[Help Thread] FMC Discussion

[Tim Reynolds]

Well, not many FMC solves in this thread recently, but I'll go ahead and post one:
Scramble: L' D2 U' F R U L' D U2 B F L2 U2 R U F' U2 (from Cube Explorer)
Solution: D F2 R U B L B R' B' L' U2 D L' D2 B' L B L2 F L2 F' L F L' U L' U' F' L' (29)

2x2x3: D F2 R U B2 * R' B' U2 (8|8) [from original 2x2x2 D F2 R B U']
F2L minus 1 pair: D L' D2 B' L B (6|14)
F2L: L2 F L F' (4|18)
ELL: F L F' L F L' U L' U' F' L' (8|26)
This leaves a corner 3-cycle; insert B' L B R' B' L' B R (3|29) at *.

Found in 40 minutes; that's two sub-30s in a row! This week I'll try a few of the scrambles that blah used in this thread, just so I can read someone else's solution after I'm done.

 

[blah]

Nice solution, but technically, that isn't ELL

 

[JLarsen]

Tim Reynolds that LL was really nice! I said I was giving up but whatever....I'll try another.

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

2x2x3 (sorta): L D' R U' R2 U F R (8/8)

F2l : B' F L F L' F U F B U B' U2 B (13/21)

OLL: R' U' l' U l F' U R (8/29)

PLL: F U F' U' F' L F2 U' F' U' F U F' L' U' (15/44) I SUCK.

Can anyone help me? Am I to understand there is not help for me?

 

[blah]

OLL + PLL = NO GO.

 

[DavidWoner]

PLL: F U F' U' F' L F2 U' F' U' F U F' L' U' (15/44) I SUCK.

This is why you suck.

Linear solution on your scramble:

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

2x2x3: D' L' F B', U' L2 D2 L F2 L' F (11)
3xcross: U F R' F' B (16)
leave 3 corners: L U L' U' L U L' U (24)
comm finish: B L2 B R2 B' L2 B R2 B2 (33)

 

[JLarsen]

The leaving 3 corners is what I find hard to do. Your solutions seem to magically end in 3 corner cycles.

 

[MTGjumper]

Someone should make some form of tutorial on the various ways to finish FMCs. My technique is flawed: hope to get lucky.

 

[JLarsen]

Someone should make some form of tutorial on the various ways to finish FMCs. My technique is flawed: hope to get lucky.

Well according to Blah that ruins FMC.

 

[blah]

The leaving 3 corners is what I find hard to do. Your solutions seem to magically end in 3 corner cycles.

Exactly what I thought every time I read solutions from the pros. I came to the conclusion that either their reading (refer to an earlier post by Johannes for the definition of "reading") is incredibly deep and accurate, or they keep trying till they get an ending that leaves 3 corners. I'm pretty sure it's a combination of both, with the former being an increasingly important component as one's experience and expertise grows. I figured that since I'm still relatively a noob, I'd work on the latter first until I have enough experience to be able to "read" a dead end before I hit it.

Someone should make some form of tutorial on the various ways to finish FMCs. My technique is flawed: hope to get lucky.

Well according to Blah that ruins FMC.

Well according to Blah that's what he does too. And that is, in his humble opinion, the best way to get good at FMC fast: Try lots and lots of starts until you get lucky.

It's exactly what Kociemba's algorithm does. Find an ending with the optimal start, then move on to find better endings with sub-optimal starts, until you hit something good enough within an hour, then smile to yourself. Being able to find optimal starts is very important for linear FMC, hardly so for "normal" FMC, being able to find LOTS of sub-optimal starts is much more important IMHO

And yes, blah is rather ashamed to use such a "technique", but elegantly or otherwise, it gets the job done He hopes to change his main technique to a more sophisticated one in the near future though, when he has more time to research on other approaches

By the way, I prefer "blah", non-capitalized

 

[JLarsen]

Very good advice thank you blah =P

 

[MTGjumper]

... and blah also prefers speaking in the third person

I may as well try and get lucky with my FMC for this week. I have 50 minutes remaining.

 

[StachuK1992]

Hey guys,

I just really started FMC yesterday, and I hoping for advice on my first real FMC solve.
I did a solve officially like a year ago, but it was a 60move DNF, so that doesn't really count.

I've started the weekly comp. scramble, and this is what I have so far:

Scramble: L2 B2 D' F2 L2 U L2 B2 F2 R2 U L' U' R2 D2 U' F' R' U2 B' R
Scramble with white front, green top.

Solution:
2x2x2:
(6) R2 F' D' R2 F R
This took me about 4min
2x2x3:
(11) U' R' F U F'
another 4ish minutes
EO:
(16) U l L' U L
took another 4 minutes do decide whether the U should be a U, a U', or a U2.

And this is where I'm stuck.
I think I have a decent start; I just don't know what to do from here. :/



I have a few initial ideas from there.


Yellow LL
Left slot first:
(23) F' U2 F' U' F U F'


Right slot first:
(24) M' U2 M F' U F
U F U' F'

Green LL
x y
(21) R2 U' R' U R2
(24) U' R U'
(27) R' U' R
(37) L R' U' R U L' U2 R' U2 R
+ U' + Uperm = 49 :/

Or, I would just do this:

R2 F' D' R2 F R
U' R' F U F'
R2 B' R'
x'
U2 L' U2 L U' L' U L
R U' R'U2 F'U2 F U R U2 R'
G perm (34 + gperm = 46 = not too good :/)

I haven't even bothered to look at the inverse scramble.

 

[guusrs]

Someone should make some form of tutorial on the various ways to finish FMCs. My technique is flawed: hope to get lucky.

Well according to Blah that ruins FMC.

There is an "Fewest Moves Techniques" entry in the speedsolving WIKI.
Some months ago I made a first version of something that might give an overview....
Maybe you guys could help me to enhance this wiki page

Gus

 

[cuBerBruce]

I would like to ask that people not discuss active online fewest move competition scrambles in this thread. People who take part in these competitions are supposed to be coming up with their solutions by themselves without any help or hints from anyone else. IMO, people should be free to read this thread without fear of inadvertently gaining information about an active competition. At the very least, I think spoiler tags should be used if you're going to say anything about an active competition.

 

[StachuK1992]

I would like to ask that people not discuss active online fewest move competition scrambles in this thread. People who take part in these competitions are supposed to be coming up with their solutions by themselves without any help or hints from anyone else. IMO, people should be free to read this thread without fear of inadvertently gaining information about an active competition. At the very least, I think spoiler tags should be used if you're going to say anything about an active competition.

Oh. I'm very sorry.
I won't submit FMC this week, then, or I'll submit an FMC for the inverse, since I didn't ask for that.

 

[JLarsen]

About your solution Stachu:

Step 4 isn't good for FMC. I recommend getting the f2l and ending with a 3 corner cycle by any means possible. I can't do it....but I've heard it's what you're supposed to do.

 

[fanwuq]

Wow! So many people are getting into FMC recently!
To be honest, I've never really focused so much on a solve that would make my head hurt or read much about FMC techniques. I just casually try out random options that look interesting and hope I get lucky. It works pretty well, but I think I've hit a wall at a low 30s average. If I have more time, I would focus on thinking harder on the 2x2x3 block, learning more Heise techniques, optimal corner twisters, more keyhole tricks, Guus's LL techniques, get better at 5 cycle and edge insertions, full ZB...
There are many things to learn that might not guarantee success; mostly it's just trying many options, do unconventional moves, and hope to get lucky.

Stachu,
2x2x3 block is ok, but can be a bit better. It's not so important for this solve. You need to get more creative with finishing the skeleton. Try to just solve F2L minus 1 slot from your 2x2x3 then solve all the edges while taking advantage of available blocks.

 

[DavidWoner]

For those of you who are trying to understand how to leave 3 corners, try taking a look at this solve and try to understand what I did there. It's not magic, merely having control over the pieces. Here is another one with an ok continuation, but also an example of how doing step 4 can help out. I wouldn't say step 4 is "useless" but I don't think it will work out for every solve.

There is also a guide on heise's page about how to leave 3 corners, but its not the greatest.

 

[JLarsen]

For those of you who are trying to understand how to leave 3 corners, try taking a look at this solve and try to understand what I did there. It's not magic, merely having control over the pieces. Here is another one with an ok continuation, but also an example of how doing step 4 can help out. I wouldn't say step 4 is "useless" but I don't think it will work out for every solve.

There is also a guide on heise's page about how to leave 3 corners, but its not the greatest.

I tried to learn it from Hiese. Kind of got me a good start I guess, but it's still trivial.

I checked out your solution. You paired up blocks on the LL while solving the edges.

 

[MTGjumper]

MAJOR SPOILERS

This weeks FMC:

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

I found this 2x2x3: L' D2 U B U L2 B2 L F
Which I extended into a 3xcross: F R' F' R U
Also found alternative 3xCross: L' D2 U B U L2 B2 U L F R F' U2 F

Any experienced FMCers care to show me how to finish this/manipulate pieces during these blocks to get a decent finish?

(I finally ended up with a linear 36: R' U2 R2 F' D' B2 R' B' U F U' B F' D R' D' R' F R F' R F' R' F2 R F' R' D' F' D' L D L' F D F' )