Freestyle Blindfold corner algs

Michael_Wee

Regular
Joined
Jul 18, 2008
Messages
60
I watched Rowe Hessler's video on freestyle corner blindfold algs and realised that solving 3 corners at a time is going to be faster that classic pochmann corners and thought that if some had all the corner algs could they post it and if they were willing to find it and post that will be helpful

thanks in advance
 
Most of the time commutators solve *2* pieces at a time. It's 3 only when your cycle ends. Chris Hardwick has all 3-cycles from URB buffer, I can't find the list though.

But really, you don't need algs for them. They're intuitive freestyle commutators, just learn how to use commutators.
 
I have Chris' list. I asked for it in a PM :p I'll get the link and post it here later.

Edit: Here it is http://dbeyer.110mb.com/BHcorners.txt :)

Wow that is a lot of algs! I wonder if its more than ZBLL :eek:

Anyway the only problem I have with freestyle corners is misoriented corners, breaking into new cycles, memo, and parity, which holds me back from using freestyle for corners xD
 
does anyone know where is the best place to learn more about commutators i really suck at it now
 
is there any website that can help me because i really suck at commutators and lack basic understanding of it
 
There is a How to Board here on this forum. Look there for some explanations of commutators.

Be looking for our site that should be up soon (I hope). We will go a little bit more in depth so that you can see that the optimal commutator system is just as simple as F2L (long term goals).

Later,
DB
 
does anyone know where is the best place to learn more about commutators i really suck at it now
I suggest doing it at home, at a quiet, clear desk. Great place to try out commutators and understand them.
Also a good place: School, when you can get away with it in class.
ahh... yes, very wise! Home is the best place to learn about commutators. I prefer to either use FMC companion when at my computer or a real cube when laying in bed.
 
Does anybody really understand commutator tricks?
Or are you guys looking at each case and seeing how to set it up to do a commutator, which you all know as an 8 mover?

For example every optimal corner commutator in optimal length is 8-12 moves.
Do you understand how to determine the move count? Do you understand the tricks?

8 movers are rather simple. ABA'B'
A being the insertion (3 moves)
B being the interchanging
A+B+A'+B+ = 8

9 movers have a setup turn. The setup turn actually cancels with either the part A or part B of the commutator.

Such as RB'RF2R'BRF2R2 = R; [B', RF2R'] = R B' RF2F' B RF2R' R'
= R B' RF2R' B RF2R2 <-- Canceling the setup move by merging the two R's making it 9 moves.

There are cases that can only be done in 10 moves. You might look at me, if I showed you some cases, and say oh well I'd do this. It's the same move count, or why not do this. These are called orthogonal cases. These cases are optimally solved by just doing a quarter turn setup, and then solving with an 8 mover. Chris and I call these orthogonal cases because as you look at the cycle you will see three mutually opposite pieces. The pieces as a whole can be interchanged only with double turns.

Such as URB -> ULF -> DRF
Now that specific corner cycle is actually a 12 move count. We are looking just at the pieces, not the sticker permutations for now.
See how the URB and ULF are interchangeable by U2.
The ULF and DRF are interchangeable by F2
The URB and DRF are interchangeable by R2

Now the distinction between the 12 move case, and these 10-mover "orthogonal cases" is the sticker permutation and their positions relative to one another.

There are 6 planes on a cube. The U, D, F, B, L, and R plane.
Each plane has a parallel plane. Yet in free space in order to be perpendicular you need three reference points. To be perpendicular (right angled) in free space you need a sticker from each of the sets of parallel planes. So a target sticker on the U/D plane, a target sticker on the L/R plane, and a target sticker on the F/B plane.

Such as <u>U</u>RB -> <i>L</i>FU -> <b>F</b>DR

There is another case for 10 movers that is pretty sick too. It's called a cyclic shift.
These cases might be recognized as COLL cases if you were to look at the cycles on the U layer.
You have three pieces on one layer. Lets start basic with a U layer cyclic shift.

URB -> RUF -> BLU
You see all three are on the U layer. Notice the middle piece (URB) isn't interchangeable with either of the adjacent pieces. We call this AnI. Adjacent, non-Interchangeable.

Now watch. Both of the middle piece's AnIs on the U layer can actually become interchangeable with setup turns.
L/L' make the BLU interchangeable with U2/B2
F/F' make the RUF interchangeable with R2/U2

Notice the common interchangeability of U2?
U2 is insertion in this case. LF' and F'L are the setups.
The URB is inserted to the RUF
So we need to insert there first
LF' sets the RUF to the ULF, so do U2, undo the setup.
F'L sets the BLU to the ULF, so do U2, undo the setup.

LF' U2 FL' F'L U2 L'F

Now it's called a cyclic shift, because you'll notice if you start at a different point in that alg it becomes
U2 FL'F'L U2 L'FLF' which is just a commutator with a 4 move insertion. Odd how changing the starting point has such a different effect eh?!

There are also columns cases which are 11 moves, there are two ways to handle these. Doing a cyclic shift that has a canceling effect just like in a 9-mover. Or you can do a setup and an 9 mover.

Per Specials are the 12 move cases. There are only 6 of these in a buffer bound system. I showed you an example of one above. All all mutually opposite and on the parallel planes.
Later,
DB
 
I'm getting to grips with commutators and this 3-cycling malarky but have a few trouble cases. Anybody got any advice for when the corner stickers you want to cycle are two on one face, the third on the opposite face. The specific case that has eluded my humble efforts is UFL -> DRB -> UBR. Equally perplexing is UFL -> UBR -> DRB. Both seem pretty resistant to using set-up moves to get an A perm or a simpler cycle (UFL -> RDF -> URF for example)

Does anybody really understand commutator tricks?

Such as URB -> ULF -> DRF
Now that specific corner cycle is actually a 12 move count. We are looking just at the pieces, not the sticker permutations for now.
See how the URB and ULF are interchangeable by U2.
The ULF and DRF are interchangeable by F2
The URB and DRF are interchangeable by R2

Now the distinction between the 12 move case, and these 10-mover "orthogonal cases" is the sticker permutation and their positions relative to one another.

There are 6 planes on a cube. The U, D, F, B, L, and R plane.
Each plane has a parallel plane. Yet in free space in order to be perpendicular you need three reference points. To be perpendicular (right angled) in free space you need a sticker from each of the sets of parallel planes. So a target sticker on the U/D plane, a target sticker on the L/R plane, and a target sticker on the F/B plane.

Such as <u>U</u>RB -> <i>L</i>FU -> <b>F</b>DR

The case you describe here is similar to the one I am having trouble with, but I don't really understand, after reading your post, what I should do to find the 12 moves for the case.
 
Last edited:
URB -> ULF -> DRF

Do:
U F2 U' F2 U' R2 U F2 U F2 U' R2

It takes a bit of getting used to, but it is actually a simple ABA'B' commutator.

A = U F2 U' F2 U'
B = R2

Take a look at the A part, it inserts the ULF corner into DFR without disturing anything on the R layer. Then interchange with R2, then undo the A part, then undo the interchange move.

This commutator is one of the more beautiful ones on the cube in my opinion.

Chris
 
I learned comms from Ryan Heise's site, but I had the intention of using them for FMC. It took about 30 minutes for me to learn. I will admit that I know practically nothing of freestyle, so I don't know well Ryan's explanation will transfer over to freestyle.
 
I learned comms from Ryan Heise's site, but I had the intention of using them for FMC. It took about 30 minutes for me to learn. I will admit that I know practically nothing of freestyle, so I don't know well Ryan's explanation will transfer over to freestyle.

Thanks for pointing me to Ryan's site. What's not to know about freestyle? You just do what you like to get the cube solved. The more you know about cycling pieces, the less you end up messing around with akward set-up moves. Some people use a buffer to ease breaking a monster 6 or 7 cycle into some easy threes. It seems there's a fair amount of confusion as to what 'freestyle' actually means to a BLD solver. The irony is that it's exactly what it says it is.
 
Okay I have a few questions

How do you go about making upyour own commutators?
How many moves are they usually?
How would you know while during the memo when your going to have a parity.
And what is the parity or parities for freestyle corners?
 
How do you go about making upyour own commutators?
If you're asking this you're not getting it. Some people find defining a 'lone corner' and an 'interchangeable pair of corners' makes sense, but basically:
1. Put a corner from one layer into the opposite layer, without affecting any other pieces in that layer. eg. L D L' puts the FLD corner into FUL spot. No other U layer pieces are affected.
2. Now any U move will change that corner for another. eg U2. These moves form A and B of the commutator. For my example, A = L D L', B = U2. Now put together ABA'B' for a 3 cycle.
That's science. Well, group theory. Which is maths. Meh. Turns out *most* cycles can be done this way. Some need more nifty A algorithms which insert a corner into a layer.

How many moves are they usually?
Corners: 8-12 moves. Try to get setup moves cancelling the last move of B'.

How would you know while during the memo when your going to have a parity.
I have no idea. I've never memorised a 3x3x3. I can't blindsolve. Something to do with an odd number of pieces in the cycle. Ask a pro.

And what is the parity or parities for freestyle corners?
Hmm. Again showing your hand by asking an odd question... BLD parity is when the edges and corners cannot be solved independently as there is a 2-edge, 2-corner swap (kinda like a T perm for example) somewhere. This means the usual BLD trick of treating corners and edges separately for memo and solving can't work. Your memo falls short and you have to remember to perform some sort of edegey-cornery swap at some point. I recommend some set-up moves and a LL perm like T. You'll pop it if you're anything like me.

Again, I'm a nobody who has only ever solved a 2x2x2 blind, only done it once, and it was lucky. (2 ready placed pieces or something) So act upon my advice with care.
 
Back
Top