COLL

[Dennis]

So I was inspired by Jason Baums ZBF2L post and his COLL recognition...
in this [thread=1176]thread[/thread] I asked him about some COLL and he gave me an example of the recognition formula given on all COLL pages I have seen, e.g

L R
F F

the main reason i moved from COLL to OLL was recognition, I never really understood the above given formula. And frankly, I did'nt really udnerstood Jason's explanaition either...

So please, explain this to me as I was a 3 year old child. Give me a few examples.

And while were at this topic, please provide me your COLL links

 

[Lt-UnReaL]

The COLL alg you do will orient the top layer + switch 2 corners that need to be switched. I'm not really sure what the code means either...When you move onto PLL you'll have 1 of 4 algs that are all edge cycles, because all the corners will be correctly permuted, and you'll have a 1/12 chance of a PLL skip.

Lars Vandenbergh's site has COLL: http://www.cubezone.be/coll.html

 

[Dennis]

I know what COLL does, what I don't know is how to interpret the recognition letters given at the four upper corners, i.e understand how these letters will permute the corners.

Here are the COLL links I know of:
http://www.cosine-systems.com/cubestation/vh/vhlltechniques.php
http://www.cubewhiz.com/coll.html
http://www.its.caltech.edu/~leyanlo/coll.html
http://www.speedcubing.com/final_layer_corners_no_edge_flips.html
http://www.cubezone.be/coll.html
http://www.geocities.com/chrisbcubing/COLL.html
http://lar5.com/cube/yMain.html
http://www.cubeloop.com/php/cube/intermediate.php?chapter=coll (my own, which I don't use... Old stuff based on Petrus)

 

[Dirk BerGuRK]

Hey Dennis

I really had trouble recognizing the cases for COLL also (except I was recognizing CMLL, basically the same thing). So I browsed and came across this: http://www.speedcubing.com/corners_first_corners.html

This is not a list of COLL algs, but it taught me how to recognize which corners to switch using the system you described. It's a little brief, but after I looked at it for a while it made sense and I began to instantly recognize each case. If you need a further description let me know and I will do my best to explain it.

 

[csfield]

The one you gave means the front corners have the same color stickers showing and the other two are the colors that aren't opposite that color. For recognition, you just think of the matching ones. If you want to see what the case for an algorithm looks like, just do the inverse of the alg on a solved cube. I wouldn't recommend learning any COLL if you aren't already fast with OLL (all of them) and fast (sub 2) with your edge-only PLL's. The only real advantage of it is for the PLL skips (it's really fun to recognize a COLL quickly and get a PLL skip). I think the slightly slower algs combined with the slight recognition delay pretty much evens out with the gain from the faster PLL's. Then if you master COLL, you still need to learn some kind of edge control to take full advantage.

 

[csfield]

I think I understand why you are having trouble. When you look at it, you don't need to figure out what the permutation is. You just recognize the OLL and the pattern of the stickers. As an example, I'll go through the cases for the "double-sune" since there are only 4 and they are easy to recognize.

The four stickers will have one or two pairs of matching corners. If there are two, I hold the cube so the are in the front and back (the front corners match each other and the back corners match each other). If the "H" is sideways, I do the regular double-sune: (RUR'U)(RU'R'U)(RU2'R'). Otherwise I do F(RUR'U')(RUR'U')(RUR'U')F'. If there is only one matching pair, I hold the cube so the "H" is sideways (like normal double-sune) and the matching corners are on the right or in front. Then I do (RUR'URU)(L'UR'U'L) or F(RU'R'U)(RU2'R'U')(RUR'U')F', respectively. I don't need to know where the pieces are going, I'm just looking for the OLL and the pattern of the colors of the corners.

 

[deadalnix]

Messing up today, I found

fRUR'U'RUR2U'R'U'RU'f'

For the Pi chessboard patern.

Is there a list of COLL cases for Pi that do not rely on sune+nicklas combinations ? These are good, but do not chain well with the last step.

 

[obelisk477]

1) ridiculous bump.
2) should be f R U R' U' R U R2 U' R' U R U' f'. You switched the U's at the end.