• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

efattah

Member
Joined
Feb 14, 2016
Messages
711
For the first time in a while I did 50 LMCF solves with a 3-phase timer, timing corners/E2L/L6E.
Based on the 50 solves, with no rejected solves:
Corners: 4.88sec
E2L: 6.91sec
L6E: 3.72sec

Now I am not a top solver, and have slow TPS and not very good lookahead.
The point is that looking at my corners time (4.88) and comparing to a top 2x2 solver (1.50), I am 3.25 times slower.
Looking at my L6E (3.72) and comparing to a top roux solver (1.3), I am 2.86 times slower.

It takes me (15.52/4.88) = 3.18 times longer to solve the 3x3 cube than it takes me to solve just the corners.
If you apply that to a 2x2 solver (1.50sec 2x2 average), it predicts their 3x3 time is 3.18 x 1.50 = 4.77 seconds Ao50.
Even if you use the more skeptical 2.86 ratio (my L6E vs. top Roux LSE), you get 15.52/2.86 = 5.42 Ao50.

I am still making adjustments to the method to improve ergonomics. Another year or two and ergonomics should be close to Roux, but I don't think ergonomics can ever match Roux. I might end up migrating the method more towards WaterRoux given recent insights on the possibility to solve 1st block & corners in 1 look.
 

PapaSmurf

Member
Joined
Jan 4, 2018
Messages
1,103
WCA
2016TUDO02
YouTube
Visit Channel
No, it would belong in this one:
 

carcass

Member
Joined
Mar 8, 2020
Messages
395
Location
Eating some delicious pineapple pizza
WCA
2018HANC03
I had this idea of LLCO, as the name COLL was taken :(
I realized as a wee little lad that oriented corners would have far fewer cases than oriented edges, although the algs wouldn't be as fast.
First, you do a winter variation regardless of whether or not the edges are oriented(27 algs)
Next, you do a PLL, anti PLL, H case 1LLL, or the case with 2 adjacent misoriented edges.
PLL has 22 cases(with solved case), so does anti PLL, H has twice as many and the other has 4 times as many. This would have 176 algs. If you include winter variation, that is only 203 algs for a 1 look last layer.
 

PetraPine

Member
Joined
Feb 15, 2020
Messages
705
Location
Somewhere i guess.
YouTube
Visit Channel
I had this idea of LLCO, as the name COLL was taken :(
I realized as a wee little lad that oriented corners would have far fewer cases than oriented edges, although the algs wouldn't be as fast.
First, you do a winter variation regardless of whether or not the edges are oriented(27 algs)
Next, you do a PLL, anti PLL, H case 1LLL, or the case with 2 adjacent misoriented edges.
PLL has 22 cases(with solved case), so does anti PLL, H has twice as many and the other has 4 times as many. This would have 176 algs. If you include winter variation, that is only 203 algs for a 1 look last layer.
this like already exist maybe isnt named but ive among others have deff done CO,EO,PLL or CO,CP,EO+EP ect.
 
Joined
Aug 30, 2020
Messages
2,081
Location
On a long train journey, Smashin' PBs one a stop
YouTube
Visit Channel
EO midsolve isn't always bad (it's done in both CFOP and Roux)

doing eo midsolve is considered pretty sucky from what i've heard
Intuitive EO midsolve sucks(as in LEOR or Petrus).
Mehta EO(or EOLE) solves a piece along with orienting the edges and is algorithmic which is generally considered good(as in EOLR or ZBLS(I doubt if this is the most optimal example lol) which solve the UL and UR edges and the last F2L pair respectively)

TLDR: doing EO intuitively mid solve sucks but algorithmic EO midsolve is generally considered as good.

I had this idea of LLCO, as the name COLL was taken :(
I realized as a wee little lad that oriented corners would have far fewer cases than oriented edges, although the algs wouldn't be as fast.
First, you do a winter variation regardless of whether or not the edges are oriented(27 algs)
Next, you do a PLL, anti PLL, H case 1LLL, or the case with 2 adjacent misoriented edges.
PLL has 22 cases(with solved case), so does anti PLL, H has twice as many and the other has 4 times as many. This would have 176 algs. If you include winter variation, that is only 203 algs for a 1 look last layer.
CFCE with some influencing.
Hmm..interesting
this like already exist maybe isnt named but ive among others have deff done CO,EO,PLL or CO,CP,EO+EP ect.
Actually, carcass is proposing a LSLL method while you're talking about an LL method.
 

PetraPine

Member
Joined
Feb 15, 2020
Messages
705
Location
Somewhere i guess.
YouTube
Visit Channel
Intuitive EO midsolve sucks(as in LEOR or Petrus).
Mehta EO(or EOLE) solves a piece along with orienting the edges and is algorithmic which is generally considered good(as in EOLR or ZBLS(I doubt if this is the most optimal example lol) which solve the UL and UR edges and the last F2L pair respectively)

TLDR: doing EO intuitively mid solve sucks but algorithmic EO midsolve is generally considered as good.


CFCE with some influencing.
Hmm..interesting

Actually, carcass is proposing a LSLL method while you're talking about an LL method.
Petrus has algorithimic eo lol
 
Joined
Aug 30, 2020
Messages
2,081
Location
On a long train journey, Smashin' PBs one a stop
YouTube
Visit Channel
Petrus has algorithimic eo lol
Ok, but petrus EO has 1 more disadvantage. It involves 4 colours whereas mehta, roux or CFOP EO involves only 1 or 2 colours to recognize which is better. Therefore, MehtaZZ or whatever you proposed is not really an advantage compared to normal Mehta.

I just wanted to prove that getting rid of mid solve EO isn't much of an advantage in some situations lol.
 

PetraPine

Member
Joined
Feb 15, 2020
Messages
705
Location
Somewhere i guess.
YouTube
Visit Channel
Ok, but petrus EO has 1 more disadvantage. It involves 4 colours whereas mehta, roux or CFOP EO involves only 1 or 2 colours to recognize which is better. Therefore, MehtaZZ or whatever you proposed is not really an advantage compared to normal Mehta.

I just wanted to prove that getting rid of mid solve EO isn't much of an advantage in some situations lol.
That's not a "disatantage" solving eo earlier/for more pieces makes you get more of an advantage as the solve goes on equaling it out.
There are other reasons but what you pointed out makes no since as a point
ASWELL as eo bieng at the BEGINNING of the solve so there is no pause. And normal mehta involves three colors just as petrus does so that doesn't make since either
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,876
I had this idea of LLCO, as the name COLL was taken :(
I realized as a wee little lad that oriented corners would have far fewer cases than oriented edges, although the algs wouldn't be as fast.
First, you do a winter variation regardless of whether or not the edges are oriented(27 algs)
Next, you do a PLL, anti PLL, H case 1LLL, or the case with 2 adjacent misoriented edges.
PLL has 22 cases(with solved case), so does anti PLL, H has twice as many and the other has 4 times as many. This would have 176 algs. If you include winter variation, that is only 203 algs for a 1 look last layer.
This is COALL and some of the last layer algs are kinda gross… although that's also true for ZBLL and maybe if someone wants to put in the effort to optimise the algs it might be viable.
 

BenChristman1

Member
Joined
Oct 26, 2019
Messages
3,505
Location
The Land of 11,842 Lakes
WCA
2019CHRI11
SS Competition Results
did you guys know this off the top of your head, and if not, how do you find out if a method has already been discovered?
Once you cube for longer and longer, you start to get much more familiar with certain methods and alg sets, but if you don’t know about a method, you can look on the SS Wiki.

 

PetraPine

Member
Joined
Feb 15, 2020
Messages
705
Location
Somewhere i guess.
YouTube
Visit Channel
made this method a while ago, here is the most recent version of it.
pinetrus, a petrus method for 4x4.
don't know how to do slice moves in alg.cubing lol so this is the best I could do.
steps:
1.solve opposite centers(yau)
2.solve three cross edges(yau)
3.solve other centers (yau)
4.solve the two back pairs (one f2l one f3l) and a random edge in the FD slot.
5.pinetrus style edgepairing using back edge as keyhole.
6.3x3 stage
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
Can you guys explain M CELL to me? Perhaps an example solve? I really don't get the method.

The wiki page seems to be incorrect. I think it is missing the part where you solve the FR edge. Overall, in the main variants, the goal is to get to F2L-2. In this case, F2L-2 means to have the DF edge and DFR corner unsolved. Then finish the solve with something like L5C and L5E. Below is an example solve, using what he called SS style for F2L-2 and the advanced variant with a lot of algs. Actually, the style he describes in the post appears to be building the 1x2x3 on D and the 2x2x2 on top. I'm going to modify that to be 1x2x3 on left and 2x2x2 on the right.

Scramble: L2 D2 L D2 L F2 D' B L' U' R2 U2 F2 U F2 L2 D L2 U F2 D2

1x2x3: z' U F' B l U2 F'
2x2x2: R' U r U r R U' r' U2 R' U' R2 U r
FR edge: R U R' F R' F' R (very unlucky case here and I think a goal is to have the DFR corner also placed but misoriented)
L5C: U' R U2 R' U R U2 R' U R U2 R'
L5E: M' U M U2 M' U' M U' M' U' M

I don't think much development was put into this method. It would be a lot of algs to generate.
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
Is it really like a 1LLL system or is it more of a ZZ-CT type thing when they say one look last layer? Also, how many algs would be in L5E?

I was confused about that also. I'm not sure why the wiki says that it is a 1-look 2-alg LL system. I've been meaning to ask shadowslice about that. L5E is a little over 200 algs. Before the method development competition I was actually developing that alg set for my MI3 method. MI3 and M-CELL are similar. The difference being that MI3 is 1x2x3 -> 2x2x2 -> Choose from among several variants. It isn't F2L-2.
 

carcass

Member
Joined
Mar 8, 2020
Messages
395
Location
Eating some delicious pineapple pizza
WCA
2018HANC03
Let me be sure I have the right idea: Roux style FB, BRD 2x2x2, FR edge solved and DFR corner permuted but not always oriented. Then, BBLS or CDRLL, L5E. It doesn't seem so bad, however i don't know of any fast BBLS algs. I think if the algorithms were developed more this could be very fast.
Honestly I had never heard of BBLS before looking into this lol.
 

Silky

Member
Joined
Apr 5, 2020
Messages
873
Let me be sure I have the right idea: Roux style FB, BRD 2x2x2, FR edge solved and DFR corner permuted but not always oriented. Then, BBLS or CDRLL, L5E. It doesn't seem so bad, however i don't know of any fast BBLS algs. I think if the algorithms were developed more this could be very fast.
Honestly I had never heard of BBLS before looking into this lol.
I believe technically it can be any corner in DFR position but it's generally easier to place the the correct corner. If you haven't checked out MI3 heres a link https://sites.google.com/site/athefre/mi3. For whatever reason M-CELL wasn't really developed even though there seemed to be good potential. MI3 seems to be a similar but with a far more developed LL.
 
Top