Cmll

[Rawn]

I'm having trouble recognizing these cases quickly. If anyone knows CMLL or similar sets of algorithms, how do you recognize the cases??

 

[Kenneth]

The order of the colours of the stickers.

Take the standard pi-case, the Bruno R' U2 R2 U R2 U R2 U2 R'. For pi-cases you look only in the U-face and this one got opposite colours for the two corners pointing in the same direction and one colour on the other two corners. H-cases are also U-face only while the rest of the cases includes stickers on the sides as well.

 

[waffle=ijm]

It takes a while to recognize. Just practice. I don't use CMLL for Roux. I just use COLL algs (40) I know from trying ZB

First I look at the orientation case which then slims it down to 7 out of 40 algorithms (4 out of 40 for H-case). This should be instant. Since there are only 7 orientation cases

Then I look for same colors or opposite colors. Determine in which order they are in. 1 out of 7 algorithms.

http://www.cubewhiz.com/coll.html shows what to look for. You can just apply CMLL algs instead of COLL algs

 

[pjk]

As far as recognition goes, you recognize the same as COLL, and Jason explains it pretty simply here:
http://jmbaum.110mb.com/coll.htm

 

[Athefre]

As far as recognition goes, you recognize the same as COLL, and Jason explains it pretty simply here:
http://jmbaum.110mb.com/coll.htm

It's not as simple as that. I think association is the quickest technique for recognizing NMCMLL.

Steps:

1. When solving the first block note the bottom color of it (and also remember whether yellow or white is the color of that block), the U corners that belong on that side with the block will have the opposite of that color on top.
2. Do the same for the second block
3. As soon as you see the corners you will be looking for the orientation pattern of the yellow and white stickers, as you are doing this, find the location of the colors you noted for each block.

Example:

SCRAMBLE (taken from the WCA site)(scramble with orange on top and blue in front): B2 F2 L2 R2 B' F2 L2 R2 F2 U2 L' B L B R' D2 R B' F2 L2 R D U B2 F2 L' R F2 D2 R

Step 1: B U M R B R' r F (note that the bottom of this block is red so the the U corners for the left side will be orange on top and also note that the white U corners will be the only corners that will also have an orange sticker on them)
Step 2: r2 U M2 U' r U' M U' R U2 R' (note that the bottom of this block is orange so the U corners for the right side will be red on top, also note that the yellow U corners will be the only corners that also have a red sticker on them)
Step 3: Look at the table (http://grrroux.free.fr/method/Step_3j.html), the case for this is F2-2. You first look at the orientation of the yellow and white stickers and as you do that you find where the color orange is located on both white corners and you find where the color red is located on both yellow corners.

So do (U2)R'U2RUR'U'RU2lU'R'U and you are finished.

EDIT: Nevermind, lol, I thought I read the topic as "NMCMLL".

 

[Rawn]

Actually I've already thought of using these ideas. I was just that on the Roux website, under the algorithm table it says that he doesn't recomend using quick colour patterns to recognize cases. So I was just asking around.

 

[Kirjava]

You guys probably wanna take a look at this;

http://home.comcast.net/~quadricode/hyperorientations/

 

[Athefre]

You guys probably wanna take a look at this;

http://home.comcast.net/~quadricode/hyperorientations/

I can't seem to completely understand his recongnition system.

1. Determine the orientation of the corners with respect to the U-color (which I will call the first orientation; rows A-H in the second table that follows).

2. Look at the FUL-sticker’s color. This is purple in the diagrams, and the opposite of this color is pink.

a. For case B, look at the UFL-sticker instead.

3. Match your configuration, or second orientation, with the configuration in the appropriate row.

4. Do the algorithm provided, or use your favorite one.

I think where I get lost is between Step 2 and Step 3. I'm reading it as this:

1. Look at the orientation of the stickers that are supposed to be brought to U. In my example above, 2 are located on R-layer, 1 on F, and 1 on B.
2. Look at the UFL corner and note the color that is actually on F. Red in my example.
3. Find your case.
4. Perform the sequence.

I'm not seeing how this is much different from the normal way of recongnizing NMCMLL. You are still looking at one orientation then another. The only difference I'm seeing is that you look at the orientation of the colors that are supposed to be on U then you look at the orientation of the colors that are supposed to be on F and B. It seems like L/R orientation isn't used in his system.

Actually I've already thought of using these ideas. I was just that on the Roux website, under the algorithm table it says that he doesn't recomend using quick colour patterns to recognize cases. So I was just asking around.

I hope you aren't considering using the NMCMLL techniques I described for your CMLL cases.

 

[DonQuixote]

Well, NMCMLL really helps with fast recognition but even after one year of regular CMLL training I had some problems learning all the NMCMLL cases because they cannot really be learned and stored in a linear way. (In the B-column of the CMLL the sequences are B1, B2, B3 ... B6, but in the B1 column of the NMCMLL the sequences are C1, C5, D6, F5, so you can't do some stand alone practice)
Additionally, if you don't solve NM-blocks NMCMLL is total overkill, so I wouldn't recommend to learn it if one intends to stay with matching blocks.

Kind Regards
Martin

 

[rubiksfriend]

What would you recommend learning if you plan to graduate from matching to non-matching blocks?

 

[Athefre]

What would you recommend learning if you plan to graduate from matching to non-matching blocks?

If you see a great benefit to using non-matching blocks then you can do orientation then permutation using the NMCMLL table to help you completely understand how it works.

When I was using non-matching blocks I went ahead and started memorizing the table instead.

 

[Rawn]

That recognition system seems hard to understand, but I understand some parts of it at least.