All layer twists and cube rotations should work now as described in the notation documents:

○○○○○○● R
○○○○○●● TR
○○○○●●● T3R
○○○●●●● T4R
○○●●●●● T5R
○●●●●●● T6R
●●●●●●● CR

○○○○○●○ NR or N6L'
○○○○●○○ N3R or N5L'
○○○●○○○ N4R or N4L' or better MR
○○●○○○○ N5R or N3L'
○●○○○○○ N6R or NL'

○○○○●●○ N2-3R or better VR
○○○●●●○ N2-4R or better V3R
○○●●●●○ N2-5R or better V4R
○●●●●●○ N2-6R or V5R or M5R or better WR

○○○●●○○ N3-4R or better M2R
○○●●●○○ N3-5R or better M3R

●○○○○○● SR
●●○○○●● S2R
●●●○●●● S3R

... and so on.



Actually, there is a thread about 6x6 Notation and other large cubes you may want to check out.


BTW, since version v6.2.56 the shape of 7x7 is spherical like the real V-Cube 7.

Love it!!!

Hmm .... For the 7x7x7 cube applet i actually prefer the normal shape. Can this be made an option via some parameter?? I would love that

By the way, great to see most or all issues with the first version(s) fixed so fast already

- Per

I definitely like the pillowed applet. How large are the spheres?

Anyhow, I have the system set up, I just have to make the definition files for the cubes.

But along the task, I couldn't resist creating cube poetry, which uses a half-nonsense scheme to make every character a move.
Challenge to Per, qqwref, Werner, and anyone who dares: Make a poem that produces a nice pretty pattern.

I've seen that thread, but that is simply a suggested notation. SilverPotato is by no means in charge of higher order cube notation. As you can see within both that thread and this one (and I'm sure any others that use 666 or 777 notations), there is still no real standard and most people use different ones.

Alright, I finished the definitions for a well-accepted system, which covers most natural moves and only has arbitrary choices whenever necessary, which several of us on #rubik have spent time deciding. As far as I know, it will be used for the official V-Cube solutions.

I'm calling it "Simple General Notation," which is very descriptive and has a nice acronym.
http://cube.garron.us/applets/SiGN_test.htm

Moves are:
○○○...○○● R
○○○...○●○ 2R
○○...○●○○ 3R
...similar for {n}R
(Uppercase = single slice)

○○○...○○●● r
○○...○○●●● 3r
○○...○●●●● 4r
...similar for {n}r
(Lowercase=outer block turn)

●●●●●●● x
(Standard cube rotations, even if the direction of M is just stupid. I'm nearly serious about considering introducing N as an alias for M'.)

○○○●○○○ M'
○○●●○○ 2m'
○○●●●○○ 3m'
○●●...●●○ m'
...similar for {n}r
(Consistent with uppercase = single slice, lowercase=variable multi-slice block size)

Few issues:
r, =2r, already breaks the transferring of 3x3x3 algs.
I'd like a single letter, or something not involving 2, for 2R. I don't want to write 4BLD algs as URU' 2R2 UR'U'. It might help to have a single symbol for your V, indicating "drop the outer layer."
Slice and anti-slice turns not covered yet.

In the extended XML: Werner's Nx-yR is x-yR. Also will have x,y,zR. I'll also throw in some aliases for completeness, like 1r.

I'm also going to make an XML for official WCA notation, even if I don't like some of it.

Werner's applets are amazing (the curved 7 is really pretty in white), but I've wanted to be able to write applets in the notation I like, and now I finally can.

EDIT: Changed around M and m, 'cause M should probably be single-slice.

Does it have to rhyme ?

I'm not a poet,
and i know it

- Per

Sure, I just added links at the bottom of the applet pages, so you can switch between cube models ('Cube 7' is the flat version and 'V-Cube 7' is the pillowed one).
Additionally you can also see how the same algorithm/pattern looks on a cube with fewer or more layers.


V-Cube 7 (applet measures)
Side length: 72.0mm
Part length: 10.285714mm
Bevel length: 1.0mm
Sphere center for the front face: x,y,z: 0.0mm, 0.0mm,-162.0mm
Sphere radius: 203.47726mm
The radius of the sphere is chosen, so that it touches the corners of the cube minus the bevel length

Maybe someone can tell us the exact/real values, so I can improve it?


Your 'Cube Poetry' idea is awesome!
Now we can check famous words and see what pattern we get and also see if it is pretty or not.

New pattern, I think. This can be improved by using slice turns plus cube rotations in the first sequence (saves two moves), but it's not worth it to do it by hand. Also it's possible the first sequence might not be slice turn optimal in which case you might be able to save more moves.

Six Stripes
D' F2 U L2 U' R2 U L' R B' R' B F' L2 F2 U R2 F2
D N2-6R N3F' N5F' N2-6R' N3F N5F D'
R B' U R' N3F2 N3B2 R U' B R' U N3F2 N3B2
B2 U' N3F N3B' U B U' N3F' N3B U B
F2 D N3F N3B' D' F' D N3F' N3B D' F' (61 btm)

I took a look at your notation file sign.xml. Unfortunately, there are problems in this file, which cause that some constructs of the notation can not be parsed by the applets.
Since this is a complicated matter, I think it is best, if you contact me directly.



The Superset ENG notation, which my applets use by default, has a different design than the WCA notation. This is because I have a different background for my notation than the WCA people.

'Superset ENG' is based on the notation published in the books by Christoph Bandelow:
'Einführung in die Cubologie', Vieweg, Braunschweig and Wiesbaden 1981, ISBN 3-528-08499-5
'Cube and Beyond', Birkhäuser, Basel and Boston 1982, ISBN 3-7643-3078-3

I think the WCA notation is based on the works by David Singmaster et al.

Bandelows notation is able to describe cube moves, group theory constructs (such as repetition and inversion), as well as the state of a cube (the location and orientation of its parts), using only a small number of different concepts. The notation has some properties, which make it easy to write parsers for it.

With 'Superset ENG', I have leveraged Bandelows notation, so that it supports additional moves, and more group theory constructs (such as commutation, conjugation and reflection). 'Superset ENG' scales to cubes with any number of layers.
I am going to implement 'Superset ENG' in all the applets (currently some use an older notation named 'Combined ENG'), so that you can see how I have implemented the scaling.

I know, that you need to be true to the Singmaster heritage, with the 'SiGN' notation. Please plan for larger than 7-layered cubes right from the start, because (I hope) Verdes is going to produce cubes with up to 11 layers soon.
For example, make sure, that you don't need to add additional concepts, when the number of layers increases. Also make sure, that clarity doesn't degrade (by too much) for cubes with large numbers of layers.

Also, it would be great, if 'SiGN' could describe group theory constructs and cube states as well. This way, 'SiGN' would be useful for a wide range of usages, such as solution guides, speed cubing, pretty patterns, and mathematical subjects.


Before not becoming too off-topic, I am posting another pattern:

7x7 Four Smiles:
SR (NR2 NU2 NR2 NF2 NU2 NF2) SR'
(NR2 N3D2)2
(NF2 N3U2)2
ND M3F2 ND2 M3F2 ND (21 btm)

... or in 'SiGN' notation:
R L' (2R2 2U2 2R2 2F2 2U2 2F2) L R'
(2R2 3D2)2
(2F2 3U2)2
2D 3s2 2D2 3s2 2D (23 btm)

Hi

Here's my 7x7x7 multisuperflip. Based on the similar 5x5x5 version.

Hint: Based on setup moves and this 6-flip:
B2 U2 F2 U B' D' R' D F2 D2 B' F2 D L D F' (16 - cube explorer)

- Per

Just a quick 5x5 pattern combining Per's supermidgeflip with a superflip:
5x5x5 X-flip
F R' B' D' B L2 B2 R' L2 B U B R2 L' F2 L2 MR2 MF2 MD2
B L' F' D' F R2 F2 R2 L' F U F R' L2 B2 R2 MR2 MF2 MD2
TU TR2 TF TB TR TB2 TR TU2 TL TB2 TR TU' TD' TR2 TF TR' TL TB2 TU2 TF2 (58 btm)

and the long 7x7 version, very long:
7x7x7 X-flip
F R' B' D' B L2 B2 R' L2 B U B R2 L' F2 L2 N3-5R2 N3-5F2 N3-5D2
B L' F' D' F R2 F2 R2 L' F U F R' L2 B2 R2 N3-5R2 N3-5F2 N3-5D2
TF TR' TB' TD' TB TL2 TB2 TR' TL2 TB TU TB TR2 TL' TF2 TL2 MR2 MF2 MD2
TB TL' TF' TD' TF TR2 TF2 TR2 TL' TF TU TF TR' TL2 TB2 TR2 MR2 MF2 MD2
T3U T3R2 T3F T3B T3R T3B2 T3R T3U2 T3L T3B2 T3R T3U' T3D' T3R2 T3F T3R' T3L T3B2 T3U2 T3F2 (96 btm)

Hi i improved the 7x7x7 four smiles.

Breakdown:

1 - eyes
N3U NR2 NL2 N3U2 NR2 NL2 N3U

2 - the mouths
M3R2 M3F2 NU WF2 NU2 WF2 NU M3F2 M3R2

Total (16 - btm)

- Per

Six Smiley Faces: [3L 3R', 2U] [3m, 2D'] [2L 2R', 3D']
Six Clowney Faces: [3L 3R', 2U] [3m, 2D'] [2L 2R', 3D'] [E', M']
Six Frowney Faces: [3L 3R', 2U] [3m, 3D'] [2L 2R', 2D']

By the way, the applet will take [3L 3R', 2U] [3m, 2D'] [2L 2R', 3D'].

Simple trick I saw qqwref use, who informs me he got it from one of Per's pretty patterns.

Hi

Could you make all those face patterns orthogonal as well ?

- Per

Hi

Just a small suggestion on the applets. it would be useful to have a "no pattern" link on the applets. It can be a bit annoying to clear the querystring manually if you are going to play around with them. Yes i know i can use the rewind button and play with it/them. I mean when i'm testing out pattern notation i need the applet with previous querystring chopped off it

- Per

Okay, I give up. How do you make it come up in white?

Yes, the applets do support commutation of type A B A' B' as [A,B].

Conjugations of type A B A' is supported too, just write <A>B.
Here is an example of Pers improved 7x7x7 smiles using conjugators:

N3U <NR2 NL2>(N3U2) N3U · <M3R2 M3F2>(NU WF2 NU2 WF2 NU)

The syntax is configurable using the XML file. If you don't like to write a conjugation as <A>B, you can change it, for example into [A;B], or whatever you fancy.

@Per:

I have now added a "Clear" link to the applet page.

Hm. Maybe I should just make the script field editable on the applet? This would make the creation of scripts easier and would allow to give better feedback by the applet in case of a typo.

btw. I sneaked in cubic versions of V-Cube 6 and V-Cube 7 a few days ago: I am calling them Cube 6 and Cube 7.

You need to edit the XML file of the applet for this. The easiest way to do this, is by loading the XML file into CubeTwister.

Make sure that you don't use extreme white (rgb 255,255,255) and extreme black colors (rgb 0,0,0) - they won't look good on the cube. Its better to use a very bright light gray (rgb 248,248,248) and very bold dark gray (rgb 16,16,16).

Yesterday, I added a white V-Cube 7 to the template document of CubeTwister. The colors of the stickers aren't right yet, but its a good starting point. You can copy and paste it into your XML document.

Cheers,

Hi

I have a new 7x7x7 sparse multisuperflip.

And an improved normal 7x7x7 multisuperflip

I realised the corners do not have to be solved when flipping half the edges. This saves quite some turns.

The alg flipping 6 edges and messing up corners goes like so: F D R D2 B U B L B2 D

With this insight i will attempt to improve some other recent pattern - obvious which one
Here's the result: 7x7x7 X superflip
Very nice improvement. Here's the breakdown:

1 - F D R D2 B U B L B2 D (start part 1)
2 - NL2 NR2 NU2 ND2 NF2 NB2
3 - D- B2 L- B- U- B- D2 R- D- F-
4 - NL2 NR2 NU2 ND2 NF2 NB2 (end part 1)
5 - N3R2 N3L2 N3U2 N3D2 N3F2 N3B2 (start part 2)
6 - TF TD TR TD2 TB TU TB TL TB2 TD
7 - N3R2 N3L2 N3U2 N3D2 N3F2 N3B2
8 - TD- TB2 TL- TB- TU- TB- TD2 TR- TD- TF- (end part2)
9 - T3R T3L T3U2 T3F T3U' T3D T3F2 T3R2 T3B2 T3L T3U2 T3F' T3B' T3U T3R2 T3D T3F2 T3U T3R2 T3U (part 3)

Now 4 +5 condenses to VR2 VL2 VU2 VD2 VF2 VB2 (saving 6 block turns)

- Per

what about inverting long strings?
say i wanted to do (RUF) (U) (RUF)'

Seriously, do people ever try things? http://www.randelshofer.ch/test/ShowCubePlayer/Cube7.html?(RUF)%20(U)%20(RUF)'
(Also: http://cube.garron.us/applets/SiGN_test.htm?alg=%5BRUF:_U%5D.)

Thanks Per, we can now improve the 5x5x5 X-flip too:
F D R D2 B U B L B2 D NR2 NL2 NU2 ND2 NF2 NB2
D' B2 L' B' U' B' D2 R' D' F' NR2 NL2 NU2 ND2 NB2 F2
TU TR2 TF TB TR TB2 TR TU2 TL TB2 TR TU' TD' TR2 TF MR TD2 TB2 (50 btm)

I optimized the last bit by canceling out a NF2 with a TF2. Is F2 U R2 F B R B2 R U2 L B2 R U' D' R2 F M' D2 B2 a slice-turn-optimal superflip?

Any quick solution for this? Besides doing the superflip with various layers 3 times of course.

Lucas has shown me a list of stm-optimal superflips, so we can optimize the 5x5x5 X-flip more:
F D R D2 B U B L B2 D NR2 NL2 NU2 ND2 NF2 NB2
D' B2 L' B' U' B' D2 R' D' F' NR2 NL2 NU2 ND2 WF2
MR TD TL2 TU2 MB TU' MF2 MR TD TF2 TU MF2 TD TL2 TD (46 btm)

EDIT: Noah: If you're talking about what I think you are, then no, there is no faster known solution. You can move-cancel a bit though, and it's awfully pretty on the V-Cube 7 applet:
U M5F M5R2 D B2 U2 M5R U' M5F M5R2 D L2 D M5F2 U
NB2 TU' M3F2 TD' TL2 TD' M3R2 M3F' TU M3R' TU2 TB2 TD' M3R2 M3F'
N3U MF MR2 T3D T3B2 T3U2 MR T3U' MF MR2 T3D T3L2 T3D MF2 T3U T3B2 (46 btm)

EDIT II: Found a cool 5x5 pattern which I hope hasn't been done before. Can't think of a nontrivial 7x7x7 variant at the moment. 4 Big Loops:
D U' R' L' U2 F' B' U' F2 R2 L2 B2
NU TB2 TL2 TR2 TF2 TU TB TF TU2 TL TR TU' U2 ND
(NR U' R' U NR' U' R U) (NL U' L' U NL' U' L U2) D'
(NL D' L' D NL' D' L D) (NR D' R' D NR' D' R D) U' (60 btm)

EDIT III: Lucas suggests the last pattern is prettier on a 7x7x7. He's right.

Hmm, it takes too long to run that search on ACube while im at work, so i have no idea. I would guess not. The htm optimal sequence already has a slice turn in it. If better exists, quite a few recent patterns can be improved marginally. The X superflip patterns can be set up on any odd sized cube and then using the 10-turn 6-flip as a building block - for various 3x3x3 simulation layer breakdowns. The way i did my solution was to first use a 1-5-1 breakdown, then 2-3-2 and finally 3-1-3 (straight superflip).
I will later have a go at the 6-stripes pattern and try to improve it. It looks hard

- Per

Thanks Michael! And I even managed to forget the picture in gelatinbrain I captured (and I spent so long on it too!)

Hi

Quite an easy yet striking contribution this time. Just combing spots of different sizes to a suitable effect. Some setup turns to orient the faces properly.

I call it 7x7x7 ripple.

(edit)
And the 7x7x7 reverse ripple.
(/edit)

- Per

Done and done.

Two 'diagonal rectangle'-based patterns involving quick centers and lots of edge commutators.

One for 7x7x7: Twisted Thin Rectangles:
(D' F R) VU' NR' N4U' N3R' NU N3D N4R' N3U ND N3L N4U NL VD' WR (R' F')
U R U' NR' U R' U' NR
F L D2 L' NU2 L D2 L' NU2 F' D
U' R NU2 B U2 B' NU2 B U2 B' R'
U2 NR F' L2 F NR' F' L2 F U' (59 btm)

And one for 6x6x6: Peaks and Cubes:
SR2 VB VF VR' VL' VB' VF' VR VL NR' NL' NF NB NR NL NF' NB' SR2
U' R B U2 B' N3U2 B U2 B' N3U2 R'
L F U2 F' N3U2 F U2 F' N3U2 L' U
D' L B D2 B' N3D2 B D2 B' N3D2 L'
R F D2 F' N3D2 F D2 F' N3D2 R' D (62 btm)

Finally a simple but pretty 6x6 pattern: Swissmad:
VU2 M2F2 M2R2 NU2 NR2 NL2 NF2 NB2 N3U2 VF2 M2R2 NB2 NR2 NL2
TB2 T3R2 T3U2 T3F2 (18 btm)

im gonna put this out there for someone to try,

space invaders theme

ive toyed with it but havent gotten results worth posting

I can't think of any space invader-y thing you can do on a 7x7 or smaller, so I haven't bothered to write out a move sequence since there's no applet that can display it. I did attach a pattern I did in a few minutes on Iso though.
Easy 7x7 pattern (which we might have seen before), Six X:
TU TD2 TF' TL2 TD TF M3R' M3U TB' TL' TD2 TB TL2 TR'
VU VD' N3F N3B' NU' ND N3F' N3B NF NB' N3U' N3D NF' NB (28 btm)
It would be nice to have a list of all the 6x6 and 7x7 patterns we'd made already

Maybe you want to check out this pattern archive?
It contains all 6x6 and 7x7 patterns posted here.
I am constantly updating the page and will try to keep up with the posts in this thread.


@QUINBLZ
Below is a quick solution which builds
7x7 Four Pac-Man Monsters:
R2 SF2 L2 D R2 SF2 L2 U
T3R2 · NF2 MF2 NB2 · T3R2 D T3L2 · NF2 MF2 NB2 · T3L2 D'
(NF2 MR2)2
(VR2 M2U2)2
(VF2 M2U2)2

It's a simplified version of VeryWetPaint's 7x7 Blinky, Inky and Clyde pattern.
Amazing, the 7x7 cube is large enough to display simple pixel graphics of classic video games.

An old friend actually telephoned just to tell me how bad he thought the Inky, Blinky, and Clyde pattern was!

Your cube applet has nice colors and good contrast, so it makes most patterns look better.

@ VeryWetPaint

I think, the images on your site look very nice and do have a good contrast.
Unfortunately, the color schemes of some cube models are hard to distinguish, and not very easy to photograph.

I don't know, maybe it would be possible to tweak the colors using Photoshop.
But for my standards, the images on your blog look just great.

I seem to love the simple 25-pixel patterns.

Six Targets: http://www.randelshofer.ch/test/ShowCubePlayer/VCube7.html?M3UWLM3RM3D2WUM3LWDM3UWRMUMLMDMR

Σ3MW: http://www.randelshofer.ch/test/ShowCubePlayer/VCube7.html?WUWLM3RWDWRM3LM2UM3LMRN3U'M3RML2MDMR
Who wants to make ΛV<>?

Chasing Arrows: http://www.randelshofer.ch/test/ShowCubePlayer/VCube7.html?F-B-L2D2M3LWUMRWDN3UM3RMLN3U-WLM3RMUWRM3LMDD2L2BF
(Not sure if I like it better like this.)

Minimalist: Two Separated Dots. (Kastellorizo or Per can make up the love story behind this.)

Two Triangles

EDIT: More effiicient triangles, after qq tried to chastise me for not using CE.

I found a new, more efficient way of "pixeling" (drawing on the centers with EME'M' type moves), so I think I can further optimize some of the patterns from the database:

6 Swastikas:
WU N4R' N3D NR' VU' N3R' ND NU N3L VD' NL V3D NU' WR (14 btm) or
NU N4U N3L VD' NL N3U N4L N3D NR' VU' N3R' ND N4D WR (14 btm)

6 Small Swastikas:
WU N4R' VU' ND NL N3D NU N3L NU' ND' N3R' ND N3U NR' M2U' WR (16 btm)

Ripple:
U' D' R L B2 (WR' WU WR WU' V4R' N3D' N4R N3U NR NU N3R N4U N3L' V4U') B2 L' R' D U (24 btm)

Reverse Ripple:
U D R' L' F2 (WR' WU WR WU' V4R' N3D' N4R N3U NR NU N3R N4U N3L' V4U') F2 L R D' U' (24 btm)

Anaconda (Type 5):
L' (NF' D2 NF D' NL' D2 NL D) L
R (NB U2 NB' U NR U2 NR' U')
L U' D' B2 (NU N3R' V3D' N4R' N3D N3L N3U NL V3U' ND V4L') B2 D U L' R' (39 btm)

Finally a snake pattern based around the Reverse Ripple, the Reverse Ripplesnake:
R D2 B WU R' L2 WF L2 R B' D2
NR TD2 TB M3U TR' TL2 M3F TL2 TR TB' TD2
N3R T3D2 T3B T3R' T3L2 MF' T3L2 T3R MU' T3B' T3D2 T3R' (34 btm)


Oh, and Waran? I know you like keeping your cube database consistent, but I really don't understand some of the things you do. For example take a look at the pattern I called Swissmad (you call it "6 Crosses"). You managed to actually UNoptimize it (I'm pretty sure I wrote it to be 18btm, not 19), you changed the name to something that doesn't really describe it very well, and you also chose to animate the 24btm one instead. Also I don't understand your penchant for placing middots (·) all over the place, since if I want to view a pattern on the applet I have to manually remove them because they make it crash. Don't think I don't respect your efforts, but I really wish you wouldn't modify the patterns so much, especially if you are going to INCREASE the move count by doing so.

Oh yeah, and I optimized your version of the 6 Crosses too.
WU M2R' NU' ND NR' NL M2U' WR T3R2 T3U2 T3F2 T3R2 (12btm)

@ Lucas: I love your "two separated dots" although that patter could just as easily be done on a 5x5x5.

@ qqwref: The swasticas are nice (hopefully they don't offend anyone) but the Type 5 Anaconda looks incredibly awesome!

If we call them ancient Roman floor mosaics maybe people will be less likely to be offended.

Also, if you use the orientation pictured, people are slightly less likely to be offended. The Nazis had it turned the other way.

Rubik's Die.
I came up with the idea a few days ago, found an okay patterning for 7x7x7, and challenged qq to make an alg for it. I had some personal preferences for it, though, so I just made my own alg (:P):

U' R' L2 B2 MU N3R' MD N3R B2 L2 R U
U' D' (TU2 R' TU2 R NU ND' NR' NU' ND NR R' TU2 R TU2') D U
U' R D2 MU NR' MD NR D2 R' U
U2 B' R D2 N3U NR' N3U' NR D2 R' B U2
L R D' B N3D' NR' N3D NR B' D R' L'

Used my test page now.

But, move-cancelled, and on Randelshofer's applet: Die.
I haven't bothered to nest for more move-cancellations, as it wouldn't help fantastically. But I think you could easily go from 59btm to 40 or even 30 by dealing with setups and comming carefully.

That's right, I try to keep the database consistent, so patterns are rotated around the same axes and in the same directions. This should make it easier to compare algorithms and patterns.

The middots are grouping a follow-up of twists with the same axis. Other notations may have tokens for such cases, like the Slice move SR for R L' or additional Anti-slice moves like AR for R L.

Unfortunately middots may cause problems on some browsers when submitted in URL's. As far as I know, the Macintosh versions of Firefox and Safari do handle middots. Maybe you want to check an alternative browser?

However, I am considering to make the script field editable on the applet. This should solve the problem and is also more comfortable while developing algorithms.

Sorry, when I accidently UNoptimized something or when the longer instead of the shorter algorithm was animated. The database is constantly growing and I am often using scripts to translate or process big amounts of algorithms.

You are right, 'Swissmad' describes this pattern better, so I changed its name and also added the optimized 18btm algorithm.

Hmmm ....

Swissmad or swissmade ?? Are all swiss mad?
Hmmm so much activity here. I dont want to post any small optimisations. When i have entirely new contributions i will post it here.

- Per

Hi

Hi i sequenced the pattern picture by VeryWetPaint.

1. FU-R-B-D2LF2L-D2BRU-F2R2F2U2F- (outermoat part)
2. TRTU-TLTFTU-TR2TU2TRTUTR-TU2TD-TLTDTF2TL2TU (middlemost part)
3. T3BT3RT3B-T3R2T3BT3U-T3BT3UT3B2T3R2 (innermost part)

See the pattern in all it's glory!!

- Per

Ooh. So close!

You certainly got the essence of it, but it's missing one detail: avoiding similar colors in adjacent rings. Playing your sequence in the Cube Player shows ring 2 and ring 3 both contain yellow-orange-blue. Then rings 4 and 5 both contain white-red-green.

Visually this is just a nit-picky detail, but it took a lot of extra time to solve the cube this way so it worth sharing.

Unfortunately I lack the skill to sequence it myself, but I came close with this sequence (it just adds another ring-exchange, like your step 2). The front looks right, but the back is slightly shmutzed.

I'll have to see about sequencing the one I did, pictured above. The colors aren't in the same order as in your picture, but there aren't any places where two adjacent rings have the same color touching. I'm a bit tired now, so I'll do that in the morning.

Hi

Yes, i realized my sequences were not quite right, but at least it has 6 colors on each face. Yours is smudged at the back probably because you did the steps in the wrong order. There was a center rotation on one alg. That alg should have been done when it didnt matter. If i shuffle my steps i would have also have smudges

-Per

This pattern sure seemed familiar, so I went back to check if I already had a sequence for it. I did, just never put it into the applet Here it is: (Cube in)5 Cube
R WU R' B2 R WU' R U' WF2 D B2 L2
TD M2F TD' TR2 TD M2F' TD TF' M2R2 TB TR2 TU2
T3R T3U T3R T3B2 T3R2 T3U T3R T3B' T3R T3B2 T3R2 (35btm)

Hi

Thanx for correcting my small mistake. You also used slice turn optimied sequences (for the first 2 parts).
The last 11 turns can be done in 10 turns, like i did. Saving a louy turn

-Per