The FMC thread

[DavidWoner]

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?

What kind of scrambles did you use?

 

[kinch2002]

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.

I'm very late in replying to this, but meh. Rather than suggesting you shouldn't rotate at all, I meant that you could scramble in bld orientation so you don't even have to 'learn' that new red=R, orange=L thing. And then rotate lots while solving. And yay you're being CN at FMC - good choice

 

[irontwig]

What kind of scrambles did you use?

25 random moves.

 

[kinch2002]

Inverse: 8.21, 8.16, 7.62, 8.22, 8.27, (8.93), 8.10, 8.53, 8.78, 8.29, 7.78, (7.32) = 8.19
Normal: (6.72), 7.22, 7.49, 9.16, 7.73, 8.50, 8.28, 8.35, 7.03, (DNF), 7.42, 8.29 = 7.95
Ratio = 1.03

25 random moves

 

[guusrs]

I first write down the inversed scramble before I use it.
It lowers my failure rate substantly

 

[Kryptonite]

Hey guys! Finally back into FMC (it's been nearly 8 months...). I DNF'd last week cause I only caught the end, but I got a solution submitted for this week, hurrah!

Also, I did irontwig's test... Just to warn you guys, I have no speed background, and the ONLY algorithms I know are the commutators I've done over and over. The first solve method I learned, believe it or not, was human thistlewaite.

(DNF), 37.79, 31.57, (30.01), 34.58, (DNF), 30.89, 31.56, 31.58, (DNF), 36.14, 30.21 =32.70

Now the real kicker.... the times to solve the cube from my DNF were about 4 minutes each :0 (4:24.79 and 3:57.25). Though one of them left a simple 7f* that I was able to figure out in about a minute (didn't time it).

Apparently I'm going to need to learn to speed solve before I ever go for a 1-hour...

 

[irontwig]

Even if you'r not too into speed just random block building and then solving the rest with comms seems like it should take way less 4 minutes.

 

[Erik]

My 29 move solution at Cachan open 2011:

Scramble: U2 L' U2 L' R' B2 F2 L' D2 U L2 B R D' B R B2 D2 U F2 R' (21)
Inverse scramble: R F2 U' D2 B2 R' B' D R' B' L2 U' D2 L F2 B2 R L U2 L U2

It was hard to find something matching instantly, so I decided to try something I never tried before: orienting all edges first (ZZ).
D2 R F orients all edges nicely though I couldnt find anything nice after this. Using this as premoves (F' R' D2) on the inverse scramble got me this:

Pre: F' R' D2
2x2x2 block: U2 L' U' L
2 more 1x2x2 blocks: R U R2 U2
All edges: R2 D' R' D R2 F' R' D2 leaving 5 corners.

I decided to first invert this so I have 5 corners left on the normal scramble to avoid time pressure of inverting the whole solution at the end.
Skeleton:
D2 R F . R2 D' R D R2 U2 R2 U' R' L' U L U2 (16) after 16 minutes.

After 40 minutes I found this at the . F' L2 F R' F' L2 F R which cancels nicely, leaving 3 corners after 21 moves.
Stupidly I didn't find anything for the last 3 corners (FAIL) and had to solve them with a stupid niklas....... -.-

Final solution:
D2 R L2 F R' F' L2 F R' D' R D R2 U2 R2 U' R' L' U L U2 ... B L' F' L B' L' F L (29) sub-30, but it could've probably at least be a 26 when I wouldn't have failed so much at the last 3 corners.

 

[Kryptonite]

Haha, don't get me wrong, I know my time is pretty atrocious. I've just never even thought about going faster, and I had no idea I was THAT slow. For the most part though, I just don't do normal solving. I pretty much only do FM attempts and Backtracking.

 

[Cubenovice]

Kryptonite: with minimal study / practice of F2L with 4 look last layer you should be able to get your solves below the 60 second mark.
A good place to start is Badmephisto’s site and youtube channel with very good explanation of F2L and LL methods.

I am also a slow solver, my average is around 40s and like you I do not practice “normal” solving.
But having some knowledge of CFOP and algs allows me to solve on “auto-pilot” and keep focus on the FMC scramble at hand.

Nice to see another Human Thistlethwaite solver!
I used it for FMC for a while just to find the method’s potential: typically low 40’s, some sub 40’s with best results being 35 and 36 HTM

No scrambling times yet, currently at work so cannot cube…

 

[Kryptonite]

Yeah, Thistlethwaite is really interesting! I've never actually used it in an FMC, because it seemed like other mothods just had more going for them, but it's a fun exercise. It used to be my primary method of solving when I botched a scramble or move set along the way. I figured it would be pretty hard to achieve a sub-30 though.

 

[JasonK]

Yeah, Thistlethwaite is really interesting! I've never actually used it in an FMC, because it seemed like other mothods just had more going for them, but it's a fun exercise. It used to be my primary method of solving when I botched a scramble or move set along the way. I figured it would be pretty hard to achieve a sub-30 though.

Thom Barlow (Kirjava) has got sub-20 with it Go to 7:13 for Thistlethwaite.

 

[Kryptonite]

Impressive. Though, I meant sub-30 moves, not seconds

 

[JasonK]

Impressive. Though, I meant sub-30 moves, not seconds

That would make sense, considering this is an FMC thread... My bad :fp

 

[Cubenovice]

Question about solving corner 5-cycles via insertions.

I was under the impression that the choice of your first 3-cycle influences the possible insertions for the remaining 3-cycle.
Now for fmc.mustcube 370 I got to 5C after 18 HTM, I sticker them 12345.

I find several 3-cycles that cancel 2 moves:
145 leaving 123
345 leaving 125
123 leaving 345
234 leaving 145

I checked all four of the resulting 24 HTM L3C skeletons and in all of these I cannot do better than 2 moves cancel for a final result of 30 HTM

Is this just coincidence or is there some cube law preventing me from finding anything better for this 5-cycle?

 

[Julian]

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

Got this scramble for speed yesterday. Decided to try FMC, got this in ~30 mins with no notes.

D' L B L' // 2x2x2 + 2x2x1 + CE pair
R' D F D R F2 R2 // F2L-1
D R D R' D2 R D2 R' // F2L
R' D' B D R D' R' B' R // OLL

Final solution (27 HTM): D' L B L' R' D F D R F2 R2 D R D R' D2 R D2 R2 D' B D R D' R' B' R

I was just about to start looking for insertions, when I found the PLL skip.

 

[Kryptonite]

@WTF2L Lol, np man, realistically sub-30 in the cube community usually doesn't mean moves. And the video was awesome anyway! haha.

I was under the impression that the choice of your first 3-cycle influences the possible insertions for the remaining 3-cycle.

This is true, and you can show it with cycle theory. Your goal is to cycle (12345), which can say be made up of (125)(534):
1 -> 2,
2 -> 5 -> 3
3 -> 4
4 -> 5
5 -> 1

However, you can't change the order, beacuse (534)(125) =/= (12345):
1 -> 2
2 -> 5
5 -> 3
3 -> 4
4 -> 5 -> 1
so (534)(125) = (12534)

We can use cycle notation, however, to find what we do need if we want (125) to be the second insertion.
X(125) = (12345)
X(125)(521) = (12345)(521) [Notice, (521) MUST be on the same side on both sides of the equation!]
X(1)(2)(5) = (1)(234)(5)
X = (234)

To find optimal insertions you must find ALL (123), (234), (345), (451), (512) cycles, then use combine accordingly to cancel the most moves. You may find it's better to use two good insertions (say both cancel two) than your best insertion (say once cancels three, but it's proper pairs don't give cancellations), or vice versa.

This also means that a 5C is five times as complex as a 3C.

 

[Julian]

This also means that a 5C is times as complex as a 3C.

How many times?

 

[Kryptonite]

Whoops, five times. Edited my post above, thanks!

For FMC #337 I wrote down sixty eight 8f insertions from a 19 move skeleton with a 5C remaining.

I found
5 that canceled three moves
9 that canceled two moves
8 that canceled one moves
The rest didn't cancel, and I didn't write them all down.

This is a little excessive; I mostly did it for the practice of doing everything, so next time I would understand why and when I didn't have to.

Also, if you insert within an insertion, the one that finishes first counts first, not the one that starts first.

 

[kinch2002]

For 5 cycles, I see two ways of doing it - one totally thorough, the other one good if you have less time left.

You have the five cycles - (123) (234) (345) (451) (512). Go through every move in your skeleton, marking down how many moves each of those cycles would take. Then look at all combinations and find the best combination. Combinations are (123)(451), (234)(512), (345)(123), (451)(234), (512)(345).

You must note this important fact: If you see that there is a nice 3 move cancellation on say a (123) cycle, you can't just then look on your piece of paper for the best (451) cycle and put that in, because depending on whether the cycle is after or before the (123) cycle, it might not work. So for this case, you'd have to look for either a (345) insertion that is before the (123) insertion, or a (451) insertion that is after the (123) insertion. This is all assuming you just want to go through the solution once and never have to rescramble with the new skeleton that contains a first insertion.

This method is good, but not entirely thorough, because there's always a chance that you could do an insertion inside the first insertion and cancel more moves. For this way of doing things, you'd have to look at all the possibilities for a first insertion (which is probably ~100) and look through the new skeleton (well the moves of the insertion anyway) that leaves just 3 corners for all those cases, just in case there is a massive cancellation there.

EDIT: I think most people tend to do the first way, but they also rescramble with the new skeleton before looking for a 2nd insertion, which I think wastes a lot of time, because they might end up checking through five different skeletons (one for each of the best first insertions) when in fact all they're gaining over the method I've described above is the option to insert a comm inside a comm