[Help Thread] FMC Discussion

[mark49152]

No, that's not what NISS is.

Right, I was confusing it with just solvng the inverse scramble. Sorry.

 

[jayefbe]

I have always thought NISS was just using the inverse scramble and finding moves that are not needed and cancelling them leaving you with only the core of the solution. Thus, it would make sense that you could get a solution so different that it has a different core.

Obviously that's not how it works (after 5 days and multiple people saying so). The rules of the event are in place to specifically keep solves from simply being the inverse scramble.

 

[Noahaha]

I also wasn't refering to NISS when I said it is human intellegence.. I didn't say that NISS wasn't, but I didn't say it was. Now, I will say that yes it takes a small amount of intellegence to do most anything. If one were not intellegent enough, they could not even decipher notation.

I will go analyze some NISS solves and see how it works and get back to you about how much intellegence I think it uses. I've actually learned quite a bit from you guys explaining my misconceptions. Thanks.

Dude. Just what? I mean, seriously??? You are trying to figure out if the best FMC method requires INTELLIGENCE?

Unbelievable.

Spoiler

THE SCRAMBLE IS NOT USED IN THE SOLUTION.

EDIT: I'm going to explain how most people go about their FMC solves because it's clear you don't know how...

Step 1: Scramble the cube and try to use blockbuilding to solve as much of the cube as possible. Repeat until you find the most efficient solution that solves the most pieces.

Step 2: use insertions to solve the remaining pieces.

Step 3: repeat 1 and 2 on the inverse scramble

Where is intelligence NOT used???

Finding an efficient skeleton (step 1) requires Cleverness, skill and intelligence for sure.

Finding good cancellations for your insertions also requires a lot of skill.

There you go.

 

[elrog]

Step 1: Scramble the cube and try to use blockbuilding to solve as much of the cube as possible. Repeat until you find the most efficient solution that solves the most pieces.

Step 2: use insertions to solve the remaining pieces.

Step 3: repeat 1 and 2 on the inverse scramble

Have you not just described the normal way of solving a cube in less moves without the use of NISS? If all NISS is is solving the inverse of the scramble, it basically just a strategy to see if you can get a better situation to solve that will also work on the scrambled cube. If this is the case, the way everyone describes it is very misleading.

 

[Noahaha]

Have you not just described the normal way of solving a cube in less moves without the use of NISS? If all NISS is is solving the inverse of the scramble, it basically just a strategy to see if you can get a better situation to solve that will also work on the scrambled cube. If this is the case, the way everyone describes it is very misleading.

www.speedsolving.com/wiki/index.php/Fewest_Moves_techniques

 

[cuBerBruce]

Have you not just described the normal way of solving a cube in less moves without the use of NISS?

Right, in that Noah's steps 1, 2, and 3 made no mention whatever of using premoves, let alone NISS. NISS inherently makes use of the concept of premoves.

If all NISS is is solving the inverse of the scramble, ...

And as I said before, it's not.

Trying to solve the inverse scramble instead of the normal scramble is a general FMC technique that was well known long before the concept of NISS was became well known.

 

[okayama]

hey guys, have you ever seen an optimal insertion for corners with 9 moves? here it is:

L2 F2 U2 B2 U L2 D U2 B2 U' L2 B' R' U' L2 D L' F2 L B L
L' B' L' F2 L D' L2 U R B L2 U B2 U2 D' L2 U' B2 U2 F2 L2

inv:
F‘ L‘ B2 U‘ R‘ B2 D‘ B U2 - 2x2x3 [9/9]
F‘ D2 F - F2L-1 + Pair [3/12]
L‘ F L‘ F‘ L2 D - F2L [6/18]

I was happy to find a sub20 3cycle, what you don't find that often. But on the first try I wasn't able to find even a insertion without cancellation. So I tried to do it with two insertions. But even with a first insertion with 5 move cancellation the secound one was again with just one or two moves. So I used a MGLS to finish the solution in time:
R2 U2 R D2 R‘ U2 R D2 R - 3cycle (MGLS) [9/27]

IF can't find a solution. So I did no bad work. Can anyone try to find a better solution with two insertions? I can't believe that this is the best solution for this skeleton

inverse Solution:
F‘ L‘ B2 U‘ R‘ B2 D‘ B U2 F‘ D2 F L‘ F L‘ F‘ L2 D R2 U2 R D2 R‘ U2 R D2 R

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

Here is a better solution based on your nice skeleton.

All but 3 corners: D' L2 F L F' L F' D2 F * U2 B' D B2 R U B2 L F
A-perm: F (U' F D2 F' U F D2 F') F'

or Insert at *: F (L2 F R2 F' L2 F R2 F') F'

FYI: The same happened several times (e.g. here and here)
My previous skeleton (link) is quite rare, truely 9-move is needed for a single insertion.

EDIT: (sorry, I didn't try two insertions)

I encountered an unlucky skeleton that needs a 9-move insertion again.
Scramble: U R2 U B2 U B2 F2 U2 R2 U' L2 B L2 U F2 D2 R' F2 R2 F' (Weekly competition 2013-13)
Skeleton: D' U B * L U L' U' B' R' F R F U R' F2 R' B L' B' U D'
Insert at *: R' (F2 R' B2 R F2 R' B2 R) R

After that I found a better skeleton, but it was not enough to get the first place.
Congrats on your win, Sébastien and Attila. :tu

 

[Cubenovice]

I encountered an unlucky skeleton that needs a 9-move insertion again.
Scramble: U R2 U B2 U B2 F2 U2 R2 U' L2 B L2 U F2 D2 R' F2 R2 F' (Weekly competition 2013-13)

Very unfortunate...

Just browsed over the solutions, I was a bit surprised that only Jacob started with the 2-move 2x2x2 block.
I spotted it on the normal scramble and initially used it with premoves, then tried it on the inverse but nothing good came out of it.
Did anyone else try it but dismissed it because of poor continuation?

 

[mehdi-roshan]

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


Premove : R'

2X2X3 Block : L F L' D2 R2 F2 U2 L D2 // (9/9)
F2L minus 1 slot : U R F R' F' [#] //(5/14)
All but 3 corners : U F' U2 F R U' R' //(7/21)
Correction: R'

Insert at # = F2 U B U' F2 U B' U' //(8-3/26)

Final solution: L F L' D2 R2 F2 U2 L D2 U R F R' F U B U' F2 U B' F' U2 F R U' R2 (26 HTM)

 

[Sebastien]

My results from BW Open 2013: 26, 26

First attempt: F' U2 L' D L F' L' B D2 B2 R F2 L' B2 R' L2 D2 R' U2

Spoiler

On Inverse:
2x2x3: D' B R' D' L2 F2
switch and F2L-1: U L B' U B
switch and L3C: F U' F' U2 B * U' L' U' L U B'

* = B' R B L2 B' R' B L2

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

26 moves, optimal insertion.

Second attempt: R2 B' D2 F U2 L2 F2 R2 F' U R B2 U' B F' U R2 U' F

Spoiler

On inverse:
L7C: R' U2 R' B' U' : L * U' D' L . U2 L

. = L' D' L U' L' D L U
: = U R D' R' U' R D R'
* R D' L2 D R' D' L2 D

Solution: L' U L' D' L U L' D U L2 D R D' L2 D L' D' R' U R D R' B R U2 R

26 moves, optimal insertions.

This does actually lead to some pretty cool averages:

Current official average of 12: 30, 27, 31, 28, (32), (26), 26, 27, 27, 27, 26, 26 = 27,5

Hope that I'll be able to roll this a bit further.

My last 12 attempt overall:

- 23 (FMC Mustcube 441)
- 25/26 (German Forum Competition hier vom 18.3. bis 24.3.)
- 25/29 (German Forum Competition hier vom 25.3. bis 31.3.)
- 25 (FMC Mustcube 443)
- 25 (Speedsolving Weekly competition 2013-13)
- 25 (FMC Mustcube 444)
- 26/26 (German Forum Competition hier vom 1.4. bis 7.4.)
- 26/26 (BW Open 2013)

results in: (23), 25, 26, 25, (29), 25, 25, 25, 26, 26, 26, 26 = 25,5

Too bad I'm apparantly still too loosy even for keeping the NR for my nationality.

 

[HMS]

Finally started to get the hang of using NISS:

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

My solve: F R2 D' R2 B D L D R2 F B' R2 U F U' B U F M F2 U' M2 U B2 M B2 U F U2 F' B U' B'

F R2 D' R2 B D L D R2 F B' R2 B - xxCross (13/13)

Apply on Inverse, then continuation - B U B' F U2 F' (6/19)

Apply on original scramble, then xxCross, continuation -
B' U F U' B U F' (Niklas) (7/26)

F2 M F2 U' M2 U B2 M B2 U (ELL) (13/39)

Undo premoves - F U2 F' B U' B

3 Cancellations = 36 HTM, about half an hour

How's 36 for a first shot at NISS? Any advice? Also, is there a better way I could have written the solution out, i'm not exactly sure what the proper format is.

 

[Cubenovice]

Too bad I'm apparantly still too loosy even for keeping the NR for my nationality.

You obviously need more practice

But big props to Moritz as it was a very nice solution.

 

[Sebastien]

You obviously need more practice

Sure, I will practise more and eventually get sub20 average!

Spoiler

I really wish that FMC would work like that...

 

[Cubenovice]

In all honesty; I think that you and Tomoaki are living proof that practice pays off.

sub 30 av12 in official FMC!!!

I know how you do it but I still don't get it

 

[Sebastien]

Well, it pays off for consitency, but not for good singles. Now I got 27,5 official average of 12 but there is not even a sub26 single in there.

Ok, it is more likely to be lucky with insertions than being lucky with an LL skip, but I'm still waiting for one attempt with good cancelations after a great skeleton (just look at my second attempt this weekend), like Tomoaki got at his WR.

But I'll stop whining and provide some more cool FMC stats:

Considering this weekend's results (5 sub30s at BW Open 2013 and a 26 by Daniel in Ireland) there are now more than 200 official sub30s (202 to be precise) from 78 different people.

Most sub30s:

Spoiler

Sébastien Auroux 34
Tomoaki Okayama (岡山友昭) 14
Erik Akkersdijk 10
Milán Baticz 9
Clément Gallet 9
Daniel Sheppard 9
Jan Bentlage 7
Teemu Tiinanen 5
Javier Cabezuelo Sánchez 4
Grzegorz Łuczyna 4
Arnaud van Galen 4
Tim Reynolds 4
Mats Valk 4
Kanneti Sae Han (คันธ์เนตี แซ่ห่าน) 3
David Woner 3
Loïc Petit 3
Mario Laurent 2
Tobias Christlieb 2
Guillain Potron 2
Yu Nakajima (中島悠) 2
Henrik Buus Aagaard 2
Mirek Goljan 2
Simon Westlund 2
Michael Young 2
Stephen Adhisaputra 2
Alexander Olleta del Molino 2
Michał Pleskowicz 2
Guus Razoux Schultz 2
Linus Fresz 2
Ralph Eikelenberg 1
Evgeny Akivis 1
Sergey Ryabko 1
Stefan Pochmann 1
Robert Yau 1
Vidar Klungre 1
Dmitry Karyakin 1
Antoine Perdereau 1
Fang Qin (秦方) 1
Sebastiano Tronto 1
Hongfei Tian (田洪飞) 1
Shotaro Makisumi (牧角章太郎) 1
Mikhail Rostovikov 1
Joon Cha 1
Peter Hung 1
Ron van Bruchem 1
Anton Rostovikov 1
Olivier Polspoel 1
Thom Barlow 1
Auguste Olivry 1
Lorenzo Vigani Poli 1
Claudio Müller 1
Nicolas Fruy 1
István Kocza 1
Johannes Laire 1
Erik Jernqvist 1
Piotr Michał Padlewski 1
Shuang Chen (陈霜) 1
Marcel Peters 1
Olivér Perge 1
Jimmy Coll 1
Baiqiang Dong (董百强) 1
Olivier Stietel 1
Lars Vandenbergh 1
Sam Boyles 1
Abdelhak Kaddour 1
Jack Moseley 1
Yunqi Ouyang (欧阳韵奇) 1
Juan Juli Andika Chandra 1
Yumu Tabuchi (田渕雄夢) 1
Dennis Loose 1
Steven Xu 1
Zbigniew Zborowski 1
Dániel Varga 1
Moritz Karl 1
Andres Flügel 1
Philippe Virouleau 1
Jakob Kogler 1
Anthony Hsu 1

 

[okayama]

Sébastien,

Of course you know it, but let me daresay this here.
The WCA ranking does not necessarily reflect the FMC ability accurately, because that is based on just single records.
You got many Sub 30 results, 26.33 mean of 3 (see here), 27.5 average of 12, which are far superior to the others.
This means you are the best FMC solver in the world. Be confident, actually you are the world champion.
I wish I will be in a position to say ``that is not true anymore'' in the future, but it seems no easy matter.

I am not saying a single record is meaningless.
I believe that just luckiness does not lead a really short solution, but no little ingenuity is required for it.
Congrats Moritz, and please keep it up for further Sub 30s, which should reinforce your ER.

 

[Sebastien]

Tomoaki, thanks for your kind words!

Actually your reaction revealed to me, that my previous posts here may seem way more desperate as they were actually meant. Please reread them with a little smile.

As I already stated in the German forum, I'm happy with Moritz for his result and even if it was lucky in the end I consider his solution as way more worthy as the previous double-ER. It also motivates me to continue practising the event we like so much.

 

[Sebastien]

I got a 28 at Leicester Open this weekend. Like my 27 in Edinburgh a month ago, this was a fairly straightfoward solution again. When I say straightforward, I'm comparing it to some of the crazy complicated solutions I've done officially before.
27, 33, 33, 32, 29, 28, (26), 27, (DNF), 31, 27, 28 = 29.5 avg12

33, 33, 32, 29, 28, (26), 27, (DNF), 31, 27, 28, 26 = 29.4 avg12

I wish you lots of fun rolling this.

 

[Username]

I'm not good at FMC, but I thought of a way to do it somewhat good (i think) Could somebody tell me if this would work?

1. Do the scramble/Inverse scramble, don't do any pre moves
2. See if there are 3 corners that could be good to cycle to make an easier all but 3 corners.
3. Cycle the 3 corners as a pre move
4. Solve the cube like you would in whatever method you currently use, and look for insertions to cancel the pre move.
5....
6. Profit

Would this work?

 

[Sebastien]

Quite confusing...I don't really know how you want your step 2 to work as this requires information you cannot have at this point.

Even if what you describe would be applicable, just creating a skeleton leaving X corners and inserting them normally would be always at least as good as an applicable version of what you describe.