Thursday, August 24, 2017

Tangram Cube Initial Discoveries

I got a new puzzle yesterday from Speedcube Shop. I am checking it out.

Naming and counting the pieces.
  • Two little corners that only rotate.
  • Six big corners that can 3-cycle around the little corners. They can also swap around the long edges with an edge twist.
  • Three long edges that can flip one at a time, and move around the cube with the inner slice.
  • Three little edges that can move around with the slice, and swap one pair at a time with an edge twist. 
  • Six center pieces that double swap with each edge twist, and 3-cycle with each corner twist.
I experimented a bit with The Move but didn't find anything immediately so moved on. May come back to it though. I tried finding a typical commutator 3-cycle. I look for a set of 3 moves that changes only 1 piece in one of the layers. Didn't find one yet. But going UpR(ight)C(orner) ReplaceT(op)E(dge) DownRC ReplaceF(ront)E UpRC ReplaceTE DownRC ReplaceFE...

Wait a minute. I thought I had a 3-cycle of little edges dialed in, but now in trying to put them back I have scrambled the centers. Well, all but the red one. What did I do?

It went downhill from there. Pretty soon it was pretty scrambled. But now, without really knowing how exactly, I've got it back to the 5-cycle. hmmmm.

August 25, 2017
Before I went to bed last night I decided to name things so I can track what I am trying and how it works. I am using some of the letters from my BLD solving. Here is what we have so far.

Holding the cube with yellow up and blue front:

L = long edge at FrontLeft
A = long edge at UpBack
R = long edge at RightDown

ED = little edge at LeftUp
K1 = little edge at FrontDown
YP = little edge at BackRight

J = corner at FUR. J or J up, and J' or J down.
W = corner at BDL. W' or W up, and W or W down.

K = move slice up. K or K up, and K' or K down.

Right now I have a 5-cycle of centers and two little edges are flipped.

Things finally fell into place after spending a lot of time today trying to figure out some cycles and swaps and flips. It is solved. But I need to experiment and document much more before I want to give it a good scramble. Here is what I've learned so far.

L R L R 3-cycles E and its adjacent centers to K, K to Y, and Y to E.
Of course similar 3-cycles can be accomplished using L and A, and A and R.

J A J' A 3-cycles big corners like you would expect The Move to do, plus it 5-cycles centers. Blue to yellow to red to orange to green to blue.
Using this and similar Moves you can cycle big corners from
Blue to Yellow to Green
Blue to Red to White
Red to Yellow to Green
Red to Blue to Orange
and on and on. I suppose solving the 8 corners would be an easy thing to do. The two small corners, J and W can be solved in at most 1 twist. The six big corners can probably be solved easily using The Move.

Another way to 3-cycle big corners is with the following algorithm.
(A W' A J A W A J') one time.
It also double-swaps centers blue-yellow, and red-white.

(A W' A J A W A J') two times swaps the centers back so results in a pure 3-cycle of big corners.

(A W' A J A W A J') three times would cycle the corners back home and double-swap the centers only.

August 26, 2017
In the table below last night I only finished two rows then filled in the J half of the puzzle 3-cycles but not their algorithms. I did not yet try them to see what is double-swapped. I went to bed. I'm ready to get back to it now. I haven't solved the cube with eyes closed since Wednesday and it is now Saturday! Yikes! I want to get back to that too. But first I want to work on this table some more.

Maybe not. I just reviewed what I wrote above and you know, if I could come up with an easy way to solve centers, and then find ways to solve the rest without messing up centers, it would surely be a much better approach. So I am going to take a break from the table and do some more experimenting and maybe document the other things I have found so far.

L R L R 3-cycles E and its adjacent centers to K, K to Y, and Y to E.
Of course similar 3-cycles can be accomplished using L and A, and A and R.

J A J' A 3-cycles big corners like you would expect The Move to do, plus it 5-cycles centers. Blue to yellow to red to orange to green to blue.
Using this and similar Moves you can cycle big corners from
Blue to Yellow to Green
Blue to Red to White
Red to Yellow to Green
Red to Blue to Orange
and on and on. I suppose solving the 8 corners would be an easy thing to do. The two small corners, J and W can be solved in at most 1 twist. The six big corners can probably be solved easily using The Move.

Another way to 3-cycle big corners is with the following algorithm.
(A W' A J A W A J') one time.
It also double-swaps centers blue-yellow, and red-white.

(A W' A J A W A J') two times swaps the centers back so results in a pure 3-cycle of big corners.

(A W' A J A W A J') three times would cycle the corners back home and double-swap the centers only.

(K A K' L) two times just moves centers. O>B>R and Y>W>G
(K L K' R) two times just moves centers. O>B>G and Y>W>R
(K A K' R) two times moves centers and edges. Edges OY>BW>GR

R L K L R K' flips L and A, and E and Y.

(K A K' A) two times cycles K>E>P. K and E flip; P rolls.

(K A K' A) three times flips L and A.

(K A K A) K' (A K A K) or
(K A K A K) two times cycles E>K>Y. K and E flip; Y rolls.

(J A J' R) x 2 (A K A K') x 2 flips YP and K1.

A ((L A L) J (L A L A) J') x 3 Pure Center 3-cycle B>R>G and with setup moves we could easily do B>R>O and B>R>W and G>W>B and G>W>O and G>W>R and B>O>Y and B>O>R and B>O>W and O>R>G and O>R>Y and G>Y>O and G>Y>B and G>Y>R and on and on and on.

Now that I found a center 3-cycle and have convinced myself I have all the algorithms I need, I have fully and intentionally scrambled it. I will not need to complete the table below. In the next post I will try to put just the algorithms I plan to use and how.

Tangram CubeCorner 3-cycleDouble-swaps of centers
AlgorithmsBY RW BG WH BW RYGO
A W' A J A W A J'Yellow-blue-orangexx
A W A J A W' A J'Yellow-blue-whitexx
Yellow-red-orange
Yellow-red-white
Blue-yellow-green
Blue-yellow-white
Blue-red-green
Blue-red-white
Red-blue-orange
Red-blue-green
Red-yellow-orange
Red-yellow-green
and on and on

Tuesday, August 22, 2017

Forgot Last Sentence

At the end I forgot the last sentence and did H 2 L instead of 2 W L. Resulted in a 3-cycle. Oh, and then I used C instead of D. I said D but moved to C. Dumb.

F' R2 D' R2 F L D' R B' L' D2 B' R U B L F U2 B R' U2
PleaseLord JesusE...ment2 Guys
KidsBabes
HelpInspire
UsJesus
FromChrist'sPowerfulVictory
2WinLittleDirt bags

Used the Z Cube and went very slowly and deliberately.
Let's see if we can get it right this time.

Using the Rubik's Cube I got it. But it wasn't easy. When I got to the corners I had to stop and think hard to recall the sentence. What helped was remembering the corners, not the words.

Monday, August 21, 2017

No Sentences, No Notes, No Speed

My first attempt in a while to solve it without having taken any notes. I got 7 of the 12 edges and all of the corners right. I am trying to figure out how I did what I did.

The first two used M2. The next 4 F Perm. The last 6 used M2. Part of the mystery is solved. 2 of the 5 bad ones are flipped. I went to 2 when I should have gone to R. In trying to figure out the other errant 3 pieces I'm at a loss. I guess I'll just try it again and see if I can get it. Only be very careful about what I'm doing.

Well, I had to come up with a different string of letters and I just tried to remember the letters instead of making sentences. Somewhere along the way I think I made a mistake and I tried to fix it and continue. When it was all said and done the thing was a mess. Not totally scrambled. Needed an edge 3-cycle and a couple edges flipped. And the corners were pretty messed.

I guess I'll try again.

I finally got it. Here was my key to success. Go slow. Extremely slow. And hold the cube in such a way that there is no possibility of layers moving other than the one intended. And don't do the double layer moves. Edge F when using the F Perm is the only one I like to use it with. For what it is worth I did all solve attempts with the Zanchi. Oh, here is the scramble.
R D' R B' L' B' R' D' R' U' F' L' F D L B2 D L2 U2 L2 U D

And for what it is worth, I am no longer interested in trying to develop any speed cubing techniques. I like to move at a slow pace. Maybe if Josiah can master the white cross in 8 moves or less every time then I will point him in the direction of F2L videos. I don't know.

Sunday, August 20, 2017

Trying To Refine Setup Moves

I got an idea how I could do setup moves for the F Perm better when going to A. Move the L layer with the M layer. I did a lot of thinking and hesitating during the solve and possibly messed up the corner twist. Anyway my first attempt ended up badly. I decided to work through the whole thing with my eyes open carefully seeing how the new moves looked and felt. Then I did the whole thing eyes closed and got it.

U' L U F U' L2 B' U B2 D2 L' B L' D' L2 F R2 F D R
1 2 All Come
VeryOftenGodPraises
DizzyLittleIndiansJesus
Seriously
Twist 2&3
1Overcomes
4Eat
Patient1FightsImmorality

The eyes closed solve took about 9 minutes. I want to try again with a different cube and see if I can do it faster now. The first solve was with the Z Cube. Now let's see how I do speeding things up with the Rubik's Cube. 6 minutes. The Rubik's Cube is probably the slowest cube I have.

I didn't time this next solve, and I did not rush it. But it seemed to go fast. ECS2F solved. I've been using 1 as want or wan'. 1A sounds a lot like wanna. :D

F' U' L' D' B' L' F R' B U2 F U B' U2 L B R D' F U
KevinEats
LPRH4J
AI-x
BO
We-xVeg
1A-xFullPiece OfFreedom
twist 2&3
On the way to the Ernsts' this evening for a small group social, I made sentences for another scramble. When we got home I reviewed the notes I had jotted down on the way to town then solved it ECS2F. 

D' B2 R' D' L' U' F R2 D' L B' U2 R D2 F' U L' U L B'

Notes:
Every 1
G V I 2 S L-x P O
UP

H 2
A J P F-x
twist 1&3

Saturday, August 19, 2017

What A Mess!

I don't know where I went wrong but when I opened my eyes it was a mess! And red was in front. Maybe I started with red in front and that was why it was a mess. Nope. I checked. So I have no idea. And this was my second attempt. :(

R2 F D L' B L2 F' U F2 U' B' D' B' R2 D' R' F D' L' F' L U D
2 BE In
FredAteVic1
GreatYahweh
Saves
I Know
Jesus Es
WorthPraising

So I worked through it slowly and carefully and even opened my eyes throughout the solve to make sure all was well, and I got it. But then I tried again ECS2F and it was a mess again! I am going to try again! I want to solve this bugger!!!

Although it is 95 degrees outside I decided to go up the hill a ways to try it. Finally! I got it. ECS2F.

Friday, August 18, 2017

More Fun With My Eyes Closed

The first scramble I tried today with eyes closed ended in a bit of a mess on the first attempt. Then I worked through it bit by bit to make sure the analysis was correct and it was. So I tried again start to finish and got it. Used M2, D2, F Perm, edge flip, and corner 3-cycle. Whew.


Thursday, August 17, 2017

Yes! First Try Start to Finish Eyes Closed M2 D2 F Perm

Look at how much jumping back and forth there was on this one.


U' F' U' L' B' R2 U' F' D F' R F2 U' F2 L' D' F2 D' L D'
GenerousJesusM2
PaulUsesF Perm
ValuesForgivenessM2
CarlKnows xF Perm
EveryoneRejoicesBeforeElohimM2
HenryServesF Perm
IBelieveElohim is1CreatorGodD2

Just like I did at Raymond's when I was first solving using only the F Perm, this time I narrated out loud what I was doing. Setup, swap, put it back, set up, swap, put it back,... I think it helped keep me on track and focused.

Got another one in today this evening. I didn't do the out loud thing, but took it slow and easy.
R B2 R' B' U L' F R D L U2 F2 L2 F' D L2 D F' D R2

Henry Uses: F Perm
4 Yams: M2
1 Squash: F Perm
Bud Rejoices In Elohim: M2

Jesus Understands: D2
Alex Serves 4 Doughnuts: F Perm
1 Person: D2

Wednesday, August 16, 2017

Setup Moves For M2 and D2


M2 setupsD2 setups
BJ2Y > BFJPV > F
DL4V > DEIOU > O
EFGH > FABCD > D
OPRS > S1GL
CD2 can be used for anything except KS2 and HW4

What?!? The 3-Cycle :(


D' R2 D' B' R' D F2 L' D L' F' R D2 F D' R' D B2 D' B
Edges
YahwehGodCame2HeapLove
OscarAte x
setup Ucycle KUS
Corners
UDon't xOJesus
Ethan2Vic
Henry14

When I opened my eyes it was all good except for edges S and 3. They were flipped. So when I did the 3-cycle, S evidently was not oriented correctly.

I see it now that I have studied it but here is the thing. I don't think there are any setup moves that would make it work. Options?

1. Figure it out correctly and add a flip. Lots of extra thinking. :(
2. Instead of the 3-cycle use F Perm swaps. 1DSK


D' R2 D' B' R' D F2 L' D L' F' R D2 F D' R' D B2 D' B
Edges
YahwehGodCame2HeapLove
OscarAte x1DSK
Corners
UDon't xOJesus
Ethan2Vic
Henry14

Wow, that was a messup. Not totally scrambled, but... 4 edges and 4 corners were out of place. A couple double swaps. I need to try again on this scramble.

Tried. This time I lost focus or something. Anyway part way through I just lost track of where I was and quit. Must be break time. Let's go do some yard work.

Tried it one more time before retiring for the night. Got it. In spite of the fact that there were a couple times during the solve when I thought I'd blown it. But I kept my eyes closed and pressed on.

Should I practice more on using the setup move LiULU2LiUiL, and restoring it, or should I turn the cube and do it right-handed?

Tuesday, August 15, 2017

Starting Over—Kind Of

After convincing myself that I could use my method to solve a cube with my eyes closed consistently, I decided to take a look see at how others really do it. I checked out NoahCubes@Cubing World on YouTube. It was thorough. Very. His corner method was similar to mine only used less setup moves because he uses a different PLL permutation. His edges method is M2, and from what I gather it is a popular method these days. M2 has pros and cons. The pro is that the move M2 is the swap algorithm! The cons are the complicated cases and the end game if there is parity.

I was pretty excited about a one twist swapper so tried to use M2 when it was easy, and my F Perm when M2 wouldn't work. And the parity issue? I'll just deal with it like I always have. And it works! I modified his M2 method a bit so it works with BD as the buffer and UF as the target. That way I have the same buffer for both M2 and F Perm.

I got to thinking. Why can't you do the same thing with corners? Why can't you use D2 as a swapper to get as many corners as possible? And so with DRB as the buffer I use DLF as the target when using D2. I tried it last night and it worked!

There are some gotchas that came up along the way. As long as I do an even number of M2 or D2 before switching to the F Perm I'm ok. So sometimes I have to use the F Perm to make the number of preceding M2s or D2s even.

Why complicate matters by using three different swappers instead of just one? Because two of them are so quick and easy!!!!!

After working through a scramble seeing what it would look like if I started with Corners, then what it would look like if I started with Edges, then trying to come up with a memorization system that would help me remember when to do M2 or D2 and when to use the F Perm, oh my, what a pain. I decided to take a break and go do some yardwork. While doing yard work it all became clear. When using the F Perm algo I use Food sentences. When using either M2 or D2 I use Jesus talk. Here is how it worked out.

L U' R2 U B' U' L D' F' L D B U L D' L B' U B U R B' D2 B D'
JoeKnows
YahwehGodLovesOthersChristVictorious
EthanUsesRice
HenrySteams
GodIsPatientAndJesusForgives

The first attempt I messed up and I didn't have a solve at the end. The second attempt I kept opening my eyes to make sure all was well. It was. Maybe I'll try again with another cube and see if I can get it start to finish with closed eyes.

Here is a breakdown of how many moves it takes me to solve each piece. Each word stands for a piece of the cube.

The ones that use the F Perm algorithm
Joe 23
Knows 25
Ethan 21
Uses 23
Rice 27
Henry 23
Steams 29

The ones that use M2 or D2
M2 for edges
Yahweh 7
God 7
Loves 7
Others 7
Christ 1
Victorious 7

D2 for corners
God 15
Is 9
Patient 7
And 17
Jesus 9
Forgives 7

So altogether I saved over 200 twists by using M2 for edges and D2 for corners whenever possible.

Sunday, August 13, 2017

Popular BLD vs My BLD

Last night one thing led to another as they often do when researching online and I decided to really see how other people solve the cube without looking. NoahCubes on the Cubing World channel has a 5 video series on the subject. I'm on video 3. He is showing the Old Pochmann method. My method is based on the basic idea Pochmann used, at least partly, so the basic method is the same, but the implementation is way different. When I got started I knew only one algorithm that swaps a pair of corners and a pair of edges. So that is the algorithm I went with. I found a way to make it work for both corners and edges. Today I looked it up to see if my swapping algorithm has one of those Perm names. Yep. It is an F Perm algorithm. www.speedsolving.com/wiki/index.php/PLL#F_Permutation is the page I found it on.

Too bad I didn't learn more of those PLL algorithms. The one NoahCubes shows has a definite advantage over mine. Setup moves for corners take either 1 or 2 moves. That is because the target is isolated from the other 3 pieces in the double swap. R, F, and D can all be used freely without disturbing the 3. My F Perm swapper is no longer than the algorithm he uses, but he has the advantage in setup move count. On the other hand I'm not so comfortable using D and I don't have to use it at all in my setups. And three series of moves that drive my setup move count up are R U Ri, R Ui Ri, and R U2 Ri. Those are pretty quick and easy to do. I also use L in my setup moves sometimes. Counting each of the above 3 sequences as 1 move instead of 3 moves here's how my setup moves break down: 3 take only 1 move; 10 take only 2 moves; 5 take 3 moves; 2 take 4 moves. I guess it isn't as bad as I thought.

I just finished video 3 of 5 in the series. He told about M2 for solving the edges. At first it seemed so simple and quick. Wow! But then it got crazy requiring some complicated thinking and knowledge of Y Perm or something like that. He has talked a lot about T Perm and Y Perm and J Perm. What are those? J Perm. To swap A and D corners and A and D edges use a J Perm algo. But in BLD you modify it slightly so you are swapping corners A and S. T Perm swaps corners A and D, and edges B and D. See the T? Y Perm swaps A and C corners, and A and D edges. There you have it. I do not plan to ever learn any of these algorithms.

He also used the nasty P word—parity. My method may not be as fast for speed solvers, but it is way more simple and straightforward. You pretty much just plug along using the same one algorithm for both corners and edges. The only thing is if you do an odd number of swaps in the first part you have to start with the front piece instead of the back piece when planning out the solve because it will be in the back when you get to that point in the solve.

I wonder though if I could use a combination of M2 and my method? Use M2 until it gets crazy then jump to my method. Is that possible? Rather than try it out immediately I went on to video 4, in which Noah talks about his memorization system. He uses letter pairs. Visual letter pairs and audio letter pairs. He goes into great detail. I might try to incorporate some of this. I think the visual images could be confusing to me because how do I know the order? So I might use audio for the whole solve instead of one for corners and the other for edges. He does not make sentences like I have been doing. I have already started using letter pairs and even 3 letters to form a single word, even before watching his video. With mine I have made things up on the fly though. He has a consistent set of words for each letter pair. In video 5 he gives tips on how to get sub 2. It builds on already being comfortable with the technique he taught in the previous videos. I don't plan to use it, so I will not be watching 5 at this time. I would like to see what others have to say about BLD though. I watched a couple other guys. Evidently there are a lot of memorization techniques. Mine works for me. I've been gravitating toward joining letters into a single word anyway.

Now I guess the only thing is to see if I could incorporate M2 at all.

Friday, August 11, 2017

Lidbeck Fun

I met Brandy and Josiah yesterday. They came to school to meet me and see my puzzle collection. One of the puzzles totally baffled me. The Witeden Super 3x3x3. It has circles on the white and yellow sides. Those sides will not turn if a complete circle is not present. I tried solving it today. Got close but not quite there. So I looked up my old notes. I've solved it several times now but it is still an adventure every time.

Josiah is almost 7 years old and has been cubing only a few weeks, but can solve the cube faster than I can. He wants to get faster so I am going to try to help him. While I am at it I will try to learn some speed solving techniques too, starting with the White Cross.

The puzzle in my collection that Josiah was most interested in is the Bump cube that I have stickered so that it can be solved by colors too. And the way I did it when you solve it by colors there is a silver cross on the orange side.

Brandy and I have been emailing about cube stuff a lot. She pointed me in the direction of SpeedCubeShop.com. It looks good. Next time I want to shop around I just might check there first.

Thursday, August 10, 2017

The Results Are In

I discussed the 4 cubes I have recently and wondered if there is a better one out there that is quiet, smooth, and has easy color recognition in dim light. I even posed the question to my puzzling Facebook friends. I haven't gotten any feedback from them yet but as far as my 4 cubes go here is what I think this morning.


  1. The Zanchi 5 Stickerless is the quietest, smoothest, colorifest cube.
  2. The Rubik's Tiled is noisy and clunky, but the colors are good and for some reason in spite of the noise and clunkiness I like to use it.
  3. The Dreampark is smooth but has that sometimes annoying spring sound. The colors are good in bright light but not in dim light.
  4. The Duncan needs more trials before I decide if I prefer it over the Dreampark. I like the look and feel of the Dreampark better, but Duncan may have an edge in the sound category.
I don't know if I could do better than the Zanchi 5 Stickerless. It is for sure my favorite at this point.