' U B2 R2 U L D' R F2 U2 R D (12f*)
R B' L U2 F2 L B R' F D2 L2 F (12f*)
The distribution for the subset of cases with the four corners in the same layer are:
Code: 
13f*:     12
14f*:      6

I note that solving using two 3-cycle insertions, you have nine cases for the flexible insertion (as Tim mentioned), so you would like to get a really high cancellation on that insertion (maybe around 4-5 moves), if you have the time to check enough possibilities. (You need to allow enough time for the 2nd insertion, of course.)
 
i am a complete noob when it comes to FMC but i am interested so can someone answer a few questions?
what is probably the best FMC method? (like the fridrich of FMC)
how do you start to learn FMC? (if that makes any sense)
how should i practice and improve?
how did the world record solver get such a low move count? a bit of luck?
what method should i start with? i currently use fridrich and thats all i know (except pochmann for bld)
and can you add any other helpful info? or link me to a site full of info? (i have seen the wiki but there is still a lot of info, to much to comprehend.
 
cuBerBruce said:
Hmm... Kenneth's alg is not optimal.
Click to expand...

Na, but easy to remember, the reason I know an alg at all =)

sa11297 said:
i am a complete noob when it comes to FMC but i am interested so can someone answer a few questions?
what is probably the best FMC method? (like the fridrich of FMC)
how do you start to learn FMC? (if that makes any sense)
how should i practice and improve?
how did the world record solver get such a low move count? a bit of luck?
what method should i start with? i currently use fridrich and thats all i know (except pochmann for bld)
and can you add any other helpful info? or link me to a site full of info? (i have seen the wiki but there is still a lot of info, to much to comprehend.
Click to expand...

There is no real method, improvisation is the best. But, people mostly uses block building for the first parts (Heise, Petrus), then try to solve as much as possible to reach a case you can solve using a commutator. Hopefully the commutator is possible to insert in the skeleton to cancle moves, else you just do it last.

Just do it, make a normal cross F2L solve. If you reach a bad LL, then do it again but the pairs in a diffrent order, and you may get a better LL-case. FMC is simply trying the same scrable several times to find the shortest possible way to solve it.

Just parctise, improvement comes from experience.

Yes, luck.

Start with your Fridrich then, but try to make X-crosses or double x-crosses before the pairs (x cross is block building, a start to more advanced blocks). Blocks and commutators saves moves, if you keep doing FMC you will eventually learn =)

Link to Heises page about coms : http://www.ryanheise.com/cube/commutators.html

Also check out his other "fundamental.." pages :)
 
Last edited:
From Weekly competition 2011-12:

guusrs said:
fmc: L F D' B U B' D2 B U' B'. L' D L' B L U B2 U R B' U' B' U B2 R' B2 R B' (28)

all but 3 corners: L F D * L' D L' B L U B2 U R B' U' B' U B2 R' B2 R B' (21)
at * D2 B U B' D2 B U' B', 1 move cancels

Easy scramble, why not sub 25?
Can anybody find a good continuation on this beauty start: L .F. D L' D (5)?

prepared for my first competition in 2011 next week!

Guus
Click to expand...
I found the following skeleton very quickly, maybe within 5 min (still not sub 25, though).
I'd like to team up with you in a competition, ha-ha. :)

Scramble: U2 B D' B R U2 R U' R2 B D R2 U L B R D' R
Solution: L F D L' D B' R U R' U2 B' R B R' U R U D' F2 D B2 D' F2 D U2 (25 HTM)

Pre-scramble: U2 B2 U2

2x2x2 block: L F D L' D (Guus's start)
2x2x3 block: B'
F2L minus 1 slot: R U R' U2
More 2x2x1 block: B' R B
All but 3 corners: R' U R U'
Correction: U2 * B2 U2

Insert at *: D' F2 D B2 D' F2 D B2

Good luck on your next competition, Guus!
 
Got to try this one this evening....
Will be competing with Guus this saturday, my first FMC!

Pfff... How do you guys do this?

L F D L' D Guus' start
B U' F R D R2 D' F' R U' R B2 16 moves make F2L + EO minus corner (or minus slot) depending what you take as D face

Two versions from here:
working with orange as bottom layer
D R' U' R D' R' U
B' R' B R' B' R2 B leaves 3 corners after 30 moves

or
working with green as bottom layer
U B' U B R' U R U'
L' B2 L B L' B L B leaves 3 corners after 32 moves

Somehow having 4 edges oriented too soon throws me competely off...

Oh well, I got a 29 with my own solution :-)
 
Last edited:
You must log in or register to reply here.

Similar threads

P
Weekly 22-49 FMC Speed Solve 1 Solutions
Replies
0
Views
949
PiEquals3.14
P
R
Heise Method Discussion thread
10 11 12
Replies
238
Views
66,853
abunickabhi
abunickabhi
C
[NR] [CR2, WR3] Nachuan Sun 19.67 FMC Mean
Replies
5
Views
437
abunickabhi
abunickabhi
OtaMota
[AsR] Jaehyun Jeong 19.33 FMC Average and 17 FMC Single
Replies
17
Views
1,296
OtaMota
OtaMota
AJT17
[SAR] Brayan Alexander Sandoval Camacho 21.33 FMC mo3
Replies
3
Views
476
acrystal
acrystal
Back
Top