[Help Thread] FMC Discussion

[irontwig]

WTF inserting 8-movers for edges?

Well, there's only one case, so it's well worth considering for humans too.

 

[guusrs]

By the way, I thought people here might like to know that I got my best ever fewest moves solve today on the weekly competition 2012-18 scramble: 23 moves! I think it may not be so much that it was a lucky scramble as it was that I got a lucky solve (I definitely got very lucky), but I'm interested in seeing what other people will come up with.

Woooow Mike, Congratz. You deserve it!

If you like a lucky scramble you should try this week 1-hour scramble of fmc.mustcube.net. Actually I invite every FMC lover to give this beauty a try!

 

[Mike Hughey]

If you like a lucky scramble you should try this week 1-hour scramble of fmc.mustcube.net. Actually I invite every FMC lover to give this beauty a try!

Thanks - it was a lucky scramble. Unfortunately, as usual, I seem to be not very good at lucky scrambles - there are just too many possibilities and I get lost in them. I got a lucky 29 move solve on it due to a PLL skip in 10 minutes, but I couldn't come up with anything better after an hour.

 

[nickvu2]

I was just told that you can write down cube turns (x, y, z) in the solution but that they don't count towards the total move count. Is this right!? Looking at the HTM definition I guess so.

Half Turn Metric (HTM) is defined as:
12a17) Each move of the categories Face Moves and Double Outer Slice Moves is counted as 1 move.
12a18) Each move of the categories Inner Slice Moves and Middle Slice Moves is counted as 2 moves.
12a19) Each rotation is counted as 0 moves.

So why don't I see anyone making use of this??? Is there some advantage to translating algs, edge control, etc. rather than using cube turns to put LL on the U layer?

 

[Cubenovice]

I was just told that you can write down cube turns (x, y, z) in the solution but that they don't count towards the total move count. Is this right!? Looking at the HTM definition I guess so.

So why don't I see anyone making use of this??? Is there some advantage to translating algs, edge control, etc. rather than using cube turns to put LL on the U layer?

Correct: cube rotations do not count as moves.

Some people do use rotations in their solutions, these people may also overlook cancellations or mess up NISS solves

Writing without rotations makes everything so much easier: you can keep flipping the cube over without ever having to worry if you get the rotations right or miss a cancellation. Writing without rotations actually means less translating.

 

[nickvu2]

Correct: cube rotations do not count as moves.

Some people do use rotations in their solutions, these people may also overlook cancellations or mess up NISS solves

Writing without rotations makes everything so much easier: you can keep flipping the cube over without ever having to worry if you get the rotations right or miss a cancellation. Writing without rotations actually means less translating.

Haha, ok that makes sense. Thanks for clearing that up for me =)

 

[Mikel]

I was just told that you can write down cube turns (x, y, z) in the solution but that they don't count towards the total move count. Is this right!? Looking at the HTM definition I guess so.



So why don't I see anyone making use of this??? Is there some advantage to translating algs, edge control, etc. rather than using cube turns to put LL on the U layer?

E1a) The maximum length of a solution is 80 (moves and rotations).

They count only if you are approaching 80, so if you have say 77 moves and 4 rotations, 77+4=81 so you DNF. But if you have 70 moves and 4 rotations, 70+4=74 so you have a 70 move solution.

I guess this really only pertains to people who aren't great at FMC, aka myself.

 

[Sebastien]

Thanks - it was a lucky scramble. Unfortunately, as usual, I seem to be not very good at lucky scrambles - there are just too many possibilities and I get lost in them. I got a lucky 29 move solve on it due to a PLL skip in 10 minutes, but I couldn't come up with anything better after an hour.

I quickly managed to find a 27 move solution and spent like 2 hours in total on the scramble, found like 5 or 6 different sub30 move solution but none better than 27 :/

Can't wait to see if and how others did better!

 

[kinch2002]

I while back I mentioned a possible video FMC 'tutorial'. I finally have a camera (thanks Mollerz) so I intend to film some stuff soon. As it essentially impossible to give a tutorial on FMC, my intention is to do several videos on topics, giving a range of ideas. The overall intention is to stop people from having to answer the same questions over and over again by having a video to link people to. I know there is a fair amount of text dotted all over the internet, so I'll be having a look over that too for ideas. I'd appreciate feedback on the general topics rather than specific tips to include.

1. Basic overview: Includes some important rules, some of the methods that are used, and a few general tips. Kind of like an FAQ.
2. Starting a solve and finding continuations
3. Finishing a solve/skeleton
4. Insertions (corners inc. 4 and 5 corners left, and edges too)
5. Pseudo blocks and using the inverse scramble (including NISS ofc)

 

[kinch2002]

A couple of random fun scrambles I tried tonight

Spoiler

[21:12] <+RobertY> 10.89 OH
[21:12] <+RobertY> B2 U2 F2 L2 D B2 F2 D2 U' R2 L2 F' D' L' D R B R F D2 R
CFOP ish solve, so I'll add rotation in there at the start. Rob only did up to F2L#2 the same during the solve. I found the rest by randomly playing around.
Cross: z D' B
1: F U' F'
2: U' L U' L'
3+4: U' R' U' R U' R' U' R2 U' R'
LL: R U B R B' R' B R B' R' U' R' U'
28 moves

Spoiler

[22:23] <+Brest> F2 U' R2 B2 D' L2 D L2 U2 R2 U' R D U2 L2 D' F2 L B' U' R' U'
R' B' L' F' D' F2 R U2 F' found by Robert-Y
L2 U L2 B L2 B' L2 U' L' found by me
18 moves to 3 corners
Insertions finder says 25 moves total

 

[guusrs]

A very nice scramble at the weekly competition on this forum (2012-20).
I had a crazy solve.
I' will post it next monday...

 

[cubernya]

What's the UWR for most moves cancelled to final solution ratio? Just wondering because I just got a 28 move solution with 12 moves cancelled (12:28 = 42%)

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

U' R2 D' B // EOLine (4/4)
R' F2 R L2 U' L2 U F2 // Left Block (8/12)
U' R' U2 D' R U2 L U' R' U L' D // Right Block + Edges (12/24)

U' R2 D' B R' F2 L2 [.] R U' L2 U F2 U' R' U2 D' R [,] U2 L U' R' U L' D
. = U' L2 U R U' L2 U R'
, = L D2 L' U2 L D2 L' U2
U' R2 D' B R' F2 L2 U' L2 U R U' L2 U R' R U' L2 U F2 U' R' U2 D' R L D2 L' U2 L D2 L' U2 U2 L U' R' U L' D
U' R2 D' B R' F2 L2 U' L2 U R F2 U' R' U2 D' R L D2 L' U2 L D2 U' R' U L' D
28 turns (12 cancelled)

 

[okayama]

What's the UWR for most moves cancelled to final solution ratio? Just wondering because I just got a 28 move solution with 12 moves cancelled (12:28 = 42%)

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

U' R2 D' B // EOLine (4/4)
R' F2 R L2 U' L2 U F2 // Left Block (8/12)
U' R' U2 D' R U2 L U' R' U L' D // Right Block + Edges (12/24)

U' R2 D' B R' F2 L2 [.] R U' L2 U F2 U' R' U2 D' R [,] U2 L U' R' U L' D
. = U' L2 U R U' L2 U R'
, = L D2 L' U2 L D2 L' U2
U' R2 D' B R' F2 L2 U' L2 U R U' L2 U R' R U' L2 U F2 U' R' U2 D' R L D2 L' U2 L D2 L' U2 U2 L U' R' U L' D
U' R2 D' B R' F2 L2 U' L2 U R F2 U' R' U2 D' R L D2 L' U2 L D2 U' R' U L' D
28 turns (12 cancelled)

I'm not sure what is the UWR exactly, but at least I got 20-move solution with 11-move cancellation.
See my solve in Weekly competition 2010-21. 11:20 = 55%.

 

[Pyjam]

And me, 18 moves cancelled for 25-move final solution = 72%

http://www.speedsolving.com/forum/showthread.php?13599-The-FMC-thread&p=739690&viewfull=1#post739690
(Mustcube 390)

Useless record.

 

[Erik]

NEMO Open 1st place: 28 moves

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

Solution

2x2x2: L F2 B L U L (6)
Extend to triple X-cross: B2 R U2 R B R (6)

Switched to the inverse now using of course premoves: R' B' R' U2 R' B2 L' U' L' B' F2 L'

Pair plus OLL: B' L U' L' U L' (6)
Cancel into J-perm: B2 R' B2 L B' L' B2 R L B' (10)

Full solution:
L F2 B L U L B2 R U2 R B R B L' R' B2 L B L' B2 R B2 L U' L U L' B (28)

 

[kinch2002]

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

Solution

2x2x2: L F2 B L U L (6)
Extend to triple X-cross: B2 R U2 R B R (6)

I thought this finish is quite obvious
F2L-1 as above
B L' B' L U B' U' B' U B U' B2 (24)

 

[Erik]

Osh**** why didn't I see that????

 

[Sebastien]

Osh**** why didn't I see that????

'cause pure LL-Skips are so lame

Edit: 35, 34, 30, 30, 27, DNF, 27, 29, 28, 26, 29, 29 = 29.8 Average of 12 with my last 12 official attempts! Can't wait to roll this!

Edit2: U' L U L' B2 U B U' B2 as finish = 21 move solution. WTF

 

[Cubenovice]

Edit2: U' L U L' B2 U B U' B2 as finish = 21 move solution. WTF

And so £&é%king straightforward too...
Crazy...


THX for making me feel extra stupid posting my fail stuff now.

Short story:
Could have been a 31 HTM but as usual I fail under time pressure.

Long story

Spoiler

F' U' L F B2 D2 L2 D B2 R' F2 U D L2 U B2 U L2 U2 F2

R' pseudo square (R2)
F2 L2 U' - find premove U2
L D' B' D - 2x2x3 – corner and a square
L2 - pseudo F2L-1 - corner (L2)

So we have
premoves L U2 R2
R' - square
F2 L2 U' L D' B' D - 2x2x3 – corner and a square
L2 - F2L-1 - corner in 12 moves
but couldn't find anything decent so eventually settled for
U2 F U2 F' U2 L' U L - leaves 5 corners in 20 HTM
so corrected everything for premoves and started stickering for insertions
and only then noticed the twisted corner...

So then I decided to try something completely different halfway in my solution:
premoves L U2 R2
R' F2 L2 U'
and from there
F' L' F D B U2 B' D' leaves 3E3C in 15HTM
then only had a few minutes left for insertions.
And under time pressure I messed everything up again…

Final skeleton:
* R' F2 L2 U' F' L' F D : B U2 B' D' L U2 . R2 leaves 3E3C in 15HTM (A perm corners)

At . insert L U L’ D2 L U’ L’ D2, no cancellation but beats the A-perm, found from the back and decided to immediately check for edges.
At : the edges can be solved via optimal U-perm but I set it up wrong + forgot to undo a set up move:fp
Cycle L2 U - U2 B’ F R2 F’ B - U’ L2 does not work when you leave out the U’
And it was the wrong cycle anyway; should have been L2 U’ – U2 F’ B L2 B’ F – U L2
On top of it seems I also forgot to copy the ‘ at the last D in the skeleton too.

Later I found F2 R B’ R’ F2 R B R’ for corners at * to cancel one move making this a possible 31 HTM.

And then Insertion Finder tells me that I need to study edge cycles…
There is a 10 move edge cycle that cancels 4 moves for an 28 HTM end result.
Can you find it?

Spoiler

1 corner 3-cycle(s) and 1 edge 3-cycle(s) insertion(s) needed
[@1] R' F2 L2 U' F' L' F D B U2 B' D' L U2 R2
Insert at @1: F2 R B' R' F2 R B R'
After the 1st insertion: F2 R B' R' F2 R B R2 F2 [@2] L2 U' F' L' F D B U2 B' D' L U2 R2
Insert at @2: U' R L' B L2 B' R' L U L2
Fewest moves: 28. 5 moves cancelled
The final solution: F2 R B' R' F2 R B R2 F2 U' R L' B L2 B' R' L F' L' F D B U2 B' D' L U2 R2
Runtime: 52.216004 milliseconds

 

[guusrs]

Edit2: U' L U L' B2 U B U' B2 as finish = 21 move solution. WTF

WTF!
Such a nice obvious 21 move linear solution was just in front of us and we all didn't see it.
I think we do not deserve a new world record.

@Erik: this was the 2nd time you overlooked a 21 move solution after a splendid start and 2nd time this year you were very close to a new world record! After all those years you still amaze me (-;