Blindsolving Discussion and Help

[RyanReese09]

Hey Ryan,

Before I explain how to do UBR -> LDF -> RUF you first have to understand how the more "standard" rotation of this case works.

Let's first do UBR -> DBR -> UFL

You'll see in a moment that this is the exact same case that you just asked me about, only at a different angle.

First I would see that the pieces form a right triangle. One leg of the triangle is formed by the line from UBR to DBR. The other leg is from UBR to ULF. This leaves the hypotenuse to be from ULF to DBR, the polar diagonal through the cube.

Once I see a triangle that has the hypotenuse as the polar diagonal I care only about two things:
1) Find the AnI corners (the "column")
2) Move the other corner so that it is still diagonally across from the column across a face

1) The AnI corners are UBR and DBR.
2) I have to move the other corner, ULF, such that it is still across from the column across a face. The move to do this is L. Now, instead of UFL being across from the URB piece across the diagonal on the U face, now the corner is at FLD and is across from the DBR corner (the "bottom" of the column) across the D face.

The rest of the commutator is a 9 mover.
UBR -> DBR -> FLD

Setup turn is U. This makes the cycle: UFR -> DBR -> FLD which is done via:
U L' U' R2 U L U' R2

Combine this with the setup move U, which will cause a 1 turn cancellation, and you get the 9 mover:
U2 L' U' R2 U L U' R2 U'

Now, combine this 9 mover with the initial L move setup to get the full 11 mover column alg which is:
L U2 L' U' R2 U L U' R2 U' L'

This alg is also very finger trick friendly for me.

-------

Ok, I guess I need to get more proficient with A9's if I wish to grasp columns. I understand everything you said in that post, but I can never tell if it's A9 (unless I rule out every other possible case, which I find is stupid that I have to resort to that), and then on top of that I can never find the correct alg (doing the 1 move setup messes with my visualization of my commutator that I have to make ). So I guess I'll learn A9's and then come back to these columns, because a setup into an A9 isn't hard to understand at all, and I know what I have to do for columns. Thanks Chris, as always.

Now onto your case: UBR -> LDF -> RFU

Find the triangle. Legs are from UBR to RFU and from RFU to LFD. Hypotenuse is from UBR to LFD. This is an awkward angle, so rotate the cube with x.

This makes the cycle: BRD -> LFU -> RBU
Or, if you prefer to viewpoint shift as I do rotate every sticker once clockwise to get:

DBR -> UFL -> UBR or UBR -> DBR -> UFL. Now it's the exact same cycle as above.

Trust me in that it sounds SO much harder than it is. Columns are so easy to recognize once you know to keep the column intact during the first setup turn. All you do is move the "other corner" either from the "top" of the column to the "bottom" or vice versa. In your example we had to move the other corner from the "top" of the column" to the "bottom" of the column with our setup turn L. As long as you're comfortable with that particular 9 mover case that follows, then the only real thought is either: A) The other corner is at the top of the column, ok move it to the bottom, or B) The other corner is at the bottom of the column, ok move it to the top.

So in my cycle I gave ,the Ani pieces are the URB and LUF, (aka the column) so I'd move the 3rd piece (the polar opposite to URB in this case) up with L'? This would make it opposite of URB? That's fine yes? L2 would work as a possible setup to move it across of LUF (though stupid to do it since L' is simpler setup?)

Time to grasp A9's..should have done this a while ago since they are common.

 

[amostay2004]

So it's 5 moves to insert it always..any general rules to know for the insertions that are different then the rules for 8 movers? I assume the 3rd corner that's being used for the interchange can't be effected. And is the 5 move insertion always something like RU style where only 2gen moves are used? Or should I find the 5 move insertion using any combination of moves to just get the insertion in the correct twist?

BYU didn't cover columns well at all in my opinion.

There're only 3 types of insertion, and two of them can be done in 3 moves. Only the diagonal insertion (eg UBR > DBR, BRU > RBD, RUB > BRD, you should know that 3 of them are the same thing) cannot be done with 3 moves, and they tend to exist in A9 and columns cases.

Remember LBL? Think of it as inserting a corner for the first layer when the white is facing top (assuming white for cross obviously). So yes, you could do [R U2 R' U' R U R', D2] and it solves a columns case. Anyway so far I know of 3 ways to insert diagonally that can be pretty fast: R U R2 U' R', R2 U R2 U' R2, and R U' R' U2 R U R'. You can use them to solve hard A9/columns cases, so use them wisely

 

[RyanReese09]

O__O that's extremely helpful amos.

What are the differences in them? Like, any reason to choose one over the other? Columns are like 8 movers, except with a 5 move insertion instead! This is so helpful. I love all of you no homo

The R U R2 U' R', I tried doign that, D2, and undoing everything but I got a scrambled cube after?

I am liking the R U' R' U2 R U R'.

My LBL method had first layer on top, using R' D' R D for corner insertion (I eventually did it intuitively but for white on bottom I did R' D' R D until I got an intuitive case.

I assume the other two insertions are the obvious R U R' (that twist) or F' U' F (I'd rotate obviously), that twist?

 

[amostay2004]

O__O that's extremely helpful amos.

What are the differences in them? Like, any reason to choose one over the other? Columns are like 8 movers, except with a 5 move insertion instead! This is so helpful. I love all of you no homo

The R U R2 U' R', I tried doign that, D2, and undoing everything but I got a scrambled cube after?

I am liking the R U' R' U2 R U R'.

My LBL method had first layer on top, using R' D' R D for corner insertion (I eventually did it intuitively but for white on bottom I did R' D' R D until I got an intuitive case.

I assume the other two insertions are the obvious R U R' (that twist) or F' U' F (I'd rotate obviously), that twist?

The R U R2 U' R' insertion is the alg I gave you earlier, it's to insert UBR to RFB, so the interchange should be F2, not D2. I don't know of any differences between them, just do what is fastest.

The other two insertions are R U R' and R U' R' (F U' F' is the same type as R U R')

 

[Egide]

Hey amos didn't you mean URB to RFU?

 

[RyanReese09]

The R U R2 U' R' insertion is the alg I gave you earlier, it's to insert UBR to RFB, so the interchange should be F2, not D2. I don't know of any differences between them, just do what is fastest.

The other two insertions are R U R' and R U' R' (F U' F' is the same type as R U R')

Extremely helpful. Hopefully I can use these insertions to solve most column cases (and some A9).

I do thank you a lot.

Edit-And yeah I assume you mean RFU.

 

[aronpm]

A 5 move insertion is not worth it for an A9.

 

[amostay2004]

A 5 move insertion is not worth it for an A9.

True, when I started off I used a lot of those 5 moves insertion, but now that I can do A9s fairly easily I'm gonna switch some of my algs to optimal. They're still very nice for columns cases though.

And yea I did mean RFU sorry

 

[cmhardw]

To be fair to Amos, if you use a 5 move insertion on a columns case it's only 12 turns. If you can do those 12 turns faster than the optimal 11, then why not?

 

[Sakarie]

To be fair to Amos, if you use a 5 move insertion on a columns case it's only 12 turns. If you can do those 12 turns faster than the optimal 11, then why not?

Who's been unfair to Amos? Aron only said that it wasn't worth doing on an A9, 12 moves instead of nine.

When looking for corner algorithms, I did a list:
UFL - r U L2 U' r'
URF - l' U' R2 U l
BUL- L U L2 U' L'
BRU - R' U' R2 U R
ULB - r' U' L2 U r
UBR - l U R2 U' l'
FLU - L' U' L2 U L
FUR - R U R2 U' R'

The sticker notated is where the piece will go, like "FUR" means that a piece will be inserted in URF-spot, without touching anything other on the F-layer. I know that half of these cases are just the other half with x/x' rotations, but I preferred writing it like this.

 

[cmhardw]

Who's been unfair to Amos? Aron only said that it wasn't worth doing on an A9, 12 moves instead of nine.

Good point, I mis-read that. I thought Aron meant for the columns case.

When looking for corner algorithms, I did a list:
UFL - r U L2 U' r'
URF - l' U' R2 U l
BUL- L U L2 U' L'
BRU - R' U' R2 U R
ULB - r' U' L2 U r
UBR - l U R2 U' l'
FLU - L' U' L2 U L
FUR - R U R2 U' R'

The sticker notated is where the piece will go, like "FUR" means that a piece will be inserted in URF-spot, without touching anything other on the F-layer. I know that half of these cases are just the other half with x/x' rotations, but I preferred writing it like this.

That's pretty cool, I'll have to take a good look at this when practicing later tonight. I've only done this move-blocks-around style insertion when doing a Per Special, I think these insertions could be great ways to speed optimize some of my slower cases!

 

[RyanReese09]

I thought when doing a column, you do 1 move setup, A9, then undo which is 11 mo ves. The alg given for column cases (which I can do all 18 of them! ) are 12 moves. What are you guys talking about when you say it's not worth doing 9 instead of 12?

 

[riffz]

They're saying that the R U R2 U' R' type of insertion is very useful for creating speed optimal (not move optimal) commutators for column cases, but generally creates a much too lengthy algorithm if you try to employ it for a case that could be solved as an A9.

Example: UBR -> RFU -> LBU

You *could* do [B', R' U' R2 U R] to solve this but generally there exists a faster solution such as this move-optimal one:

x U [R2, U L2 U'] U'

EDIT: I would like to point out that I did not introduce that 5 move insertion. The first time I saw it was when Joel posted it on this forum.

 

[riffz]

I use that insertion for almost every columns case, except the one where both of the targets are on D. Then I find a move-optimal setup to an A9 to be the fastest:

ULB -> DBL -> DFR

L2 U [L2, U R2 U'] U' L2 cancels to give L2 U L2 U R2 U' L2 U R2 U2 L2

 

[RyanReese09]

Gotcha, hopefully last question for a bit..

BUt for A9's I know it's an A9 if there is an interchange but no 3 move insertion? Always no 3 move insertion? In BYU BH thread, he says "but either there's a piece already in that layer, or there is no 3-move insertion (A) part."

Does that mean if all 3 pieces are in the same layer (as the interchange?) & there is no 3 move insertion, then it's an A9?

 

[riffz]

BUt for A9's I know it's an A9 if there is an interchange but no 3 move insertion?

Nope. Take UBR -> BUL -> RDF for example:

L [U2, L D' L'] L' cancels to L U2 L D' L'U2 L D' L2

As for your other questions, I can't answer them. I've never really known a set of rules for identifying different BH cases. I just kind of *know* by looking at them.

 

[RyanReese09]

Well what general steps of logic do you take to identify it?

And yay!
Can recognize and perform Columns, per specials, Cyclic shifts, Pure (for the most part unless I blank on recognition (in which case I can just rule out the other cases)/can't find the comm), and orthogonal.

Only those nasty A9's left .

 

[Sakarie]

I don't think that there's a nice and easy general rule on how to identify an A9 (or is there?). How I did was that I learnt how to identify all the other cases, since there are pretty strict rules on every case with more than 9 moves. But when doing enough solves or commutators, you'll hopefully learn how to see it.

 

[riffz]

I guess you could do it by elimination, since everything else is easier to identify.

 

[aronpm]

There are only 4 A9s, so just learn how to recognise them.