Interest in a BH method website?

[Mike Hughey]

Any information for name convention for 8 move algs ?

I honestly don't see any reason why you need a naming convention for them. I honestly didn't bother to learn the names. I know the general shapes and what they look like, and can move them around on the cube. Since I can see them really quickly, I don't see any need for actual names for them. I think the names are mainly good for explaining them to people who can't see them.

If you can already seen them, I figure you've already grown past the need to be able to name them. Again, unless you're teaching them to others, in which case the naming could be useful.

I remember I found those names somewhat helpful back when I was learning center commutators. (Chris has been using most of those names for years.) The names helped me see what was going on, so I could understand the algorithms. Then I promptly forgot the names, once I could just see the algorithms.

 

[deadalnix]

Yes, but as you explain, it's easier to explain with names. And I don't understand the name (even if it's not a problem to find the good alg).

 

[deadalnix]

Hi chris, here ar some improvements for BH corners page :

L2 D R2 D' L2 D R2 D' : B'RBL2B'R'BL2
U2 F' D2 F U2 F' D2 F : RDR'U2RD'R'U2
U2 L D2 L' U2 L D2 L' : B'D'BU2B'DBU2
D F2 D' B2 D F2 D' B2 : B2R'F'RB2R'FR
L' F2 L B2 L' F2 L B2 : B2UFU'B2UF'U'
U2 R D2 R' U2 R D2 R' : F'D'FU2F'DFU2
D2 L' U2 L D2 L' U2 L : D2BUB'D2BU'B'
U2 B' D2 B U2 B' D2 B : LDL'U2LD'L'U2
F U2 F' D2 F U2 F' D2 : D2R'U'RD2R'UR
F' R2 F L2 F' R2 F L2 : L2URU'L2UR'U'
R2 B L2 B' R2 B L2 B' : D'L'DR2D'LDR2
D R2 D' L2 D R2 D' L2 : L2B'R'BL2BRB'
R2 U' L2 U R2 U' L2 U : FLF'R2FL'F'R2
R2 D' L2 D R2 D'L2 D : BLB'R2B'L'BR2
U' L2 U R2 U' L2 U R2 : FLF'R2FL'F'R2
R' F2 R B2 R' F2 R B2 : B2DFD'B2DF'D'
U F2 U' B2 U F2 U' B2 : B2L'F'LB2L'FL

I hope you'll use them (And i hope I have made no mistake)

 

[Mike Hughey]

Hi chris, here ar some improvements for BH corners page :

I agree with these. In fact, these are the ones I learned. (At least, the first few are - I'm too lazy to check them all.) I guess that shows everyone I wasn't completely honest. If I was really sure that the moves were optimal, I didn't bother checking your list sometimes to see if I was right. Apparently I didn't do that for any of these either.

Chris, I think it's worth it to go for QTM optimal (as second priority, after HTM optimal). If you're going for optimal anyway, you might as well go all the way. And these are certainly every bit as easy to see as the ones you have, once you get used to them.

 

[MatsBergsten]

I must say I have no time for learning BH-corners now (mainly because I am rebuilding
part of our house and have no spare time at all).

But I cannot resist, I have to check whenever I have a scramble and need to use one
of my worse algs. And more often than not it is an alg I immediately want to learn.
Commutators are so nice

Maybe I end up with only commutators and then become center invariant too so I can
solve the corners first on 4bld & 5bld

Thanks for your effort, B & H!

 

[dbeyer]

I am now working on the Wings portion of BH. I am actually listing the most effective QTM solutions.

R'F2R b' R'F2R b
vs
b U'FU b' U'F'U

Both are quite fast, as you can use a quick cube rotation and a Wide-Turn to make the solution into a U/R layers alg (that is to include U, u, d, L, l, r, R)

(3L')U2R d' R'U2R d x'
vs
y r U'LU r' U'L'(3D)

This is just a simple 8 move example. I will give none (or at least very very few) solutions on wings that involve cube rotations.

Later,
DB

 

[flee135]

I have a few things I need to get straight about corner commutators. If anything I state is wrong, please correct me or add on to what I say if I miss something. These all deal with recognition. I need more information about this because I am still a little shaky on recognizing which cases are which. I am only truly comfortable with executing pure commutators at the moment, so maybe when I learn some more later, it will come to me.

1) For pure commutators, you will see two interchangeable pieces, with a third piece on a different plane that can be swapped with one of the other pieces.
2) I have trouble recognizing A9 commutators. How do you know what your set-up move is and how do you know it will cancel with another move prior to execution?
3) Orthogonal commutators involve cycling three stickers on three different planes, none of which are parallel, such U, F, and R.
4) Cyclic shifts have three pieces on the same plane, but the piece in the middle is not interchangeable with either of the other two pieces. So how about these other two pieces? If they are interchangeable, then does that always make the middle piece interchangeable with something?
5) Columns are when there are two pieces interchangeable, but the third piece cannot be swapped with any of the two.
6) Per specials are when all the pieces can be interchanged with double turns. Or something to that effect. I notice that there are two stickers on the same side, and the other one is opposite and diagonal.

Please correct any of these if they are wrong, or help me out with any questions. Thanks for any help provided!

 

[cmhardw]

Daniel and I are doing something quite different for edges than for corners, and I will probably end up doing the same for corners eventually. For edges we are posting multiple algs for many cases, sometimes as many as 3 different algs for a case.

I won't overwrite algs with more QTM optimal ones, but rather include both so that people can choose. I don't always use the most QTM optimal alg in every situation, because I do to some extent prefer to use the easier to execute alg. I would gladly use an alg with an interchanging M2 rather than an alg with all quarter turns. I have learned from the speed of M2 that interchange moves by double slice turns tend to be faster in some cases.

Any algs that people post as alternatives I will include on the site as alternatives, rather than replacing.

I am still working on the edges, formatting about 50 algs per day. It is turning out to be much more work than I expected. I am hoping to finish within the next few days. Meanwhile Daniel is typing up the algs for bigger cubes, and when he is done with the first set I will format and post while he types up the following set, etc.

Stay tuned for more updates! To be honest I want to get the edge algs posted first, then I will go through and add any suggested alternatives that people provide. Remember that the goal of BH is to construct the correct algorithm for a given case from the case name. If the correct alg for you is different than the alg we list, then use your alg. Not everyone solves F2L the same way, so I see no reason at all that people should use the BH method the same way. To be perfectly honest there are a handful of corner cases where I use a B9 alg rather than the optimal 8 move alg because the B9 is much faster. Eventually we may add these alternatives on the site, but for now my priority is getting the edge algs posted.

Thanks all for the comments, any suggested algs will be added to the site after the edges are posted, so keep them coming!

Chris

 

[Mike Hughey]

To be perfectly honest there are a handful of corner cases where I use a B9 alg rather than the optimal 8 move alg because the B9 is much faster.

I'm really tempted for the case (URB ULF DBR) to use the algorithm I've used for years for 3OP (I learned from Macky's site): (R2 D R2 D' R2 U2) * 2. Since I hate columns anyway, and it's only one move more (but admittedly lots of QTM), and I've had so much practice with the longer one, I figure I may never get faster with the columns algorithm, no matter how much I practice it.

I think offering alternatives makes a lot of sense. But I still think it would be nice if the primary algorithms listed were maximally efficient. But don't get me wrong - I think your priorities are well-placed. I would just like to see this happen when you run out of other good things to do.

 

[cmhardw]

I'm really tempted for the case (URB ULF DBR) to use the algorithm I've used for years for 3OP (I learned from Macky's site): (R2 D R2 D' R2 U2) * 2. Since I hate columns anyway, and it's only one move more (but admittedly lots of QTM), and I've had so much practice with the longer one, I figure I may never get faster with the columns algorithm, no matter how much I practice it.

Mike, I used to feel the same way. Try to stick with it though, with the columns algs. I personally prefer the setup into the A9 case (URB ULF DBR) I would do as L U R2 U L' U' R2 U L U2 L'. I admit that it took me a long time to get used to the columns algs, but just try that alg as an alg that you were trying to learn to finger trick. I can guarantee you that if you stick with it you will fall in love with the columns algs. They are difficult to see at first, I agree 100%, but they are so much faster than any other alternative once you are used to them.

If it helps, I used to use Columns algs when I could see them easily, then use my old algorithms during competitions or solves where I felt I was going really fast and might get a good time. Now I use columns algs exclusively only because I stuck with it and am so used to them. I can't think of any columns case that I would do with my alg algorithms, but then again I've been practicing them for a long time. Try to transition yourself into the columns algs slowly, but I do recommend to use your old algs when you see those cases much easier. Just try to slowly wean yourself onto the columns algs. They are so super nice.

I think offering alternatives makes a lot of sense. But I still think it would be nice if the primary algorithms listed were maximally efficient. But don't get me wrong - I think your priorities are well-placed. I would just like to see this happen when you run out of other good things to do.

I agree with that actually. I will probably mark the pages to show that the optimal QTM alg is Optimal with an asterisk or something, but provide other alternative algs with the same HTM. I do agree that would be a good idea.

Chris

 

[rahulkadukar]

Please put the Edges up fast so that I can give up TuRBO

 

[Mike Hughey]

Mike, I used to feel the same way. Try to stick with it though, with the columns algs. I personally prefer the setup into the A9 case (URB ULF DBR) I would do as L U R2 U L' U' R2 U L U2 L'. I admit that it took me a long time to get used to the columns algs, but just try that alg as an alg that you were trying to learn to finger trick. I can guarantee you that if you stick with it you will fall in love with the columns algs. They are difficult to see at first, I agree 100%, but they are so much faster than any other alternative once you are used to them.

If it helps, I used to use Columns algs when I could see them easily, then use my old algorithms during competitions or solves where I felt I was going really fast and might get a good time. Now I use columns algs exclusively only because I stuck with it and am so used to them. I can't think of any columns case that I would do with my alg algorithms, but then again I've been practicing them for a long time. Try to transition yourself into the columns algs slowly, but I do recommend to use your old algs when you see those cases much easier. Just try to slowly wean yourself onto the columns algs. They are so super nice.

I have already been following your advice. I've stuck with the columns algorithms for a while, and I admit I'm starting to see them. It still typically takes me 10-15 seconds to do one, but I'm doing it. It's cost me some really fast times on a couple of solves, I think. But I can see where it could be fast. I'll hang in there with it for a while longer.

 

[dbeyer]

The great thing about the High-move count algs is that there are only so many of them.
8 turns: 198 cases
9 turns: 126 cases
10 turns: 30 cases
11 turns: 18 cases
12 turns: 6 cases

54 cases use 10+ moves.
12 Orthogonals
18 Cyclic Shifts
18 Columns
6 Per Specials

The system is really awesome. The similarities in patterns are so simple. All the cyclic shifts are quite nice. I like them. The key is recognizing the patterns quickly, execution will come with time.

I'm getting back into it myself actually. Everybody enjoy.
Later,
DB

 

[cmhardw]

Update: As of 10:51pm EST I have formatted 220 of the edges algs, so I'm exactly half way through. I'm going to try to keep working tonight to do as many more as I can. Still looking at 2-3 days from now I'm thinking to finish all of them.

--edit--
Going to bed. 160 cases left to go. Might take 2 more days and the page will be done.

Chris

 

[deadalnix]

Chris, you are my heroes

 

[cmhardw]

OK everyone, Enjoy! :-D

BH Edges

--edit--
I'm leaving the case counts off the page temporarily as I am trying to verify and double check my numbers. I'll add them back once I'm done.

Chris

 

[Mike Hughey]

Wow, Chris - looks like you've been working hard! Thanks for this!

I'm sure I won't get to learning edges anytime real soon, but I must admit I'm becoming a believer. Today I got 4 successful 3x3x3 BLD solves under 1:50, and I had a very close DNF (I forgot to flip 2 edges) that was 1:27. If those two edges hadn't been flipped in the original scramble, I could have had my first sub-1:30 solve. Clearly I'm capable of much faster solves with BH corners than I was able to do with 3OP. And I suspect that as bad as I am at M2, and as slow as I am in general, learning optimal edges will probably help me greatly on edges, too.

With BH corners, if there are no twisted-in-place corners and there is no parity, it feels almost like cheating. I must have done the corners on that 1:27 solve in under 20 seconds, memorization + execution. It's just ridiculous. But unfortunately, it can still be pretty bad for me if I have both twisted corners and parity. It didn't seem like the disparity was as bad between good and bad solves with 3OP.

 

[Pedro]

no 2-gen algs for the U perms? I can do them much faster than the STM optimal...

anyway, great work, Chris

I'm not sure yet if I'll ever try this...maybe when (if) I finish my letter-pairs list

 

[cmhardw]

no 2-gen algs for the U perms? I can do them much faster than the STM optimal...

anyway, great work, Chris

I'm not sure yet if I'll ever try this...maybe when (if) I finish my letter-pairs list

Hey Pedro, Mike, glad you like the site. Daniel really did all the work to write down the algs. All I did was to format them into a webpage and occasionally add my preferred alg to a case as well as Daniel's when we use different ones.

Keep in mind that learning the edges is just like corners. Treat it like learning to recognize F2L cases, and you see a lot of the same kinds of things for edges that you see for corners. Only some of the cases (ex. being the B9's and the Slice-Planes) are completely different.

Some of the A9's break down into different subcases than what you see for corners, but otherwise it's all the same idea.

Now Daniel and I are working on a page explaining the BH method, and how to recognize the patterns. Think of the pages with the algs simply as a lookup reference if you want to check the alg you are using for a particular case.

You can't write a math book without a forward on set theory, and similarly we can't post the method without a reference of the algs for all cases. Those sites are only the reference sites though, the method pages are soon to come.

Chris

 

[Rubixcubematt]

yay, well done chris! I liked your explanation on 4x4 edges in your 4x4 BLD tutorial, which is very helpful explanation on BH corners. awesome job on posting it up!