Older cubers discussions

[newtonbase]

Interesting. For U2, Ub buffer requires y & y2 rotation for sides. For Ul buffer, y & y' rotation for the side. Hm, which one will be efficient?
I would say Ul since you can always use "U" instead of y for D layer swap alg but not sure.

I wasn't really thinking about efficiency when I chose it. I just picked it because it's at the back next to my x centre buffer Ubl. I'll have a play with the other one. It's not too late to change.

 

[mark49152]

I don't see how Ul versus Ub would make a difference to rotations. The only reason to rotate during U2 is to avoid having to do S slices during inserts. So you only need two orientations 90 degrees apart - for example, I switch between F and R faces facing me. I don't know why you would ever need a y2 rotation.

 

[mitja]

Same here. Corners - Wings - Centers, execution in the opposite order.

@mitja / regarding fingers, yes, better practicing tracing/memo without using anything but audio or images.

Thanks, I will try this order. Now, I still have to decide about UBr xcentre buffer that I use now. I use UBL for 3BLD, so don't know what is better for my xcentres on 4x4. I like r2 for wings for my fingertricks, but not sure about centres. The alg for top centre is on r slice, thats nice and for target setups I use also y and y', so the slice alternates from l to r. I guess I will stay with UBr it is not that different.
Do you guys always use these:
l slice:
BUl - (B' U R' B U') r2 (U B' R U' B)
UFl - l' (B' U R' B U') r2 (U B' R U' B) l
FDl - l2 (B' U R' B U') r2 (U B' R U' B) l2
DBl - l (B' U R' B U') r2 (U B' R U' B) l'
r slice:
UBr - r2
FUr - F d (R U R') d' (R U' R') F' r2
BDr - r2 F (R U R') d (R U' R') d' F'
I succesfuly manage to avoid them in 2/3 of pair combinations.( not UBr of course)
But maybe would be better just to learn them as fast as possible and not waste time thinking through those cases?
My way is to convert l slice case to r slice case( by using F2, B2 or U2 setup). Then use those 3BLD Mark's advance M2 tricks :
FU-s = M2 s M s’ M (where s/s’ are the setup/undo for the s sticker)
s-FU = M’ s M’ s’ M2
BD-s = M2 s M’ s’ M’
s-BD = M s M s’ M2

 

[mark49152]

@mitja: You can also solve the l slice targets by setting up to UBr. For example, to solve BUl, rather than setting up to FUr with U2, I would set up to UBr with l U2.

Setting up to FUr or BDr might be nicer in some cases though. I might start playing with that.

For adjacent l slice targets I assume you know the comm [r U2 r', D2] and inverse?

 

[openseas]

I don't see how Ul versus Ub would make a difference to rotations. The only reason to rotate during U2 is to avoid having to do S slices during inserts. So you only need two orientations 90 degrees apart - for example, I switch between F and R faces facing me. I don't know why you would ever need a y2 rotation.

Maybe I was not doing efficiently - you can give me a comment. I find it easier to do m slice rather than s/s', that's why I rotate (y or y') for all "u" target of T-center. Same goes for "d" target, rotate "y" then do m slice. Then again, it makes more sense to use Ub since I can use m slice without rotation, lol...

My way is to convert l slice case to r slice case( by using F2, B2 or U2 setup). Then use those 3BLD Mark's advance M2 tricks :
FU-s = M2 s M s’ M (where s/s’ are the setup/undo for the s sticker)
s-FU = M’ s M’ s’ M2
BD-s = M2 s M’ s’ M’
s-BD = M s M s’ M2

Yes, you can use the same advanced M2 techniques for FUr & BDr cases.

@mitja: You can also solve the l slice targets by setting up to UBr. For example, to solve BUl, rather than setting up to FUr with U2, I would set up to UBr with l U2.

Oh, this is nice! I should try this!

 

[mark49152]

Maybe I was not doing efficiently - you can give me a comment. I find it easier to do m slice rather than s/s', that's why I rotate (y or y') for all "u" target of T-center. Same goes for "d" target, rotate "y" then do m slice. Then again, it makes more sense to use Ub since I can use m slice without rotation, lol...

Yes I also rotate but I always have one of the same two faces at the front. That way there is never more than a single y or y' needed. Insert at both front and back. You mentioned y2 and I'm not sure why you would ever do that since all you would achieve is turning the M slice around to insert at front instead of back, or something like that, which seems pointless.

 

[JohnnyReggae]

Try this out for a 4x4 scramble

F B2 R2 Uw' R2 Rw' B2 L2 Rw' Fw' D B F' Rw D Fw L R F B2 Uw' L Fw2 R' Fw2 B' R F2 Fw' D2 U2 Fw2 L' F' U L Uw2 L' R' F'

I got a PB with it of 52.28 and should have been quicker .... but it almost feels like cheating

 

[openseas]

Yes I also rotate but I always have one of the same two faces at the front. That way there is never more than a single y or y' needed. Insert at both front and back. You mentioned y2 and I'm not sure why you would ever do that since all you would achieve is turning the M slice around to insert at front instead of back, or something like that, which seems pointless.

Yes, I realized that after reading your comment - I need to switch my buffer from Ul to Ub now

 

[mitja]

@mitja: You can also solve the l slice targets by setting up to UBr. For example, to solve BUl, rather than setting up to FUr with U2, I would set up to UBr with l U2.

Setting up to FUr or BDr might be nicer in some cases though. I might start playing with that.

For adjacent l slice targets I assume you know the comm [r U2 r', D2] and inverse?

Thanks this looks great.
What about FDl? l2U2 maybe?

 

[mark49152]

Yes, I realized that after reading your comment - I need to switch my buffer from Ul to Ub now

That was my original point. If you are rotating, those buffers are equivalent, because in both cases you just switch between two adjacent front faces - one using the m slice to insert from u layer (front insert) or d layer (back insert) and the other using the r slice to insert from e layer (front or back). So really there's no advantage to be gained from switching if you've invested time in Ub.

Thanks this looks great.
What about FDl? l2U2 maybe?

Set up to UBr with l' U2.

 

[openseas]

@mitja: You can also solve the l slice targets by setting up to UBr. For example, to solve BUl, rather than setting up to FUr with U2, I would set up to UBr with l U2.

Set up to UBr with l' U2.

editing my comments: tried to do but realized it changes U layer (U2) followed by l or l'. So, it seems to screw up T & X center of l layer. Probably I'm doing something wrong. Can you give an example?

 

[mark49152]

These case, can we use only for even numbered target? (for r2) When it is odd target, r slice in U layer has opposite color - so U2 will change l and r slice color in U layer, am I doing something wrong?

You can use on odd or even, but I am talking about setups for advanced r2 not for solving the piece individually.

Let's say your next two pieces to solve are FLu and FDl. You can solve FLu on its own with [U' L' U: r2] but then you would have to solve FDl individually which is quite a long solution.

Instead you can set it up to UBr and solve both together. So [l' U2: [U' L' U, r2]]. Note that the inner bracket is now a commutator not a conjugate; the second r2 solves FDl before you undo the setup.

Expanded, that would look like this: (l' U2) (U' L' U) r2 (U' L U) r2 (U2 l)

 

[mitja]

You can use on odd or even, but I am talking about setups for advanced r2 not for solving the piece individually.

Let's say your next two pieces to solve are FLu and FDl. You can solve FLu on its own with [U' L' U: r2] but then you would have to solve FDl individually which is quite a long solution.

Instead you can set it up to UBr and solve both together. So [l' U2: [U' L' U, r2]]. Note that the inner bracket is now a commutator not a conjugate; the second r2 solves FDl before you undo the setup.

Expanded, that would look like this: (l' U2) (U' L' U) r2 (U' L U) r2 (U2 l)

Thats exactly what I use and I learned it on 3Bld first. At the beginning it was confusing, but eventually it became automatic and it is more or less 3-cycle edges(wings). For me it is also nice because it is in pairs and I don't need to think about flipped M or r layer.

 

[SpartanSailor]

Quick question guys... I'm new to BLD solving in general, having only started as part of a 2018 goal. I have figured out the OP/OP method of 3BLD solving and understand conceptually the M2/OP process, but haven't decided to pursue learning the few differences for M2 yet.

I'm still interested to follow along with some of the Big BLD discussion.... but what are the "wings"? I assume those are the edges on the big cubes and solving the "wings" is I assume the same as solving the edges? Just trying to keep up with some of the discussion.

 

[newtonbase]

Quick question guys... I'm new to BLD solving in general, having only started as part of a 2018 goal. I have figured out the OP/OP method of 3BLD solving and understand conceptually the M2/OP process, but haven't decided to pursue learning the few differences for M2 yet.

I'm still interested to follow along with some of the Big BLD discussion.... but what are the "wings"? I assume those are the edges on the big cubes and solving the "wings" is I assume the same as solving the edges? Just trying to keep up with some of the discussion.

Yes. Wings are the edges on a 4x4. On a 5x5 they are the edges either side of the middle edge (midge).

 

[mark49152]

@SpartanSailor : Yes, wings are the individual edge pieces. Two wings together make an edge. The word "edge" is more commonly used to refer to all the pieces directly between two corners and corresponds to the 3x3 edge after reduction. Note that unlike 3x3 edges, wings cannot be flipped.

You will often see the following terminology:

4x4 - two wings matched together form a "dedge"
5x5 - two wings and a "midge" together form a "tredge"
6x6 - two inner wings and two outer wings together form an edge, and similar for bigger cubes. I have only rarely seen these called a "quedge".
...etc.

 

[mitja]

Yes two wings of 4x4 is one edge of 3x3. Welcome to Bld. I also started a year ago, but I don't do it regularly, but would love to find more time for it. What is nice about M2 is that is very intuitive. At first I only learned the important M slice Algs( and only two are longer and also one reverse to another), the rest can be done intuitively. I did read all of them and checked if I can do them by myself. Thats all you need. First ,check some youtube tutorials so you visually understand the system of setup and interchange.

 

[mitja]

I see we all answered at the same time

 

[mitja]

editing my comments: tried to do but realized it changes U layer (U2) followed by l or l'. So, it seems to screw up T & X center of l layer. Probably I'm doing something wrong. Can you give an example?

Why don't you try it on 3x3 first. It is much easier to follow the concept. But the concept is a 3-cycle. And what is nice, i would call it intuitive 3-cycle. I browsed through many 3-cycle algs for edges and compared them with this advanced M2 cycles. And many are simmilar, many are also shorter and optimized, but eventuality not that much longer to be worth memorizing as a must. I prefere this intuitive aproach.

 

[mark49152]

I browsed through many 3-cycle algs for edges and compared them with this advanced M2 cycles. And many are simmilar, many are also shorter and optimized, but eventuality not that much longer to be worth memorizing as a must. I prefere this intuitive aproach.

That is what is great about M2. It is simple enough to learn as a beginner method, and once you start to exploit advanced M2 to the full, what you have is effectively 3-style, albeit constrained to M slice interchanges. That's very efficient. It's also a great platform for incrementally adding new comms that interchange on other faces/slices wherever that gives significant optimisation; but in many cases the M slice comms are simply not worth replacing.