• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 35,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

The FMC thread

WoowyBaby

Member
Joined
May 27, 2018
Messages
362
Location
United States
WCA
2018ISOM02
YouTube
https://www.youtube.com/channel/UCzgIKWl
An Interesting FMC Skeleton-

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

L2 F L' // EO (3)
U' L2 B F // Five Pairs (4)
U2 B L2 F2 // Combine (4)
U2 D // Two Pairs (2)
B' U2 D2 F // Combine (4)

AB2C2E in 17

This is an example of what I call “freebuilding”, where you don’t have a specific blockbuilding substep or goal, such as 2x2x2 or F2L-1. I believe I shared a 23 move solution recently in this thread using this idea. I think this way of solving is “objectively superior” but I want to know what you think about it.

But also, I don’t know many 2c2e algs, only like the 10 move J-Perms, and really I’m just not great at insertions in general, so if you could show me a nice way to finish my FMC practice solve that’d be great :)

I saw that a lot of good fmc solves almost always begin with EO. Is that just a generally good way to start the solve instead of immediately looking for blocks?
EO is a very nice way of beginning a solve, especially if it’s only 3 moves or so, and it also usually makes blocks easier to build. EO is actually the very first thing I check whenever I start FMC.
 
Last edited:

xyzzy

Member
Joined
Dec 24, 2015
Messages
1,592
I've been doing a bunch of practice solves, and I just can't break the sub 30 barrier.

Here's my latest one, I got pretty close but couldn't find a good insertion:

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

R2 U' L U' F2 //2x2x2
B' L2 B2 L B //2x2x3 + preserve pairs
(U B' U2 B) //F2l-1 + preserve pair
(L2 U L U') //square + pair on last layer
(F' L' F U L' U' L) //orient and place edge pair and square

Skeleton: R2 U' L U' F2 B' L2 B2 L B L' U L U' F' L F U L' U' L2 B' U2 B U' *
Insertion: * U' B2 U F U' B2 U F'

32 moves
Couldn't find a better insertion with more than 1 move cancel. Checked insertion finder later and it also only gave 1 move cancels. Unlucky.
Having a bunch of pairs around doesn't mean you have to preserve all of them. Usually preserving them is good, but sometimes you just get five pairs or something ridiculous and there's no reason to keep all of them. Having two pairs around after 223 is fine, but don't spend too many moves trying to preserve them.

Here's one trick I like (but probably isn't that useful in general—it's not even useful here). Any time you have a pair connected to the centre, but with a wrong edge next to it, you can always solve the opposite edge and align with premoves.

R2 U' L U' F2 B' L2 B2 L B // 223 (10/10)
L F' L' F L' // EO, finish cross, insert corner and edge in different slots (5/15)
D (D) // keyhole magic (2/17)
// then finish the skeleton or do whatever, e.g.
L' U' L U' L' U' L U L U2 L2 U' L2 U' L2 U2 L // LSLL (17/34)

---

Going back to your solution, after your F2L-1 you have all of the red F2L solved except for one corner, and two flipped edges on orange along with a pair. The two flipped edges aren't opposite colours, so some kind of fruruf will be able to solve the edges; fruruf also preserves a pair, but unless you have that already memorised, you probably don't know whether it'll preserve the orange-blue-yellow pair. So let's try it and see what happens:

R2 U' L U' F2 B' L2 B2 L B // 223 (10/10)
(U B' U2 B) // F2L-1, two bad edges (4/14)
(B D L D' L' B' L2) // fruruf to solve edges; ab4c (7-1/20)

20 to 4c isn't very likely to give you sub-30, but you could keep this skeleton as a backup. Another alg for the same LL edges case (opp flip, not opposite colours) is sexysledge, which also preserves a pair somewhere (along with flipping two pairs, but you can see that the two flipped edges you have aren't part of pairs so that doesn't matter):

R2 U' L U' F2 B' L2 B2 L B // 223 (10/10)
(U B' U2 B) // F2L-1, two bad edges (4/14)
(L2 U L U' L' U' F U F') // sexysledge to solve edges; ab3c (9/23)

23 to 3c is much more likely to give you a sub-30; you just need the insertion to cancel two moves and you'll get a 29. Indeed, IF says the optimal insertion cancels two moves:

R2 U' L U' F2 B' L2 B2 L [@1] B F U' F' U L U L' U' L2 B' U2 B U'
Insert at @1: L B2 L' F2 L B2 L' F2
Fewest moves: 29. 2 moves cancelled
The final solution: R2 U' L U' F2 B' L2 B2 L2 B2 L' F2 L B2 L' F' B U' F' U L U L' U' L2 B' U2 B U'

I saw that a lot of good fmc solves almost always begin with EO. Is that just a generally good way to start the solve instead of immediately looking for blocks?
I rarely ever use EO starts and I strongly prefer blockbuilding starts. 80% of the time when I do an EO start and build blocks from there, I end up having two of the "wrong" faces left to solve… (E.g. EO on the F-B axis, and I solve a block on DB, leaving U and F faces free. But F and F' moves destroy EO!)

But yes, I think community consensus has shifted towards EO starts being good in general (and I'm just whining because I suck at it).

An Interesting FMC Skeleton-

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

L2 F L' // EO (3)
U' L2 B F // Five Pairs (4)
U2 B L2 F2 // Combine (4)
U2 D // Two Pairs (2)
B' U2 D2 F // Combine (4)

AB2C2E in 17

This is an example of what I call “freebuilding”, where you don’t have a specific blockbuilding substep or goal, such as 2x2x2 or F2L-1. I believe I shared a 23 move solution recently in this thread using this idea. I think this way of solving is “objectively superior” but I want to know what you think about it.

But also, I don’t know many 2c2e algs, only like the 10 move J-Perms, and really I’m just not great at insertions in general, so if you could show me a nice way to finish my FMC practice solve that’d be great :)
Two different 2c2e insertions ("pick one", not "do both").

L2 @1 F L' U' L2 B F U2 B L2 F2 U2 D B' @2 U2 D2 F
@1 = [U : U2 L F2 L' U2 B2 L' D2 L B2 R'] // T perm; 12 moves
@2 = [U' : L2 F L F' L D2 R' B R D2] // J perm; 12 moves cancelling 3 = 9

Didn't find any other place in the skeleton to insert short 2c2e algs. The T perm alg is R2 u R2 u' R2 F2 u' F2 u F2 U'; you can replace every other u/u' with U/U' too, but this is unlikely to cancel more moves. (Usually, if you can cancel more moves, you can also insert a 10-move J perm instead of an 11-move T perm.)

In my experience, unrestricted blockbuilding is cool when it works, but really annoying when it doesn't. With too many blocks everywhere, you might be tempted to waste moves preserving all of them.
 
Last edited:

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,140
Location
singapore
WCA
2014WENW01
Solutions+thoughts for my 24.33 NR (AsR #2) mean at Medan Farewell 2019:


R' U' F L2 B2 R F2 R' D2 L2 R U2 F U R' F' L2 D2 F' R2 U L2 F2 R' U' F, 25 moves

(B) R U' B'//EO (4/4)
D L (R2 L2 B2 U2 L' U L)//DR (9/13)
(D2 L2 D' R2 D F2)//6e (6/19)

R U' B' D L F2 D' R2 D L2 D2 * L' U' L U2 B2 L2 R2 B'

*=D2 L2 D2 R2 ** U2 R2 (6-5/20)
**=R2 F2 L2 D' L2 F2 R2 U' (8-3/25)

Final: R U' B' D L F2 D' R2 D' F2 L2 D' L2 F2 R2 U R2 L' U' L U2 B2 L2 R2 B'

This is my first official solve with domino reduction, insertions were lucky though

R' U' F D B2 U2 L2 F2 U B2 D L R' B D2 R2 F' U R' F' R' U R' U' F, 25 moves

F' (R' U F')//EO (4/4)
(U D R') U' D L2 R D' R'//DR (9/13)
(B2 U' R2 B2 U')//2e2e (5/18)
(U R L' B2 D U' L D' U L R')//solved (11-4/25)

Final: F' U' D L2 R D' L' U' D L' U D' B2 R' L B2 R2 U B2 R D' U' F U' R

I found a 21 to 3c backup solution that ended with 26 final so I was fairly relaxed for the rest of the attempt. Found this 2e2e at 40 minutes. Turns out optimal is 24 with some nested cancellations. I don't think I would have found that even if I had more time.

R' U' F R2 B U2 R2 F U2 B U2 F' D B' R D R' F L2 F2 D U2 B2 R' U' F, 23 moves

L (L U' L)//EO (4/4)
(U F' U') F2//222 (4/8)
D B' D (D2)// (4/12)
(R2 B' D' R2 D)//5c (5-2/15)

L F2 D B' R2 D B R2 * D2 U F U' L' U L'

*=R ** U R' D2 R U' R' D2 (8-3/20)
**=R D R' U R D' R' U' (8-5/23)

Final: L F2 D B' R2 D B D R' U R D' R2 D2 R U' R' U F U' L' U L'

I knew that I needed a 26 to break NR mean, so when I found this 15 to 5c at 20 minutes I knew I was most likely going to succeed. My past 4 official 5cs have been +12 moves optimal so it's nice to finally have a good cancellation
 

LambdaPi

Member
Joined
Jun 17, 2019
Messages
8
Location
United States
YouTube
doesn't fit lol
I noticed that in many of people's EO often involves a NISS. When do you decide that you want to go inverse and is it really that much better?

Solutions+thoughts for my 24.33 NR (AsR #2) mean at Medan Farewell 2019:


R' U' F L2 B2 R F2 R' D2 L2 R U2 F U R' F' L2 D2 F' R2 U L2 F2 R' U' F, 25 moves

(B) R U' B'//EO (4/4)
D L (R2 L2 B2 U2 L' U L)//DR (9/13)
(D2 L2 D' R2 D F2)//6e (6/19)

R U' B' D L F2 D' R2 D L2 D2 * L' U' L U2 B2 L2 R2 B'

*=D2 L2 D2 R2 ** U2 R2 (6-5/20)
**=R2 F2 L2 D' L2 F2 R2 U' (8-3/25)

Final: R U' B' D L F2 D' R2 D' F2 L2 D' L2 F2 R2 U R2 L' U' L U2 B2 L2 R2 B'

This is my first official solve with domino reduction, insertions were lucky though
For example in that solve, you used a switch to solve EO, however there is a 4 move EO on the normal scramble as well, and I can't immediately see why one is better than another. You do a B move on the inverse, so does that have something to do with the DR step, like with you orienting 3 corners and placing one E layer edge on the E layer? Or is it something else I am missing?
 
Joined
Feb 23, 2019
Messages
389
Location
The FitnessGram Pacer Test is a multi stage...
I noticed that in many of people's EO often involves a NISS. When do you decide that you want to go inverse and is it really that much better?



For example in that solve, you used a switch to solve EO, however there is a 4 move EO on the normal scramble as well, and I can't immediately see why one is better than another. You do a B move on the inverse, so does that have something to do with the DR step, like with you orienting 3 corners and placing one E layer edge on the E layer? Or is it something else I am missing?
It’s important to remember that FMC isn’t linear in the 1 hour format that we have in competitions. Wong probably tried all of the possibilities that he thought were viable. I personally go on the inverse when I find something not-so-great on the normal scramble. Please correct me if I’m wrong, but one has a lot of time to play around with things during FMC, and what you’re tackling about ties into that.
 

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,140
Location
singapore
WCA
2014WENW01
I noticed that in many of people's EO often involves a NISS. When do you decide that you want to go inverse and is it really that much better?



For example in that solve, you used a switch to solve EO, however there is a 4 move EO on the normal scramble as well, and I can't immediately see why one is better than another. You do a B move on the inverse, so does that have something to do with the DR step, like with you orienting 3 corners and placing one E layer edge on the E layer? Or is it something else I am missing?
I tried the 4 move EO on normal, as well as the 3 move EO on the inverse, but I couldn't find any good continuations off that

I then noticed if I do a B on inverse it will leave a 3 move EO on normal so I gave that a try too.

If that did not work out I would most likely have tried EO on a different axis but luckily it worked out.
 

asacuber

Member
Joined
Nov 4, 2015
Messages
1,398
Location
India
WCA
2015SANT44
YouTube
channel/UC5znrjIR5WXZMlAubzUAG-w
I tried the 4 move EO on normal, as well as the 3 move EO on the inverse, but I couldn't find any good continuations off that

I then noticed if I do a B on inverse it will leave a 3 move EO on normal so I gave that a try too.

If that did not work out I would most likely have tried EO on a different axis but luckily it worked out.
How do you determine how many moves it takes to do EO on the inverse without actually going to the inverse?
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
1,592
How do you determine how many moves it takes to do EO on the inverse without actually going to the inverse?
(Assuming white-top green-front; adjust accordingly otherwise.)

If the white-green edge is bad on normal, then the UF edge is bad on inverse; if the white-blue edge is bad on normal, then the UB edge is bad on inverse; etc. It also works the other way: if the UF edge is bad on normal, then the white-green edge is bad on inverse, and so on. If you make use of this, you'll never have to actually do the inverse scramble to find where all the bad edges are.
 

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,140
Location
singapore
WCA
2014WENW01
(Assuming white-top green-front; adjust accordingly otherwise.)

If the white-green edge is bad on normal, then the UF edge is bad on inverse; if the white-blue edge is bad on normal, then the UB edge is bad on inverse; etc. It also works the other way: if the UF edge is bad on normal, then the white-green edge is bad on inverse, and so on. If you make use of this, you'll never have to actually do the inverse scramble to find where all the bad edges are.
This, I saw that doing (B) resulted in 2 blue bad edges so I knew it was a decent start
 

asacuber

Member
Joined
Nov 4, 2015
Messages
1,398
Location
India
WCA
2015SANT44
YouTube
channel/UC5znrjIR5WXZMlAubzUAG-w
(Assuming white-top green-front; adjust accordingly otherwise.)

If the white-green edge is bad on normal, then the UF edge is bad on inverse; if the white-blue edge is bad on normal, then the UB edge is bad on inverse; etc. It also works the other way: if the UF edge is bad on normal, then the white-green edge is bad on inverse, and so on. If you make use of this, you'll never have to actually do the inverse scramble to find where all the bad edges are.
This makes sense. Thanks :)

Edit: Haven't spent much time on this scramble but 2 people got 21s on this: R' U' F L2 D R2 D B2 L2 R2 F2 D R F2 D' R2 F R' U L' D2 L' U2 R' U' F
Northern Virginia Summer 2019, scr 3.
 
Last edited:

Cale S

Member
Joined
Jan 18, 2014
Messages
2,401
Location
Iowa, USA
WCA
2014SCHO02
YouTube
mathtornado7
Edit: Haven't spent much time on this scramble but 2 people got 21s on this: R' U' F L2 D R2 D B2 L2 R2 F2 D R F2 D' R2 F R' U L' D2 L' U2 R' U' F
Northern Virginia Summer 2019, scr 3.
The 21 solution:
L F2 D B F R' // 222 + pairs
L B L B2 L' // F2L - 1
L D L' D' L D L' // F2L
L' D' F' D F L // LL

definitely not something I would ever try :p
 
Joined
Sep 15, 2017
Messages
423
WCA
2017BRYA06
YouTube
channel/UCj_pE_D
This makes sense. Thanks :)

Edit: Haven't spent much time on this scramble but 2 people got 21s on this: R' U' F L2 D R2 D B2 L2 R2 F2 D R F2 D' R2 F R' U L' D2 L' U2 R' U' F
Northern Virginia Summer 2019, scr 3.
The 21 solution:
L F2 D B F R' // 222 + pairs
L B L B2 L' // F2L - 1
L D L' D' L D L' // F2L
L' D' F' D F L // LL

definitely not something I would ever try :p
Lol yeah, I found that in like the first 20 minutes of the attempt. I already had a 28 and a DNF(36), the DNF was because I thought I had the insertion written wrong, so fixed it, but it turns out it was right all along lol. Then on the last attempt, I was still pretty mad about the DNF (because I wanted to qualify for nats) so I was just angrily messing around with the cube and found this lol. I jumped up really quickly (kinda making a scene lol) and turned it in, Blake saw this and was like.. hey maybe I should try something different, and found this.
 

Lvl9001Wizard

Member
Joined
Jan 23, 2019
Messages
12
Rather than being scared by DR everytime I do EO first, I decided to do some practice...took 2 random scrambles from CS timer, would appreciate advice on how to do things differently and more efficiently

Note: for the sake of time I did not bother with NISS, and only did EO on the F/B axis, placed E layer edges and oriented U/D corners.
(Both still took 20 mins each rip)

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

F' D F // EO
D' U' L'
D' U' L' U L
U' L' U L // DR (15 moves)

U R2 U' L2 R2 F2 // why not lol 3e4c


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

U2 F R L U B // EO
D' U R U R'
U L D2 L' // DR (15 moves but not many good continuations)

What's a "good" movecount to do DR in if you are aiming for sub 30? For reference, 2x2x3 in 9 or 10 moves would be a "good" movecount for sub 30
And sorry for not doing any research lol (there's tons of DR solves in this thread but I cbs looking at all of them) but what's the average number of moves taken to do DR given the whole solution was <25 moves.

Thanks
 

porkynator

Member
Joined
Oct 27, 2010
Messages
1,264
Location
Belluno, Italy
WCA
2011TRON02
YouTube
PorkyDays
I think 10 is very good for DR. One should probably aim at 12 or less. But I admit I only use DR when I stumble upon a good one, not so often :)

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

F' D F // EO
D' U' L'
D' U' L' U L
U' L' U L // DR (15 moves)
After your EO you could have done:
D2 U' L'
D2 U2 L' U L //DR (11 moves)
 

LambdaPi

Member
Joined
Jun 17, 2019
Messages
8
Location
United States
YouTube
doesn't fit lol
Inspired by lvl9001wizard I too decided to try a dr solve. Did the first one on the competition website (spent a lot more than 1 hour) and got this:

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

(L B' D2 L' F') //EO
(D' R) D R2 L2 F2 U D2 R //DR (i thought huh not bad 14 moves)
D' F2 U2 R2 U D' L2 D' //22 to 2e2c

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

I would consider this a pretty lucky solve (especially since i am abysmal at the part after DR, and not too good at the DR step either), however the 2e2c really tripped me up and made this a bad solve.

I have no practice doing 2e2c and I only know the 11 move jperm for that situation. So would I need to just learn a bunch of those algs for 2e2с insertions?

I ended up just putting a pure R perm at the end on the niss inverse so the final solution is
D R2 L2 F2 U D2 R D' F2 U2 R2 U D' L2 D' D' F2 L' F' D' F' D F L F' D2 F D2 F' R' D F L D2 B L'

(2 D' magically cancelled to D2 so 35 moves)
I was also considering a sune+corner com combo, but didn't try that yet.

Ok i guess for my first complete DR solve, but I would appreciate any tips or suggestions you guys can give.
 
Last edited:

LambdaPi

Member
Joined
Jun 17, 2019
Messages
8
Location
United States
YouTube
doesn't fit lol
Appologies for double post, but I just got my first sub 30 :D yay.

2nd scramble from this week's fmc comp
Scramble: R' U' F L2 U2 F2 D' L2 D L2 D2 L2 U L' D' U2 B2 R' B' R' U L B2 D R' U' F

(R2 F D2 L B') //EO (5/5)
L2 U L' U F2 D' R2 D L' //DR (9/14)
(U' R2) //3x2x1 + line (11/16)
U D F2 U D L2 D //finish to 3e (7/23)

checked for inserts
o **** 6 move com in beginning
o **** cancel 1

Skeleton: * L2 U L' U F2 D' R2 D L' U D F2 U D L2 D R2 U B L' D2 F' R2
Insertion: U2 R L' B2 R' L *

Cancel 1 move

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

28 moves

Insertion was lucky af in my opinion, considering that this was my first time doing an edge commutator un a solve.

I think this was probably my 10th fmc solve so I am very happy.
 
Last edited:

PingPongCuber

Member
Joined
Oct 20, 2018
Messages
150
Location
Colorado
WCA
2018AMSB02
Hi! I don't know much about FMC, but I was just thinking about the way that you do it in competitions. They give you a simple scramble, but you are not allowed to use any parts of that scramble, right? (please correct me if I am wrong, like I said, I don't know much about FMC).
It seems like it could be fairer if they either gave you a long scramble or a cube pre-scrambled because then the competitor would have more freedom of moves, and people wouldn't have to read the solutions and decide if it is fair enough. There could be organizers at different competitions who are stricter than others, which can make setting records complicated and debatable. What do you guys know about this? Is there something that I am missing?
 
Top