• 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 40,000+ people from around the world today!

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

New nmCmLL Layout.

dbeyer

Member
Joined
Sep 12, 2006
Messages
396
Location
Harrington, Delaware
I am sure that you are all aware of this website.
http://www.athefre.110mb.com/NMCMLL.html

I have spoken with James for a few weeks about the concepts on this page. Once I began to fully understand the system, I noticed some formatting issues. I found that the table of algorithms was rather redundant and reptitive. I actually came up with an idea to reconstruct the organization of the system.

Revision 1. The algorithm table will only have the 42 algorithms and the solved case.
Revision 2. The L/R configurations will be grouped and reorganized.
Revision 3. The code beneath the configurations will give the AUF that positions the cube to apply the algorthm.

The table of algorithms will no longer have an AUF in parentheses. This is not required, because the new table of configurations can allow a consolidation of the table.

If you look at the table of configurations, you will notice that the last case on the left column goes with the first two rows of cases on the right column. Some slight reorganization of the cases can promote a better learning process.

I will be releasing my version of the table very shortly. Any tips or feedback is quite welcomed and appreciated.
 
Joined
Oct 1, 2010
Messages
361
I'm very interessted how you solve this, especially Revision2.

P.S. Any suggestion to improve AUF-Recognition for G3-1 would be apreciated.
 

dbeyer

Member
Joined
Sep 12, 2006
Messages
396
Location
Harrington, Delaware
A1 Solved B1 RU2R'U'RU'R' C1 RUR'URU2R'
A2 LU'R'UL'U2RU'R'U2R B2 R'ULUF'U2R'UL'U'R2 C2 R'U2RU2RB'R'B
A6 FRU'R'U'RUR'F'RUR'U'R'FRF' B3 B'RBR'U2R'U2R C3 FR'F'RU2RU2R'
B4 RU2R'U2R'FRF' C4 RU’L’U’BU2RU’LUR2
D1 RU2R'U'RUR'U'RUR'U'RU'R' B5 L'URU'LUR' C5 RU'L'UR'U'L
D2 RBL'B2RB'LBR'BR' B6 RU2L'UR'U'LU'RU'R' C6 RUR'UL'URU'LU2R'
D3 R'U'RU2L'UR'U'LU'R
D4 rU'r'U'F'UF E1 R'U'F2RUL'ULUR'U2R F1 RU2R'U'RU'R2U2RUR'UR
D5 RURU'L'UR'U' E2 RU'L'UR'U2B'UBL F2 LU'RUR2F2rFr'
D6 L'U2RU'r'U2R2B'R' E3 R'UL'URU'LU2R'UR F3 FRF'LFR'F'L'
E4 FR2U'L'UR2U'LUF' F4 FU'L'UR2U'LUR2F'
G1 FRUR'U'RUR'U'F' E5 RB'RB'L'BR'B2LB'R' F5 RUR'U'R'FRF'
G2 R'F2RF'U2RU'r'U'F E6 FRUR'U'F' F6 R'FRF'RU2R'U'F'U'F
G3 R'FRUFU'RUR'U'F'
G4 F'UrUR'U2Fl'U2l H1 RU2R'U'RUR'U'RU'R' H3/5 RUR'URUL'UR'U'L
G5 R'FR'F'R2U2B'RBR' H2/4 FURU'R'UF'U'R'F'R H6 FRUR'U'RUR'U'RUR'U'F'
G6 R'U'RU'R'UF'UFR
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
This is a simpler way of organizing the cases. I probably wouldn't have gotten around to remaking the text and sequence tables if I had decided this should be done. It's nice that you have.

I spent a lot of my time discovering and developing this recognition. It's fun and is useful in FMC for those that aren't able to intuitively recognize non-matching corners. But, I've grown away from searching for good reasons to use NMCLL in Roux speedsolves. 83 cases, difficult EO recognition in 50% of solves, and .75 extra moves during Step 4. All of that to save one move on average during the first block and, if skilled enough, maybe save one more on average during the second block. I think this works a little better in methods that have edges oriented before the step, even with the problems those methods have.
 

dbeyer

Member
Joined
Sep 12, 2006
Messages
396
Location
Harrington, Delaware
Hey, guys. It has been a while since I have been on here.
I developed a very interesting recognition system. I used to think of this recognition system as a huge task (in spite of developing BH).
I realized that the detemination of the CmLL case is actually based on the rules of the cube. Such that by defining two stickers on a corner, this indicates the third.

I realized that the orientation and permutation of the L/R stickers actually limited the possible permutation of the U-layer stickers.

Now, I solve the cube with one of four 1x2x3 blocks on the L face. Then I solve the R's 1x2x3 block from one of the other two remaining blocks. Essentially I have taken the Opposite Cross technique and applied it to roux.

White and Yellow are the L/R. The D color for the first block can be red or orange. Likewise, the D color for the second block can be red or orange. The two D colors do NOT have to match.

I solve these four blocks so fluently, that I do not even realize that I solved non matching blocks or recall which color was my L block.

I started off by organizing the L/R orientations into groups.
I then color coded them.
Then I also color coded the corresponding CmLL algorithms based on the L/R orientations associated with that case.

L/R Oriented: Yellow
L/R Sune: Orange
L/R Anti-Sune: Orange
L/R Double Sune: Dark Blue
L/R Triple Sune: Green
L/R U Orientation: Light Blue
L/R Bruno: Dark Red
L/R T Orientation: Pink

Color in the spreadsheet of algorithms with this color coding. You notice that there are some patterns. Patterns are good. Patterns make learning and memorization easier. For this reason, you realize that the method is powerful, in that the same recognition system works for both Non-matching and matching blocks.
 
Last edited by a moderator:

Kirjava

Colourful
Joined
Mar 26, 2006
Messages
6,121
WCA
2006BARL01
YouTube
Visit Channel
lol "method"

Seeing and remembering these patterns for better recognition is something that comes naturally after simply using any viable NMCMLL system.
 

dbeyer

Member
Joined
Sep 12, 2006
Messages
396
Location
Harrington, Delaware
To continue, you will notice pairs of permutations grouping together. Such as E3 and E5. H1 and H6, or F2 and F4.

It is nice because Pink (T orientation) only has 2 CmLL from B, C, E, F, G, and H.
That means per Orientation of the U layer, you can immediately narrow it down.
 

mariano.aquino

Premium Member
Joined
Dec 18, 2009
Messages
104
Location
Buenos Aires, Argentina, Argentina
WCA
2011AQUI01
YouTube
Visit Channel
I started trying opposite D-color blocks and I realized just that same concept, although I'm not sure applied the same way. I use regular recon system, vertical, horizontal or diagonal patterns (or parallels or cross) of same colores stickers, and I have been able to use almost same recon for matchting block and NM ones as well. There are a few cases where I had to invert pattern (cross becomes parallel and viceversa) but it works great so far!
Is this what you're talking about, dbeyer? cause you sound like you're using an 80 pattern's chart, while i still use the same 40's from coll
 
Top