SuneOLL

[Kirjava]

You're only five years late to the party
http://games.groups.yahoo.com/group/speedsolvingrubikscube/message/24097

It says that 9 cases require 3 sunes when all cases are possible in 2. (With some magic)

Did you have a list of algs? I don't think the usefulness of this list would've been apparent 5 years ago ~_~.

 

[Robert-Y]

(There are some cases which we've listed with 3 sunes because we (or just I) felt that doing 3 sunes for a few cases would have been faster than 2 with... some magic...)

 

[Stefan]

Did you have a list of algs? I don't think the usefulness of this list would've been apparent 5 years ago ~_~.

I don't think so, and apparently only posted solutions for the 4-edges-flip with three and two sunes.
(Rw U2 R' U' R U' Rw') U' (Lw' U' L U' L' U2 Lw) (R' U' R U' R' U2 R)
(M U R U R' U R U2 R' M') (Rw U2 R' U' R U' Rw')

I believe we also talked somewhere about using only variants of the 6 moves T-OLL and about the combination of Sune and T-OLL variants. Are you interested in that as well? The T-OLL swaps corners diagonally, so if you do two, corner permutation isn't changed, just like when using Sunes.

There are some cases which we've listed with 3 sunes

Where? Not in the list in the first post here.

 

[Kirjava]

I believe we also talked before about using only variants of the 6 moves T-OLL and about the combination of Sune and T-OLL variants. Are you interested in that as well? The T-OLL swaps corners diagonally, so if you do two, corner permutation isn't changed, just like when using Sunes.

Indeed. A few posts ago I talked about covering every OLLCP case with just Sune, Niklas and FRURUF variations. The reason why I didn't do that at the time is that it's a lot of cases to find (300+) and I had enough to be getting on with.

If you had something to automagically generate the cases though....

There are some cases which we've listed with 3 sunes

Where? Not in the list in the first post here.

35, 13 and 14? (Counting doublesunes as two)

 

[Stefan]

A few posts ago I talked about covering every OLLCP case with just Sune, Niklas and FRURUF variations.

Right, sorry, overlooked that. I don't think we ever analyzed it, looking forward to how this turns out.

If you had something to automagically generate the cases though....

Are you doing this yourself so far instead of letting a program do it? I had a program for the Sunes but I don't know whether I still have it or how useful it would be. Maybe KSolve or GAP can help?

35, 13 and 14? (Counting doublesunes as two)

Oops, missed the "*2". Thanks.

 

[Kirjava]

Are you doing this yourself so far instead of letting a program do it? I had a program for the Sunes but I don't know whether I still have it or how useful it would be. Maybe KSolve or GAP can help?

We found all these by hand since the number of algs isn't that large and I trust myself to come up with nicer things to execute than a computer

However, something with a larger number of algs would require a different approach.

KSolve is a good shout - I need to learn how to use it anyway for a bunch of other things. I've never written a solver before, but it's a possibility if I'm feeling particularly masochistic.

An alternative is to set up a collaborative document (GoogleDox? MoonEdit?) that a group of us could edit and gradually fill in each case. We would need quite a large number of contributors to get it finished in any sort of reasonable time period though. One advantage of this technique is that the algorithms produced this way are more likely to be ergonomically friendly than their computer generated counterparts.

 

[mr. giggums]

Are you doing this yourself so far instead of letting a program do it? I had a program for the Sunes but I don't know whether I still have it or how useful it would be. Maybe KSolve or GAP can help?

We found all these by hand since the number of algs isn't that large and I trust myself to come up with nicer things to execute than a computer

However, something with a larger number of algs would require a different approach.

KSolve is a good shout - I need to learn how to use it anyway for a bunch of other things. I've never written a solver before, but it's a possibility if I'm feeling particularly masochistic.

An alternative is to set up a collaborative document (GoogleDox? MoonEdit?) that a group of us could edit and gradually fill in each case. We would need quite a large number of contributors to get it finished in any sort of reasonable time period though. One advantage of this technique is that the algorithms produced this way are more likely to be ergonomically friendly than their computer generated counterparts.

I would help if you wanted. I've too bored for the past couple of weeks with not much to do.

 

[oll+phase+sync]

EDIT: I found one here's the new version.

I didn't know OLL when I started reading this thread ( and still don't know every case). Most beginner guides to OLL start with: learn Trigger, learn Alg = Trigger combo , learn pattern ... repeat.

My hope was that using Sune OLL, I can solve OLL "dynamicly" (with out knowing all the patterns, similar to "niklas-COLL" )

So here is my suggestion how to start learning Sune OLL

Step 0: Startup Knowledge
7 corners only OLLs
3 edges only OLLs
(10 / 57)

Step 1: More Single Alg Sunes
4 fatSunes
4 M-prefixed Sunes ( e.G. M RUR’URU2R’ UM’)
( 18 / 57)

Step 2: Knowing the Bad ones
The basic idea at this point is that all other OLLs can be solved by just combining two Sunes/fat Sunes but
some cases are not that nice. To be able to use any kind of “intuitive” prediction, I should know these 6 unpredictable cases.
3 cases require M-prefixed Sunes (Cs and one Line/Bar )
3 cases reqire triple combos (some big Ls and one Kite )
(24/57)

Step 4: Categorization
I tried to categorize the remaining combo cases.

7 fatSune + normal Sune ( these ones do EO first and are really predictable, it's really bad there are so few of this kind )

12 combining two fat Sunes ( besides Dot-cases and fatDoublesunes , this is already much harder to predict)

14 normal Sune first + fat Sune second

In the End:
For me full dynamic prediction didn’t really work. For most of the cases I ended up remembering the pattern (including AUF) and the combo category. (wich is still much easier than learning 33 algs)

P.S.
Is the somewhere a thread about FRUR’U’F’/Sune - OLL

 

[CubicNL]

Here's a list of all the 16 sune variations written out:
RFS: RUR'URU2R'
RBS: R'U'RU'R'U2R
LFS: L'U'LU'L'U2L
LBS: LUL'ULU2L'
RFAS: RU2R'U'RU'R'
RBAS: R'U2RUR'UR
LFAS: L'U2L'U'LU'L'
LBAS: LU2LUL'UL
RFWS: rUR'URU2r'
RBWS: r'U'RU'R'U2r
LFWS: l'U'LU'L'U2l
LBWS: lUL'ULU2l'
RFWAS: rU2R'U'RU'r'
RBWAS: r'U2RUR'Ur
LFWAS: l'U2L'U'LU'l'
LBWAS: lU2LUL'Ul

I'm planning to learn this for OH and since I didn't really master the notation I made this list for myself.

 

[cubernya]

LBS: LUL'ULU2L'

I've used this for a long time. Although it may seem like a lefty's alg, it's actually great because L turns with left hand and U turns with right hand (alternating hands) so it's very fast

 

[CubicNL]

I've used this for a long time. Although it may seem like a lefty's alg, it's actually great because L turns with left hand and U turns with right hand (alternating hands) so it's very fast

I still use it because I like it over the regular sune (which is probably faster...)

 

[Kirjava]

Ok so I have a program now to find Niklas/Sune/Fruruf/whatever combinations for any OLLCP/LL case.

It'd be nice to have an EasyOLLCP list but making a list of the cases is a grind.

 

[Escher]

Ok so I have a program now to find Niklas/Sune/Fruruf/whatever combinations for any OLLCP/LL case.

It'd be nice to have an EasyOLLCP list but making a list of the cases is a grind.

I would help you make list :3

But not til end of August.

Full (excluding pointless cases) OLLCP here I come

 

[Kirjava]

I would help you make list :3

We should do it, it'd be smarter for me to learn OLLCP instead of ZBLL anyway >.>

I just need a list of cases in a certain format.

 

[Cielo]

Wow, looking forward to OLLCP!

 

[riffz]

We REALLY need a wiki or some collaboration tool for this. I'm too lazy though.

 

[Kirjava]

Don't worry, I'm on it

 

[mark49152]

Bumping because this is a cool thread. Did this project ever go anywhere?

On a vaguely related question, does anyone know of a list of OCLL cases that arise after applying simple EOLL for each OLL case? I'm interested in learning a few more OLL algs, and am looking for a systematic way to assess the time saving for each case - or indeed for which cases SuneOLL might be worth using as an alternative to 2-look.

 

[Kirjava]

Bumping because this is a cool thread. Did this project ever go anywhere?

Yeah. I used the same technique to learn OLLCP.

Then I tried extending it to 1LLL. Still in progress

 

[mark49152]

Yeah. I used the same technique to learn OLLCP.

Are the OLLCP algs documented anywhere?